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 [Add deprecations, removals and changes affecting compatibility here]
18 Changes to build and runtime requirements:
20 [Add changes to build and runtime requirements here]
22 Security related changes:
24 [Add security related changes here]
26 The following bugs are resolved with this release:
28 [The release manager will add the list generated by
29 scripts/list-fixed-bugs.py just before the release.]
36 * The GNU C Library can now be compiled with support for building static
37 PIE executables (See --enable-static-pie in INSTALL). These static PIE
38 executables are like static executables but can be loaded at any address
39 and provide additional security hardening benefits at the cost of some
40 memory and performance. When the library is built with --enable-static-pie
41 the resulting libc.a is usable with GCC 8 and above to create static PIE
42 executables using the GCC option '-static-pie'. This feature is currently
43 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
44 aarch64 with binutils 2.30 or later.
46 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
47 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
50 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
52 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
54 * In order to support faster and safer process termination the malloc API
55 family of functions will no longer print a failure address and stack
56 backtrace after detecting heap corruption. The goal is to minimize the
57 amount of work done after corruption is detected and to avoid potential
58 security issues in continued process execution. Reducing shutdown time
59 leads to lower overall process restart latency, so there is benefit both
60 from a security and performance perspective.
62 * The abort function terminates the process immediately, without flushing
63 stdio streams. Previous glibc versions used to flush streams, resulting
64 in deadlocks and further data corruption. This change also affects
65 process aborts as the result of assertion failures.
67 * On platforms where long double has the IEEE binary128 format (aarch64,
68 alpha, mips64, riscv, s390 and sparc), the math library now implements
69 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
70 These are the same interfaces added in version 2.26 for some platforms where
71 this format is supported but is not the format of long double.
73 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
74 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
75 implements interfaces for that type, as defined by ISO/IEC TS
76 18661-3:2015. These are corresponding interfaces to those supported for
79 * The math library now implements interfaces for the _Float32, _Float64 and
80 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
81 corresponding interfaces to those supported for _Float128.
83 * glibc now implements the memfd_create and mlock2 functions on Linux.
85 * Support for memory protection keys was added. The <sys/mman.h> header now
86 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
89 * The copy_file_range function was added.
91 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
93 * The ldconfig utility now processes `include' directives using the C/POSIX
94 collation ordering. Previous glibc versions used locale-specific
95 ordering, the change might break systems that relied on that.
97 * Support for two grammatical forms of month names has been added.
98 In a call to strftime, the "%B" and "%b" format specifiers will now
99 produce the grammatical form required when the month is used as part
100 of a complete date. New "%OB" and "%Ob" specifiers produce the form
101 required when the month is named by itself. For instance, in Greek
102 and in many Slavic and Baltic languages, "%B" will produce the month
103 in genitive case, and "%OB" will produce the month in nominative case.
105 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
106 are all valid and will all accept any known form of month
107 name---standalone or complete, abbreviated or full. In a call to
108 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
109 the strings used by "%B" and "%b", respectively. New query
110 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
111 used by "%OB" and "%Ob", respectively.
113 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
114 define the strings for %OB and %Ob, respectively; these have the
115 same syntax as "mon" and "abmon". These arrays are optional; if they
116 are not provided then they have the same content as "mon" and "abmon",
119 These features are provided for locales which define "alt_mon" and/or
120 "ab_alt_mon" in their locale source data. This release includes such
121 alternative month name data for the following languages: Belarusian,
122 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
124 This feature is currently a GNU extension, but it is expected to
125 be added to the next revision of POSIX, and it is also already
126 available on some BSD-derived operating systems.
128 This feature will cause existing statically compiled applications
129 to fail to load locales and fall back to the builtin C/POSIX locales.
130 See notes below for other changes affecting compatibility.
132 * Support for the RISC-V ISA running on Linux has been added. This port
133 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
134 for the following ISA and ABI pairs:
140 Deprecated and removed features, and other changes affecting compatibility:
142 * Statically compiled applications attempting to load locales compiled for the
143 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
144 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
145 support for two grammatical forms of the month names, also extends the locale
146 data binary format. Static applications needing locale support must be
147 recompiled to match the runtime and data they are deployed with. In some
148 distributions there is an upgrade window where dynamically linked applications
149 may use a new library but the old locale data and also fall back to the
150 builtin C/POSIX locales; restarting the application process is sufficient to
153 * Support for statically linked applications which call dlopen is deprecated
154 and will be removed in a future version of glibc. Applications which call
155 dlopen need to be linked dynamically instead.
157 * Support for old programs which use internal stdio data structures and
158 functions is deprecated. This includes programs which use the C++ streams
159 provided by libstdc++ in GCC 2.95. Programs which use the internal
160 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
161 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
162 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
163 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
164 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
165 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
166 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
167 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
168 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
169 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
170 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
171 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
172 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
173 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
174 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
175 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
176 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
177 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
178 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
179 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
180 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
181 future version of glibc. Unlike other symbol removals, these old
182 applications will not be supported using compatibility symbols.
184 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
185 defined by <sys/ptrace.h>.
187 * libm no longer supports SVID error handling (calling a user-provided
188 matherr function on error) or the _LIB_VERSION variable to control error
189 handling. (SVID error handling and the _LIB_VERSION variable still work
190 for binaries linked against older versions of the GNU C Library.) The
191 libieee.a library is no longer provided. math.h no longer defines struct
192 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
193 TLOSS, PLOSS and HUGE.
195 * The libm functions pow10, pow10f and pow10l are no longer supported for
196 new programs. Programs should use the standard names exp10, exp10f and
197 exp10l for these functions instead.
199 * The mcontext_t type is no longer the same as struct sigcontext. On
200 platforms where it was previously the same, this changes the C++ name
201 mangling for interfaces involving this type.
203 * The add-ons mechanism for building additional packages at the same time as
204 glibc has been removed. The --enable-add-ons configure option is now
207 * The --without-fp configure option is now ignored. Whether hardware
208 floating-point instructions are used is now configured based on whether
209 the compiler used at configure time (without any options implied by a
210 --with-cpu= configure option) uses such instructions.
212 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
213 the specified string can be parsed as a domain name.
215 * In the malloc_info output, the <heap> element may contain another <aspace>
216 element, "subheaps", which contains the number of sub-heaps.
218 * The libresolv function p_secstodate is no longer supported for new
221 * The tilepro-*-linux-gnu configuration is no longer supported.
223 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
224 and will be removed in a future release. Software that is still using
225 either header should be updated to use standard <stdio.h> interfaces
228 libio.h was originally the header for a set of supported GNU extensions,
229 but they have not been maintained as such in many years, they are now
230 standing in the way of improvements to stdio, and we don't think there are
231 any remaining external users. _G_config.h was never intended for public
232 use, but predates the bits convention.
234 Changes to build and runtime requirements:
236 * bison version 2.7 or later is required to generate code in the 'intl'
239 Security related changes:
241 CVE-2009-5064: The ldd script would sometimes run the program under
242 examination directly, without preventing code execution through the
243 dynamic linker. (The glibc project disputes that this is a security
244 vulnerability; only trusted binaries must be examined using the ldd
247 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
248 suffered from a one-byte overflow during ~ operator processing (either
249 on the stack or the heap, depending on the length of the user name).
250 Reported by Tim Rühsen.
252 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
253 would sometimes fail to free memory allocated during ~ operator
254 processing, leading to a memory leak and, potentially, to a denial
257 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
258 without GLOB_NOESCAPE, could write past the end of a buffer while
259 unescaping user names. Reported by Tim Rühsen.
261 CVE-2017-17426: The malloc function, when called with an object size near
262 the value SIZE_MAX, would return a pointer to a buffer which is too small,
263 instead of NULL. This was a regression introduced with the new malloc
264 thread cache in glibc 2.26. Reported by Iain Buclaw.
266 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
267 to the allocation of too much memory. (This is not a security bug per se,
268 it is mentioned here only because of the CVE assignment.) Reported by
271 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
272 of the number of search path components. (This is not a security
273 vulnerability per se because no trust boundary is crossed if the fix for
274 CVE-2017-1000366 has been applied, but it is mentioned here only because
275 of the CVE assignment.) Reported by Qualys.
277 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
278 for AT_SECURE or SUID binaries could be used to load libraries from the
281 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
282 succeeds without returning an absolute path due to unexpected behaviour
283 of the Linux kernel getcwd syscall. Reported by halfdog.
285 CVE-2018-6485: The posix_memalign and memalign functions, when called with
286 an object size near the value of SIZE_MAX, would return a pointer to a
287 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
289 The following bugs are resolved with this release:
291 [866] glob: glob should match dangling symlinks
292 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
293 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
295 [5997] math: Very slow execution of sinf function
296 [10580] localedata: hr_HR: updated locale
297 [10871] locale: 'mon' array should contain both nominative and genitive
299 [12349] localedata: eu_ES: incorrect thousands separator
300 [13605] localedata: shn_MM: new Shan locale
301 [13805] localedata: ru_RU: currency should use ',' as radix point
302 [13953] localedata: km_KH: locale update
303 [13994] localedata: mjw_IN: new locale
304 [14121] build: make writes .mo files in po directory
305 [14333] libc: Fix the race between atexit() and exit()
306 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
307 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
309 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
310 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
311 width Latin characters
312 [15332] localedata: es_CU: locale update
313 [15436] stdio: Don't close or flush stdio streams on abort
314 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
316 [16148] localedata: ca_ES: incorrect thousands separator
317 [16750] dynamic-link: ldd should not try to execute the binaries
319 [16777] localedata: pl_PL: incorrect thousands separator in locale
320 [16905] localedata: hanzi: new collation
321 [17563] localedata: cmn_TW: add hanzi collation
322 [17750] localedata: wrong collation order of diacritics in most locales
323 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
325 [17956] build: Build fails on missing definitions from header file
326 nss/nss.h when Mozilla NSS is used for cryptography
327 [18203] libc: realpath() does not handle unreachable paths correctly
328 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
329 [18812] localedata: kab_DZ: new Kabyle Algeria locale
330 [18822] libc: Internal functions are called via PLT
331 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
332 [19170] libc: __gmon_start__ defined in hppa in crtn.S
333 [19574] libc: glibc should support building static PIE binaries
334 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
336 [19971] glob: glob: Do not skip entries with zero d_ino values
337 [19982] localedata: fr.po: spelling mistake for error code EXDEV
338 [20008] localedata: km_KH: convert to translit_neutral
339 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
340 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
341 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
342 [20482] localedata: de_CH: abbreviated weekdays should be two letters
343 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
345 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
346 they're set, wrong errors returned by gaih_inet when lookup functions
348 [20756] localedata: [PATCH] Use Unicode wise thousands separator
349 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
351 [20952] localedata: yuw_PG: new locale
352 [21084] localedata: charmaps/IBM858: new codepage
353 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
354 [21242] libc: assert gives pedantic warning in old gcc versions
355 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
356 __regcall calling convention
357 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
358 [21326] libc: C99 functions are not declared for C++11 and later if
359 _GNU_SOURCE is not predefined by g++
360 [21457] libc: sys/ucontext.h namespace
361 [21530] libc: tmpfile() should be implemented using O_TMPFILE
362 [21660] math: GCC fails to compile a formula with tgmath.h
363 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
364 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
365 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
366 [21684] math: tgmath.h handling of complex integers
367 [21685] math: tgmath.h handling of bit-fields
368 [21686] math: tgmath.h handling of __int128
369 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
371 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
373 [21750] localedata: column width of characters incompatible with classical
375 [21754] malloc: malloc: Perform as little work as possible after heap
376 consistency check failures
377 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
378 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
379 [21791] string: Unused XXX_chk_XXX functions in libc.a
380 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
382 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
384 [21845] localedata: Added new Locale bho_NP
385 [21853] localedata: Fix abday Which looks same as day in zh_SG
386 [21854] localedata: Added New Locale en_SC
387 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
388 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
389 _dl_runtime_resolve_avx_slow
390 [21885] network: getaddrinfo: gethosts does not release resolver context
391 on memory allocation failure
392 [21899] libc: XPG4.2 sigaction namespace
393 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
395 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
397 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
398 errno value, causing getaddrinfo to fail
399 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
400 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
401 instead of EAI_NODATA
402 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
403 constant PTRACE_SEIZE_DEVEL
404 [21930] math: C-only gcc builtins used in <math.h> isinf
405 [21932] network: Unpaired __resolv_context_get in generic get*_r
407 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
408 [21944] libc: sigval namespace
409 [21951] localedata: Update hanzi collation by stroke
410 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
411 sysdeps/x86_64/fpu/e_expf.S
412 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
413 allocation in syscall loops)
414 [21959] localedata: Fix Country name for xh_ZA
415 [21960] localedata: Fix abmon for bem_ZM
416 [21966] math: AVX2 mathvec functions use FMA without checking
417 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
418 [21971] localedata: Added New Locale for mfe_MU
419 [21972] libc: assert macro requires operator== (int) for its argument type
420 [21973] math: [sparc] libm missing sqrtl compat symbol
421 [21974] libc: Remove __bb_init_func and __bb_exit_func
422 [21982] string: stratcliff.c: error: assuming signed overflow does not
424 [21986] stdio: __guess_grouping is called incorrectly
425 [21987] math: [sparc32] wrong bits/long-double.h installed
426 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
428 [22022] localedata: Missing country_name for mni_IN
429 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
431 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
433 [22026] locale: iconv_open: heap overflow on gconv_init failure
434 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
435 [22035] math: [m68k] bits/math-inline.h macro namespace
436 [22038] localedata: Fix abbreviated weeks and months for Somali
437 [22044] localedata: Remove redundant data for Limburgish Language
438 [22050] malloc: Linking with -lmcheck does not hook
439 __malloc_initialize_hook correctly
440 [22051] libc: zero terminator in the middle of glibc's .eh_frame
441 [22052] malloc: malloc failed to compile with GCC 7 and -O3
442 [22070] localedata: charmaps/UTF-8: wcwidth for
443 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
444 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
445 Jungseong and Jongseong) should be 0
446 [22078] nss: nss_files performance issue in multi mode
447 [22082] math: bits/math-finite.h exp10 condition
448 [22086] libc: pcprofiledump incorrect cross-endian condition
449 [22093] dynamic-link: ld.so no longer searches in .../x86_64
450 [22095] network: Name server address allocation memory leak in resolv.conf
452 [22096] network: __resolv_conf_attach can incorrectly free passed conf
454 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
455 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
457 [22111] malloc: malloc: per thread cache is not returned when thread exits
458 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
459 [22134] libc: [linux] implement fexecve with execveat
460 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
462 [22145] libc: ttyname() gives up too early in the face of namespaces
463 [22146] math: C++ build issue with float128 on x86_64
464 [22153] nptl: nptl: save error code before process termination
465 [22156] libc: [hppa,ia64,microblaze] Executable stack default
466 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
467 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
468 [22165] libc: [hppa] Text relocations in libc.so
469 [22180] libc: destructor registered via __cxa_atexit is called twice
470 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
471 symlinks") cause "make" segfaults
472 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
474 [22207] libc: FAIL: stdlib/test-atexit-race
475 [22225] math: nearbyint arithmetic moved before feholdexcept
476 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
477 [22235] math: iscanonical in C++ and float128
478 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
479 the svid compat wrapper
480 [22244] math: ynf and yn are wrong without the svid compat wrapper
481 [22273] libc: Improper assert in Linux posix_spawn implementation
482 [22284] libc: -pg -pie doesn't work
483 [22292] locale: localedef exits with error 4 when it should be error 1
484 [22294] locale: Allow "" for int_currency_symbol definition in locales.
485 [22295] locale: Don't warn on non-symbolic characters in locale sources in
487 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
488 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
490 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
491 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
492 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
493 [22322] libc: [mips64] wrong bits/long-double.h installed
494 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
495 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
497 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
498 [22343] malloc: Integer overflow in posix_memalign
499 [22347] libc: getrandom() returns the number of bytes that were copied to
500 the buffer even though the comments say "Return 0 on success and -1 on
502 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
503 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
505 [22370] dynamic-link: Incorrect note padding check
506 [22375] libc: malloc returns pointer from tcache_get when should return
507 NULL (CVE-2017-17426)
508 [22377] math: iseqsig, float128 and C++
509 [22382] localedata: Error in tpi_PG locale
510 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
511 inside the ASCII printable range
512 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
514 [22403] localedata: Slash needs escaping in some locales
515 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
516 [22409] network: res_hnok does not accept some host names used on the
518 [22412] network: res_dnok, res_hnok should perform syntax checks
519 [22413] network: ns_name_pton ignores syntactically invalid trailing
521 [22415] stdio: setvbuf can lead to invalid free/segfault
522 [22432] build: Non-deterministic build
523 [22439] malloc: malloc_info should compute summary statistics for all sub-
525 [22442] network: if_nametoindex could report index for the wrong
527 [22446] build: aliasing violation calling readlink in handle_request
528 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
529 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
530 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
531 __nscd_hash/__nss_hash
532 [22463] network: p_secstodate overflow handling
533 [22469] localedata: pl_PL LC_COLLATE does not use i18n
534 [22478] libc: sigwait can fail with EINTR
535 [22505] libc: ldconfig processes include directive in locale-specific
537 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
538 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
539 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
540 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
541 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
542 [22534] localedata: Collation rules for Serbian and Bosnian should be the
544 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
545 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
546 [22577] libc: missing newline after "cannot allocate TLS data structures
548 [22588] manual: manual/conf.texi: missing underscore in front of
550 [22593] math: nextafter and nexttoward are declared with const attribute
551 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
552 [22603] string: ia64 memchr overflows internal pointer check
553 [22605] libc: SH clone does not set the exit code correctly
554 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
556 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
557 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
558 modified in case of success
559 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
560 [22615] manual: manual: ambiguous wording about errno value in case of
562 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
564 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
565 binaries or if /proc is not mounted (CVE-2017-16997)
566 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
567 [22630] build: $(no-pie-ldflag) is no longer effective
568 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
569 [22635] nptl: pthread_self returns NULL before libpthread is loaded
570 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
571 [22637] nptl: guard size is subtracted from thread stack size instead of
573 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
574 [22657] localedata: hu_HU: Avoid double space in date
575 [22660] math: fmax, fmin sNaN handling on alpha
576 [22664] libc: New warning of GCC8
577 [22665] math: alpha: ceil and floor raise inexact exceptions
578 [22666] math: alpha: trunc raise inexact exceptions
579 [22667] libc: makecontext lacks stack alignment on i386
580 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
581 [22679] libc: getcwd(3) can succeed without returning an absolute path
583 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
585 [22687] math: [powerpc-nofpu] complex long double functions spurious
587 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
588 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
589 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
591 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
592 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
594 [22701] nis: Incomplete removal of libnsl
595 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
596 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
597 [22715] dynamic-link: FAIL: elf/tst-audit10
598 [22719] libc: Backtrace tests fail on hppa
599 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
600 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
601 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
602 before the first call to crypt_r () in crypt/badsalttest.c
609 * A per-thread cache has been added to malloc. Access to the cache requires
610 no locks and therefore significantly accelerates the fast path to allocate
611 and free small amounts of memory. Refilling an empty cache requires locking
612 the underlying arena. Performance measurements show significant gains in a
613 wide variety of user workloads. Workloads were captured using a special
614 instrumented malloc and analyzed with a malloc simulator. Contributed by
615 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
617 * Unicode 10.0.0 Support: Character encoding, character type info, and
618 transliteration tables are all updated to Unicode 10.0.0, using
619 generator scripts contributed by Mike FABIAN (Red Hat).
620 These updates cause user visible changes, especially the changes in
621 wcwidth for many emoji characters cause problems when emoji sequences
622 are rendered with pango, see for example:
623 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
625 * Collation of Hungarian has been overhauled and is now consistent with "The
626 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
629 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
631 - The GNU C Library will now detect when /etc/resolv.conf has been
632 modified and reload the changed configuration. The new resolver option
633 “no-reload” (RES_NORELOAD) disables this behavior.
635 - The GNU C Library now supports an arbitrary number of search domains
636 (configured using the “search” directive in /etc/resolv.conf);
637 previously, there was a hard limit of six domains. For backward
638 compatibility, applications that directly modify the ‘_res’ global
639 object are still limited to six search domains.
641 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
642 Library will now randomly pick a name server from the configuration as a
643 starting point. (Previously, the second name server was always used.)
645 * The tunables feature is now enabled by default. This allows users to tweak
646 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
648 * New function reallocarray, which resizes an allocated block (like realloc)
649 to the product of two sizes, with a guaranteed clean failure upon integer
650 overflow in the multiplication. Originally from OpenBSD, contributed by
651 Dennis Wölfing and Rüdiger Sonderfeld.
653 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
654 These are extended versions of preadv and pwritev, respectively, taking an
655 additional flags argument. The set of supported flags depends on the
656 running kernel; full support currently requires kernel 4.7 or later.
658 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
659 create a new session ID for the spawned process. This feature is
660 scheduled to be added to the next major revision of POSIX; for the time
661 being, it is available under _GNU_SOURCE.
663 * errno.h is now safe to use from C-preprocessed assembly language on all
664 supported operating systems. In this context, it will only define the
665 Exxxx constants, as preprocessor macros expanding to integer literals.
667 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
668 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
669 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
670 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
672 To compile programs that use this feature, the compiler must support
673 128-bit floating point with the type name _Float128 (as defined by TS
674 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
675 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
676 must be defined to make the new interfaces visible.
678 The new functions and macros correspond to those present for other
679 floating-point types (except for a few obsolescent interfaces not
680 supported for the new type), with F128 or f128 suffixes; for example,
681 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
682 printf or scanf formats for the new type; the strfromf128 and strtof128
683 interfaces should be used instead.
685 Deprecated and removed features, and other changes affecting compatibility:
687 * The synchronization that pthread_spin_unlock performs has been changed to
688 now be equivalent to a C11 atomic store with release memory order to the
689 spin lock's memory location. Previously, several (but not all)
690 architectures used stronger synchronization (e.g., containing what is
691 often called a full barrier). This change can improve performance, but
692 may affect odd fringe uses of spin locks that depend on the previous
693 behavior (e.g., using spin locks as atomic variables to try to implement
694 Dekker's mutual exclusion algorithm).
696 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
699 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
700 will only be built and installed when the GNU C Library is configured with
701 --enable-obsolete-rpc. This allows alternative RPC implementations, such
702 as TIRPC or rpcsvc-proto, to be used.
704 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
705 libnss_compat, are deprecated, and will not be built or installed by
708 The NIS(+) support library, libnsl, is also deprecated. By default, a
709 compatibility shared library will be built and installed, but not headers
710 or development libraries. Only a few NIS-related programs require this
711 library. (In particular, the GNU C Library has never required programs
712 that use 'gethostbyname' to be linked with libnsl.)
714 Replacement implementations based on TIRPC, which additionally support
715 IPv6, are available from <https://github.com/thkukuk/>. The configure
716 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
717 name service modules, to be built and installed.
719 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
720 support is enabled, the configured recursive resolver must support EDNS.
721 (Responding to EDNS-enabled queries with responses which are not
722 EDNS-enabled is fine, but FORMERR responses are not.)
724 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
725 servers have not supported this opcode for a long time.
727 * The _res_opcodes variable has been removed from libresolv. It had been
728 exported by accident.
730 * <string.h> no longer includes inline versions of any string functions,
731 as this kind of optimization is better done by the compiler. The macros
732 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
734 * The nonstandard header <xlocale.h> has been removed. Most programs should
735 use <locale.h> instead. If you have a specific need for the definition of
736 locale_t with no other declarations, please contact
737 libc-alpha@sourceware.org and explain.
739 * The obsolete header <sys/ultrasound.h> has been removed.
741 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
743 * The obsolete function cfree has been removed. Applications should use
746 * The stack_t type no longer has the name struct sigaltstack. This changes
747 the C++ name mangling for interfaces involving this type.
749 * The ucontext_t type no longer has the name struct ucontext. This changes
750 the C++ name mangling for interfaces involving this type.
752 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
753 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
754 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
755 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
756 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
757 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
758 fpregset_t type no longer has the name struct fpu. This changes the C++
759 name mangling for interfaces involving those types.
761 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
762 synced with the kernel:
764 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
765 are not supported on this architecture and have been removed.
767 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
768 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
769 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
771 Programs that assume the GET/SETREGS ptrace requests are universally
772 available will now fail to build, instead of malfunctioning at runtime.
774 Changes to build and runtime requirements:
776 * Linux kernel 3.2 or later is required at runtime, on all architectures
777 supported by that kernel. (This is a change from version 2.25 only for
780 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
782 * On most architectures, GCC 4.9 or later is required to build the GNU C
783 Library. On powerpc64le, GCC 6.2 or later is required.
785 Older GCC versions and non-GNU compilers are still supported when
786 compiling programs that use the GNU C Library. (We do not know exactly
787 how old, and some GNU extensions to C may be _de facto_ required. If you
788 are interested in helping us make this statement less vague, please
789 contact libc-alpha@sourceware.org.)
791 Security related changes:
793 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
794 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
796 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
797 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
799 * Avoid printing a backtrace from the __stack_chk_fail function since it is
800 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
803 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
804 fixed (CVE-2017-12133).
806 The following bugs are resolved with this release:
808 [984] network: Respond to changed resolv.conf in gethostbyname
809 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
810 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
812 [12189] libc: __stack_chk_fail should not attempt a backtrace
814 [14096] time: Race condition on timezone/tst-timezone.out
815 [14172] localedata: az_IR: new locale
816 [14995] build: glibc fails to build if gold is the default linker, even if
818 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
819 [16637] network: inet_pton function is accepting IPv6 with bad format
820 [16640] string: string/strtok.c: undefined behaviour inconsistent between
821 x86 and other generic code
822 [16875] localedata: ko_KR: fix lang_name
823 [17225] localedata: ar_SY: localized month names for May and June are
825 [17297] localedata: da_DK: wrong date_fmt string
826 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
827 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
828 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
829 [19066] localedata: ar_SA abbreviated day and month names are in English
830 [19569] network: resolv: Support an arbitrary number of search domains
831 [19570] network: Implement random DNS server selection in the stub
833 [19838] locale: localedef fails on PA-RISC
834 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
835 order of 0D36 and 0D37
836 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
838 [20098] libc: FAIL: debug/backtrace-tst on hppa
839 [20257] network: sunrpc: clntudp_call does not enforce timeout when
841 [20275] localedata: locale day/abday/mon/abmon should not have trailing
843 [20313] localedata: Update locale data to Unicode 9.0
844 [20424] manual: Document how to provide a malloc replacement
845 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
847 [20686] locale: Add el_GR@euro to SUPPORTED.
848 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
849 failures consistently
850 [21015] dynamic-link: Document and fix --enable-bind-now
851 [21016] nptl: pthread_cond support is broken on hppa
852 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
853 frame-pointer on i386
854 [21049] libc: segfault in longjmp_chk() due to clobbered processor
856 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
857 [21088] libc: Build fails with --enable-static-nss
858 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
859 generic c code is used
860 [21109] libc: Tunables broken on big-endian
861 [21112] math: powf has large ulp errors with base close to 1 and exponent
863 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
865 [21120] malloc: glibc malloc is incompatible with GCC 7
866 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
868 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
869 (-0.0) when linking with -lieee
870 [21171] math: log10, log2 and lgamma return incorrect results
871 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
872 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
873 [21207] localedata: ce_RU: update weekdays from CLDR
874 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
875 [21217] localedata: Update months from CLDR-31
876 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
877 [21243] libc: support_delete_temp_file should issue warning for failed
879 [21244] libc: support resolv_test_start() socket fd close should be
881 [21253] libc: localedef randomly segfaults when using -fstack-check due to
882 new posix_spawn implementation
883 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
884 leads to lower CPU frequency
885 [21259] libc: [alpha] termios.h missing IXANY for POSIX
886 [21261] libc: [sparc64] bits/setjmp.h namespace
887 [21267] network: [mips] bits/socket.h IOC* namespace
888 [21268] libc: [alpha] termios.h NL2, NL3 namespace
889 [21270] libc: mmap64 silently truncates large offset values
890 [21275] libc: posix_spawn always crashes on ia64 now
891 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
892 [21280] math: [powerpc] logbl for POWER7 return incorrect results
893 [21289] libc: Incorrect declaration for 32-bit platforms with
894 _FILE_OFFSET_BITS=64 causes build error
895 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
896 not support gethostbyname4_r
897 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
899 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
901 [21340] libc: Support POSIX_SPAWN_SETSID
902 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
903 [21359] network: ns_name_pack needs additional byte in destination buffer
904 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
905 against fragmentation attacks (CVE-2017-12132)
906 [21369] network: resolv: Remove EDNS fallback
907 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
909 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
910 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
911 [21393] stdio: Missing dup3 error check in freopen, freopen64
912 [21396] libc: Use AVX2 memcpy/memset on Skylake server
913 [21399] localedata: Bad description for U00EC in
914 localedata/charmaps/CP1254
915 [21411] malloc: realloc documentation error
916 [21426] network: sys/socket.h uio.h namespace
917 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
918 [21445] libc: signal.h bsd_signal namespace
919 [21455] network: Network headers stdint.h namespace
920 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
922 [21475] network: resolv: Overlong search path is truncated mid-label
923 [21511] libc: sigstack namespace
924 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
925 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
926 bits/syscall.h: No such file or directory
927 [21517] libc: struct sigaltstack namespace
928 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
929 [21533] localedata: Update locale data to Unicode 10.0
931 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
932 junk at end of line, first unrecognized character is `@'
933 [21538] libc: SIG_HOLD missing for XPG4
934 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
935 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
936 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
937 [21543] libc: sigevent namespace
938 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
939 size for general purpose registers in mcontext_t structure
940 [21550] libc: sigwait namespace
941 [21552] libc: XPG4 bsd_signal namespace
942 [21554] libc: sigpause namespace
943 [21560] libc: sys/wait.h signal.h namespace
944 [21561] libc: waitid namespace
945 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
946 '/usr/include/stdlib.h': Permission denied
947 [21575] libc: sys/wait.h missing struct rusage definition
948 [21584] libc: sigaltstack etc namespace
949 [21597] libc: siginterrupt namespace
950 [21607] math: hppa: FAIL: math/test-tgmath
951 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
953 [21622] libc: [tile] missing SA_* for POSIX.1:2008
954 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
955 stack and heap (CVE-2017-1000366)
956 [21625] libc: wait3 namespace
957 [21654] nss: Incorrect pointer alignment in NSS group merge result
959 [21657] network: Parse interface zone id for node-local multicast
960 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
962 [21666] libc: .symver is used on common symbol
963 [21668] network: resolv: res_init cross-thread broadcast introduces race
965 [21687] math: tgmath.h totalorder, totalordermag return type
966 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
968 [21696] libc: Incorrect assumption of of __cpu_mask in
969 posix/sched_cpucount.c
970 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
971 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
973 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
974 leading to relocation crash
975 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
976 `resolv_conf_matches (resp, conf)' failed.
977 [21710] localedata: Added Samoan language locale for Samoa
978 [21711] localedata: Pashto yesstr/nostr locale are missing
979 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
980 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
981 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
982 [21723] localedata: yesstr/nostr missing for Chinese language locale
983 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
985 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
987 [21728] localedata: New Locale for Tongan language
988 [21729] localedata: incorrect LC_NAME fields for hi_IN
989 [21733] localedata: yesstr and nostr are missing for zh_HK
990 [21734] localedata: Missing yesstr and nostr are for kw_GB
991 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
992 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
993 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
994 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
996 [21744] libc: Tests failing on --enable-tunables --enable-stack-
998 [21749] localedata: Wrong abbreviated day name (“abday”) for
1000 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1001 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1002 [21759] localedata: missing yesstr and nostr for Tigrinya
1003 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1004 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1005 [21767] localedata: Missing Bislama locales
1006 [21768] localedata: Missing yesstr and nostr for aa_ET
1007 [21770] localedata: Missing Field in li_NL
1008 [21778] nptl: Robust mutex may deadlock
1009 [21779] libc: MicroBlaze segfaults when loading libpthread
1010 [21783] localedata: Fix int_select international_call_prefixes
1011 [21784] localedata: Inconsistency in country_isbn
1012 [21788] localedata: Missing Country Postal Abbreviations
1013 [21794] localedata: Added-country_isbn-for-Italy
1014 [21795] localedata: Add/Fix country_isbn for France
1015 [21796] localedata: Added country_isbn for Republic of Korea
1016 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1017 [21799] localedata: Added int_select international_call_prefixes
1018 [21801] localedata: Added int_select international_call_prefixes
1019 [21804] nptl: Double semicolon in thread-shared-types.h
1020 [21807] localedata: LC_ADDRESS fix for pap_CW
1021 [21808] localedata: Fix LC_ADDRESS for pap_AW
1022 [21821] localedata: Added country_name in mai_IN
1023 [21822] localedata: Fix LC_TIME for mai_IN
1024 [21823] localedata: missing yesstr, nostr for sa_IN
1025 [21825] localedata: Fix name_mrs for mag_IN
1026 [21828] localedata: 2.26 changelog should mention user visible changes
1028 [21835] localedata: Added Maithili language locale for Nepal
1029 [21838] localedata: Removed redundant data for the_NP
1030 [21839] localedata: Fix LC_MONETARY for ta_LK
1031 [21844] localedata: Fix Latin characters and Months Sequence.
1032 [21848] localedata: Fix mai_NP Title Name
1037 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1038 24731-2:2010, is supported to enable declarations of functions from that
1039 TR. Note that not all functions from that TR are supported by the GNU C
1042 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1043 18661-1:2014, is supported to enable declarations of functions and macros
1044 from that TS. Note that not all features from that TS are supported by
1047 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1048 18661-4:2015, is supported to enable declarations of functions and macros
1049 from that TS. Note that most features from that TS are not supported by
1052 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1053 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1054 Since the GNU C Library defaults to a much newer revision of POSIX, this
1055 will only affect programs that specifically request an old conformance
1056 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1057 see a change in the visible declarations, but a program compiled with
1058 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1061 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1062 defined by all multithreaded code, but glibc has not required this for
1065 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1066 means that in a future release, the macros “major”, “minor”, and “makedev”
1067 will only be available from <sys/sysmacros.h>.
1069 These macros are not part of POSIX nor XSI, and their names frequently
1070 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1071 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1072 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1075 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1076 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1077 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1079 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1080 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1081 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1082 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1083 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1084 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1085 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1086 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1087 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1088 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1089 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1092 * New <math.h> features are added from TS 18661-1:2014:
1094 - Signaling NaN macros: SNANF, SNAN, SNANL.
1096 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1097 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1098 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1100 - llogb functions: the llogb, llogbf and llogbl functions, and the
1101 FP_LLOGB0 and FP_LLOGBNAN macros.
1103 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1106 - Comparison macros: iseqsig.
1108 - Classification macros: iscanonical, issubnormal, iszero.
1110 - Total order functions: totalorder, totalorderf, totalorderl,
1111 totalordermag, totalordermagf, totalordermagl.
1113 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1115 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1116 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1118 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1119 are added to libc. They convert a floating-point number into string.
1121 * Most of glibc can now be built with the stack smashing protector enabled.
1122 It is recommended to build glibc with --enable-stack-protector=strong.
1123 Implemented by Nick Alcock (Oracle).
1125 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1126 intended to be used instead of memset() to erase sensitive data after use;
1127 the compiler will not optimize out calls to explicit_bzero even if they
1128 are "unnecessary" (in the sense that no _correct_ program can observe the
1129 effects of the memory clear).
1131 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1132 to float instead of double. This does not affect the ABI of any libraries
1133 that are part of the GNU C Library, but may affect the ABI of other
1134 libraries that use this type in their interfaces.
1136 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1137 float_t and double_t types are now defined to long double instead of float
1138 and double. These options are not the default, and this does not affect
1139 the ABI of any libraries that are part of the GNU C Library, but it may
1140 affect the ABI of other libraries that use this type in their interfaces,
1141 if they are compiled or used with those options.
1143 * The getentropy and getrandom functions, and the <sys/random.h> header file
1146 * The buffer size for byte-oriented stdio streams is now limited to 8192
1147 bytes by default. Previously, on Linux, the default buffer size on most
1148 file systems was 4096 bytes (and thus remains unchanged), except on
1149 network file systems, where the buffer size was unpredictable and could be
1150 as large as several megabytes.
1152 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1153 for the Linux quota interface which predates kernel version 2.4.22 has
1156 * The malloc_get_state and malloc_set_state functions have been removed.
1157 Already-existing binaries that dynamically link to these functions will
1158 get a hidden implementation in which malloc_get_state is a stub. As far
1159 as we know, these functions are used only by GNU Emacs and this change
1160 will not adversely affect already-built Emacs executables. Any undumped
1161 Emacs executables, which normally exist only during an Emacs build, should
1162 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1164 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1165 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1166 “no-ip6-dotint” had already been the default, and support for the
1167 “ip6-dotint” option was removed from the Internet in 2006.
1169 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1170 flag from <resolv.h> have been removed. The option relied on a
1171 backwards-incompatible DNS extension which was never deployed on the
1174 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1175 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1176 They were already unimplemented.
1178 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1179 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1180 removed again from the IETF name lookup interface specification in RFC
1181 2553. Applications should use getaddrinfo instead.
1183 * DNSSEC-related declarations and definitions have been removed from the
1184 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1185 decode the data part of DNSSEC record types. Previous versions of glibc
1186 only implemented minimal support for the previous version of DNSSEC, which
1187 is incompatible with the currently deployed version.
1189 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1190 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1191 <arpa/nameser.h> header file because the distinction between RR types and
1192 meta-RR types is not officially standardized, subject to revision, and
1193 thus not suitable for encoding in a macro.
1195 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1196 and rhook members of the res_state type in <resolv.h> have been removed.
1197 The glibc stub resolver did not support these hooks, but the header file
1198 did not reflect that.
1200 * For multi-arch support it is recommended to use a GCC which has
1201 been built with support for GNU indirect functions. This ensures
1202 that correct debugging information is generated for functions
1203 selected by IFUNC resolvers. This support can either be enabled by
1204 configuring GCC with '--enable-gnu-indirect-function', or by
1205 enabling it by default by setting 'default_gnu_indirect_function'
1206 variable for a particular architecture in the GCC source file
1209 * GDB pretty printers have been added for mutex and condition variable
1210 structures in POSIX Threads. When installed and loaded in gdb these pretty
1211 printers show various pthread variables in human-readable form when read
1212 using the 'print' or 'display' commands in gdb.
1214 * Tunables feature added to allow tweaking of the runtime for an application
1215 program. This feature can be enabled with the '--enable-tunables' configure
1216 flag. The GNU C Library manual has details on usage and README.tunables has
1217 instructions on adding new tunables to the library.
1219 * A new version of condition variables functions have been implemented in
1220 the NPTL implementation of POSIX Threads to provide stronger ordering
1223 * A new version of pthread_rwlock functions have been implemented to use a more
1224 scalable algorithm primarily through not using a critical section anymore to
1227 Security related changes:
1229 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1230 have been created with makecontext could fail to terminate due to a
1231 missing .cantunwind annotation. This has been observed to lead to a hang
1232 (denial of service) in some Go applications compiled with gccgo. Reported
1233 by Andreas Schwab. (CVE-2016-6323)
1235 * The DNS stub resolver functions would crash due to a NULL pointer
1236 dereference when processing a query with a valid DNS question type which
1237 was used internally in the implementation. The stub resolver now uses a
1238 question type which is outside the range of valid question type values.
1241 The following bugs are resolved with this release:
1243 [4099] stdio: Overly agressive caching by stream i/o functions.
1244 [7065] build: Support building glibc with -fstack-protector or -fstack-
1246 [9842] localedata: en_CA: incorrect date format
1247 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1248 before it started waiting
1249 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1250 [15765] nptl: sem_open is wrongly a cancellation point
1251 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1252 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1254 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1255 library linked with pthread
1256 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1257 "FIXME: Ingo" issue)
1258 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1259 [17252] libc: getrandom and getentropy syscall
1260 [17730] malloc: thread-local storage is sometimes improperly free()'d
1261 after being __libc_memalign()'d
1262 [18241] stdio: failed fseek on memstream does not set errno and can fail
1264 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1266 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1267 PTHREAD_PRIO_INHERIT mutex on ARM
1268 [18784] network: res_query and related function crash for special record
1269 type queries (CVE-2015-5180)
1270 [19380] math: strtod does not raise "inexact"
1271 [19387] string: Integer overflow in memchr
1272 [19390] string: Integer overflow in strncat
1273 [19398] build: linknamespace tests fail with massively parallel build
1274 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1276 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1277 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1278 [19514] libc: [PATCH] Fix spelling errors in spelling
1279 "implement"/"implementation" in several places
1280 [19582] network: Deprecate RES_USE_INET6
1281 [19673] manual: clog10 docs appear to be erroneous
1282 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1283 causes a segmentation fault
1284 [19826] libc: invalid pointer returned from __tls_get_addr with static
1286 [20016] network: resolv: Remove hooks support from the API
1287 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1289 [20033] math: [x86_64] vectorized math function don't call the __finite
1291 [20116] nptl: use after free in pthread_create
1292 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1293 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1295 [20311] nptl: please install proc_service.h
1296 [20366] libc: Compilation errors in installed headers in strict-compliance
1298 [20370] malloc: malloc: Arena free list management is still racy
1299 (incorrect fix in bug 19243)
1300 [20386] libc: assert (X = 0) does not result in GCC warning
1301 [20432] malloc: malloc: Minimize interface required for interposition
1302 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1303 in _Unwind_Backtrace (CVE-2016-6323)
1304 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1305 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1307 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1308 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1309 [20477] network: resolv: incorrect double-checked locking related to
1311 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1313 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1315 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1316 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1317 cause transition penalty
1318 [20517] math: sparcv9 missing fdiml compat symbol
1319 [20524] manual: strverscmp is inconsistent
1320 [20525] libc: <sys/quota.h> should be based on kernel headers
1321 [20539] math: GCC 7 -static -lm fails to link at -O3
1322 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1324 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1326 [20591] network: Remove obsolete DNSSEC support
1327 [20592] network: DNS resource record type classification macros in
1328 <arpa/nameser.h> are incorrect
1329 [20593] network: Update DNS RR type definitions
1330 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1331 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1333 [20629] network: libresolv: Remove support for bitstring labels
1335 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1336 assertion: "maxidx >= 2"
1337 [20660] math: [arm] Use VSQRT
1338 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1339 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1340 [20689] libc: Test for FMA should also check for AVX.
1341 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1342 [20715] math: iszero macro breaks existing code
1343 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1344 [20728] libc: powerpc: Missing TOC stub in clone
1345 [20729] build: glibc-2.24 fails to build for i486 with -Os
1346 [20750] build: Build fails with default PIE enabled gcc-6.x
1347 [20768] math: [alpha] sqrt fegetenv namespace
1348 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1350 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1351 __FLT_EVAL_METHOD__ is 2
1352 [20790] network: rpcgen buffer overrun in get_prog_declaration
1353 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1354 [20829] libc: crypt snprintf namespace
1355 [20847] libc: tst-vfork3 failure
1356 [20855] math: Default bits/mathdef.h has inappropriate float_t
1357 [20859] math: [sh4] FP_ILOGB0 invalid
1358 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1359 U+20AC), not same as GBK
1360 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1361 [20916] math: pow handling of sNaN arguments
1362 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1363 [20919] math: Bad pow (qNaN, 0) result with -lieee
1364 [20940] math: hypot sNaN handling
1365 [20947] math: fmax, fmin sNaN handling
1366 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1367 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1369 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1370 [20973] nptl: robust mutexes: Lost wake-ups
1371 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1372 [20978] nis: strlen on null pointer in nss_nisplus
1373 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1375 [21014] string: i686 memchr overflows internal pointer check
1376 [21019] libc: [mips] n32 lseek incorrect on overflow
1377 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1378 [21026] libc: [MIPS] readahead syscall is broken on n64
1379 [21028] math: Fallback fesetexceptflag should always succeed
1380 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1381 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1383 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1384 [21061] librt: [microblaze] librt lost clock_* exports
1385 [21073] libc: tunables: insecure environment variables passed to
1386 subprocesses with AT_SECURE
1387 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1391 * The minimum Linux kernel version that this version of the GNU C Library
1392 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1393 kernel version 2.6.32 or later suffices (on architectures that already
1394 required kernel versions more recent than 3.2, those requirements remain
1395 unchanged). Linux 3.2 or later kernel headers are required on all
1398 * The pap_AN locale has been deleted. This has been deprecated for a long
1399 time. It has been replaced by pap_AW & pap_CW, both of which have long
1400 been included in previous releases.
1402 * The readdir_r and readdir64_r functions have been deprecated. It is
1403 recommended to use readdir and readdir64 instead.
1405 * The type “union wait” has been removed. It was deprecated in the early
1406 1990s and never part of POSIX. Application code should use the int type
1407 instead of “union wait”.
1409 * A new NSS action is added to facilitate large distributed system
1410 administration. The action, MERGE, allows remote user stores like LDAP
1411 to be merged into local user stores like /etc/groups in order to provide
1412 easy to use, updated, and managed sets of merged credentials. The new
1413 action can be used by configuring it in /etc/nsswitch.conf:
1414 group: files [SUCCESS=merge] nis
1415 Implemented by Stephen Gallagher (Red Hat).
1417 * The deprecated __malloc_initialize_hook variable has been removed from the
1420 * The long unused localedef --old-style option has been removed. It hasn't
1421 done anything in over 16 years. Scripts using this option can safely
1424 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1425 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1426 return the next representable value in the direction of positive infinity
1427 and the nextdown functions return the next representable value in the
1428 direction of negative infinity. These are currently enabled as GNU
1431 Security related changes:
1433 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1434 could result in a stack overflow when getnetbyname was called with an
1435 overly long name. (CVE-2016-3075)
1437 * Previously, getaddrinfo copied large amounts of address data to the stack,
1438 even after the fix for CVE-2013-4458 has been applied, potentially
1439 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1440 instead. Reported by Michael Petlan. (CVE-2016-3706)
1442 * The glob function suffered from a stack-based buffer overflow when it was
1443 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1444 Reported by Alexander Cherepanov. (CVE-2016-1234)
1446 * The Sun RPC UDP client could exhaust all available stack space when
1447 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1448 alloca plugin for GCC. (CVE-2016-4429)
1450 * The IPv6 name server management code in libresolv could result in a memory
1451 leak for each thread which is created, performs a failing naming lookup,
1452 and exits. Over time, this could result in a denial of service due to
1453 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1455 The following bugs are resolved with this release:
1457 [1170] localedata: ne_NP: update Nepali locale definition file
1458 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1460 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1461 [6796] math: fdim() does not set errno on overflow
1462 [10354] libc: posix_spawn should use vfork() in more cases than presently
1463 [11213] localedata: localedata: add copyright disclaimer to locale files
1464 [12143] localedata: chr_US: new Cherokee locale
1465 [12450] localedata: sgs_LT: new locale
1466 [12676] localedata: ln_CD: new locale
1467 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1469 [13304] math: fma, fmaf, fmal produce wrong results
1470 [14259] build: --localedir arg to configure is ignored
1471 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1472 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1473 [14934] localedata: es_CL: wrong first weekday chilean locale
1474 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1476 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1478 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1479 [15368] nptl: raise() is not async-signal-safe
1480 [15479] math: ceil, floor, round and trunc raise inexact exception
1481 [15578] localedata: kk_KZ: various updates
1482 [16003] localedata: pap_AN: punt old locale
1483 [16137] localedata: iw_IL: punt old locale
1484 [16190] localedata: eo: new esperanto locale
1485 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1486 [16742] malloc: race condition: pthread_atfork() called before first
1487 malloc() results in unexpected locking behaviour/deadlocks
1488 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1490 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1491 [17565] localedata: pt_PT: wrong (work-)week start
1492 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1493 [17950] build: Build fails with -msse
1494 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1495 [18433] libc: posix_spawn does not return correctly upon failure to
1497 [18453] localedata: charmaps/IBM875: incorrect codes
1498 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1500 [18896] localedata: he_IL: improvements for currency
1501 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1502 Romanian locale data
1503 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1505 [19056] libc: Deprecate readdir_r
1506 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1508 [19198] localedata: nl_NL: small improvements for Dutch locales
1509 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1510 nameservers (CVE-2016-5417)
1511 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1513 [19400] locale: Language missing in "iso-639.def", trivial fix in
1515 [19431] malloc: Deadlock between fflush, getdelim, and fork
1516 [19505] libc: Incorrect file descriptor validity checks in
1517 posix_spawn_file_actions_add{open,close,dup2}
1518 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1519 when using RTLD_NEXT
1520 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1521 `intl/{gettextP,loadinfo}.h'
1522 [19534] libc: execle, execlp may use malloc
1523 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1524 [19573] network: res_nclose and __res_maybe_init disagree about name
1525 server initialization, breaking Hesiod
1526 [19575] localedata: Status of GB18030 tables
1527 [19581] localedata: sr_* date_fmt string contains additional newline
1528 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1530 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1531 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1532 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1533 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1535 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1537 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1539 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1540 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1541 [19633] locale: strfmon_l applies global locale to number formatting
1542 [19642] network: Memory leak in getnameinfo
1543 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1544 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1546 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1547 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1549 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1550 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1552 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1554 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1556 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1557 pointers and lengths in error-case.
1558 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1559 report errors on UTF-16 surrogates values.
1560 [19755] nscd: nscd assertion failure in gc
1561 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1562 [19759] libc: mempcpy shouldn't be inlined
1563 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1564 [19765] libc: s390 needs an optimized mempcpy
1565 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1566 NAME_MAX limit assumption (CVE-2016-1234)
1567 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1568 [19787] network: Missing and incorrect truncation checks in getnameinfo
1569 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1571 [19791] network: Assertion failure in res_query.c with un-connectable name
1573 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1574 [19822] math: libm.so install clobbers old version
1575 [19825] network: resolv: send_vc can return uninitialized data in second
1576 response to getaddrinfo
1577 [19830] network: nss_dns: should check RDATA length against buffer length
1578 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1579 confronted with A/AAAA records of invalid size
1580 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1581 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1582 [19853] stdio: Printing IBM long double in decimal with high precision is
1584 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1585 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1586 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1588 [19865] network: Assertion failure or memory leak in
1589 _nss_dns_getcanonname_r
1590 [19868] network: nss_dns: netent code does not skip over non-PTR records
1591 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1593 [19881] string: Improve x86-64 memset
1594 [19907] string: Incorrect memcpy tests
1595 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1597 [19925] libc: termios.h XCASE namespace
1598 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1599 [19929] libc: limits.h NL_NMAX namespace
1600 [19931] stdio: Memory leak in vfprintf
1601 [19957] libc: clone(CLONE_VM) access invalid parent memory
1602 [19963] localedata: en_IL: New locale
1603 [19989] stdio: stdio.h cuserid namespace
1604 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1606 [19996] locale: langinfo.h nl_langinfo_l namespace
1607 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1608 [20010] network: getaddrinfo: Stack overflow in hostent translation
1610 [20012] stdio: libio: fmemopen append mode failure
1611 [20014] stdio: stdio.h namespace for pre-threads POSIX
1612 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1613 [20023] libc: fcntl.h timespec namespace
1614 [20024] math: [x86_64] vectorized sincos trashes the stack
1615 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1616 [20041] time: sys/time.h timespec namespace
1617 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1618 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1619 [20051] libc: ttyslot in wrong header under wrong conditions
1620 [20054] libc: gethostname not declared for XPG4
1621 [20055] libc: termios.h missing tcgetsid for XPG4
1622 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1624 [20073] libc: sys/stat.h fchmod namespace
1625 [20074] libc: stdlib.h rand_r namespace
1626 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1627 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1629 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1630 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1632 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1633 [20119] libc: Wrong mask for processors level type from CPUID
1634 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1635 AS not supporting AVX512
1636 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1638 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1639 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1641 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1642 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1643 [20178] libc: posix_spawn{p} should not call exit
1644 [20191] stdio: libio: vtables hardening
1645 [20195] string: FMA4 detection requires CPUID execution with register
1647 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1648 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1650 [20212] math: acos (sNaN) returns sNaN
1651 [20213] math: asin (sNaN) returns sNaN
1652 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1653 [20218] math: [i386] asinhl (sNaN) returns sNaN
1654 [20219] math: [i386] atanhl (sNaN) returns sNaN
1655 [20222] stdio: fopencookie: Mangle function pointers
1656 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1657 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1658 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1660 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1661 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1662 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1663 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1664 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1665 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1666 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1667 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1668 [20240] math: modf (sNaN) returns sNaN
1669 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1670 [20250] math: frexp (sNaN) returns sNaN
1671 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1672 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1674 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1675 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1676 subscript is below array bounds [-Werror=array-bounds]
1677 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1678 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1680 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1681 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1683 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1684 "invalid" exceptions
1685 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1686 [20316] localedata: id_ID: Februari instead of Pebruari
1687 [20327] string: POWER8 strcasecmp returns incorrect result
1688 [20347] math: Failure: Test: j0_downward (0xap+0)
1689 [20348] libc: FAIL: misc/tst-preadvwritev64
1690 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1691 [20350] libc: There is no test for p{read,write}64
1692 [20357] math: Incorrect cos result for 1.5174239687223976
1693 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1697 * Unicode 8.0.0 Support: Character encoding, character type info, and
1698 transliteration tables are all updated to Unicode 8.0.0, using new
1699 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1700 These updates cause user visible changes, such as the fixes for bugs
1701 89, 16061, and 18568.
1703 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1704 kernel-internal CPU set size. This means that requests that change the
1705 CPU affinity which failed before (for example, an all-ones CPU mask) will
1706 now succeed. Applications that need to determine the effective CPU
1707 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1708 setting it because the kernel can adjust it (and the previous size check
1709 would not detect this in the majority of cases).
1711 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1712 the following new symbols are used: fts64_children, fts64_close,
1713 fts64_open, fts64_read and fts64_set.
1715 * getaddrinfo now detects certain invalid responses on an internal netlink
1716 socket. If such responses are received, an affected process will
1717 terminate with an error message of "Unexpected error <number> on netlink
1718 descriptor <number>" or "Unexpected netlink response of size <number> on
1719 descriptor <number>". The most likely cause for these errors is a
1720 multi-threaded application which erroneously closes and reuses the netlink
1721 file descriptor while it is used by getaddrinfo.
1723 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1724 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1725 unnecessary serialization of memory allocation requests across threads.
1726 The defect is now corrected. Users should see a substantial increase in
1727 the concurent throughput of allocation requests for applications which
1728 trigger this bug. Affected applications typically create and
1729 destroy threads frequently. (Bug 19048 was reported and analyzed by
1732 * There is now a --disable-timezone-tools configure option for disabling the
1733 building and installing of the timezone related utilities (zic, zdump, and
1734 tzselect). This is useful for people who build the timezone data and code
1735 independent of the GNU C Library.
1737 * The obsolete header <regexp.h> has been removed. Programs that require
1738 this header must be updated to use <regex.h> instead.
1740 * The obsolete functions bdflush, create_module, get_kernel_syms,
1741 query_module and uselib are no longer available to newly linked binaries;
1742 the header <sys/kdaemon.h> has been removed. These functions and header
1743 were specific to systems using the Linux kernel and could not usefully be
1744 used with the GNU C Library on systems with version 2.6 or later of the
1747 * Optimized string, wcsmbs and memory functions for IBM z13.
1748 Implemented by Stefan Liebler.
1750 * Newly linked programs that define a variable called signgam will no longer
1751 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1752 require signgam to be set by those functions must ensure that they use the
1753 variable provided by the GNU C Library and declared in <math.h>, without
1754 defining their own copy.
1756 * The minimum GCC version that can be used to build this version of the GNU
1757 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1758 still be used to compile programs using the GNU C Library.
1760 Security related changes:
1762 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1763 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1765 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1766 the pointer guard feature. It is always enabled. Previously,
1767 LD_POINTER_GUARD could be used to disable security hardening in binaries
1768 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1771 * An integer overflow in hcreate and hcreate_r could lead to an
1772 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1774 * The catopen function no longer has unbounded stack usage. Reported by
1775 Max. (CVE-2015-8779)
1777 * The nan, nanf and nanl functions no longer have unbounded stack usage
1778 depending on the length of the string passed as an argument to the
1779 functions. Reported by Joseph Myers. (CVE-2014-9761)
1781 * A stack-based buffer overflow was found in libresolv when invoked from
1782 libnss_dns, allowing specially crafted DNS responses to seize control
1783 of execution flow in the DNS client. The buffer overflow occurs in
1784 the functions send_dg (send datagram) and send_vc (send TCP) for the
1785 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1786 family. The use of AF_UNSPEC triggers the low-level resolver code to
1787 send out two parallel queries for A and AAAA. A mismanagement of the
1788 buffers used for those queries could result in the response of a query
1789 writing beyond the alloca allocated buffer created by
1790 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1791 the overflow. Thanks to the Google Security Team and Red Hat for
1792 reporting the security impact of this issue, and Robert Holiday of
1793 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1795 The following bugs are resolved with this release:
1797 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1798 [887] math: Math library function "logb" and "nextafter" inconsistent
1799 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1800 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1801 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1802 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1804 [4404] localedata: German translation of "Alarm clock" is misleading
1805 [6799] math: nextafter() and nexttoward() doen't set errno on
1806 overflow/underflow errors
1807 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1809 [10432] nis: _nss_nis_setnetgrent assertion failure
1810 [11460] libc: fts has no LFS support
1811 [12926] network: getaddrinfo()/make_request() may spin forever
1812 [13065] nptl: Race condition in pthread barriers
1813 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1814 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1816 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1817 [14912] libc: Rename non-installed bits/*.h headers
1818 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1819 [15367] math: Let gcc use __builtin_isinf
1820 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1821 [15421] math: lgamma wrongly sets signgam for ISO C
1822 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1823 FE_INVALID with argument out of range
1824 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1826 [15786] dynamic-link: ifunc resolver functions can smash function
1828 [15918] math: Unnecessary check for equality in hypotf()
1829 [16061] localedata: Review / update transliteration data
1830 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1832 [16141] time: strptime %z offset restriction
1833 [16171] math: drem should be alias of remainder
1834 [16296] math: fegetround is pure?
1835 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1836 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1838 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1839 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1841 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1842 "Exception "Inexact" set" on ppc32
1843 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1844 [16517] math: Missing underflow exception from tanf/tan/tanl
1845 [16519] math: Missing underflow exception from sinhf
1846 [16520] math: Missing underflow exception from tanhf
1847 [16521] math: Missing underflow exception from exp2
1848 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1850 [16734] stdio: fopen calls mmap to allocate its buffer
1851 [16961] math: nan function incorrect handling of bad sequences
1852 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1853 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1854 [16985] locale: localedef: confusing error message when opening output
1856 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1857 [17197] locale: Redundant shift character in iconv conversion output at
1859 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1860 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1861 [17250] dynamic-link: static linking breaks nss loading
1862 (getaddrinfo/getpwnam/etc...)
1863 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1865 [17441] math: isnan() should use __builtin_isnan() in GCC
1866 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1867 (related to lock elision)
1868 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1869 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1870 [17887] time: strptime should be able to parse "+01:00" style timezones
1871 [17905] libc: catopen() Multiple unbounded stack allocations
1873 [18084] libc: backtrace (..., 0) dumps core on x86
1874 [18086] libc: nice() sets errno to 0 on success
1875 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1876 is too large (CVE-2015-8778)
1877 [18251] dynamic-link: SONAME missing when audit modules provides path
1878 [18265] libc: add attributes for wchar string and memory functions
1879 [18370] math: csqrt missing underflows
1880 [18421] libc: [hppa] read-only segment has dynamic relocations
1881 [18472] libc: Obsolete syscall wrappers should be compat symbols
1882 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1883 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1884 [18525] localedata: Remove locale timezone information
1885 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1886 [18568] localedata: Update locale data to Unicode 8.0
1887 [18589] locale: sort-test.sh fails at random
1888 [18595] math: ctan, ctanh missing underflows
1889 [18604] libc: assert macro-expands its argument
1890 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1891 contains a vector instruction exception.
1892 [18611] math: j1, jn missing errno setting on underflow
1893 [18618] localedata: sync Chechen locale definitions with other *_RU
1895 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1896 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1897 [18665] network: In send_dg, the recvfrom function is NOT always using the
1898 buffer size of a newly created buffer (CVE-2015-7547)
1899 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1901 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1903 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1904 [18699] math: tilegx cproj() for various complex infinities does not yield
1906 [18724] libc: Harden put*ent functions against data injection
1907 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1908 [18755] build: build errors with -DNDEBUG
1909 [18757] stdio: fmemopen fails to set errno on failure
1910 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1912 [18781] libc: openat64 lacks O_LARGEFILE
1913 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1914 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1915 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1916 [18790] math: [ldbl-128ibm] tanhl inaccurate
1917 [18795] libc: stpncpy fortification misses buffer lengths that are
1918 statically too large
1919 [18796] build: build fails for --disable-mathvec
1920 [18803] math: hypot missing underflows
1921 [18820] stdio: fmemopen may leak memory on failure
1922 [18823] math: csqrt spurious underflows
1923 [18824] math: fma spurious underflows
1924 [18825] math: pow missing underflows
1925 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1926 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1927 [18870] build: sem_open.c fails to compile with missing symbol
1929 [18872] stdio: Fix memory leak in printf_positional
1930 [18873] libc: posix_fallocate overflow check ineffective
1931 [18875] math: Excess precision leads incorrect libm
1932 [18877] libc: arm: mmap offset regression
1933 [18887] libc: memory corruption when using getmntent on blank lines
1934 [18918] localedata: hu_HU: change time to HH:MM:SS format
1935 [18921] libc: Regression: extraneous stat() and fstat() performed by
1937 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1938 binaries (CVE-2015-8777)
1939 [18951] math: tgamma missing underflows
1940 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1942 [18953] localedata: lt_LT: change currency symbol to the euro
1943 [18956] math: powf inaccuracy
1944 [18961] math: [i386] exp missing underflows
1945 [18966] math: [i386] exp10 missing underflows
1946 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1947 [18969] build: multiple string test failures due to missing locale
1949 [18970] libc: Reference of pthread_setcancelstate in libc.a
1950 [18977] math: float / long double Bessel functions not in XSI POSIX
1951 [18980] math: i386 libm functions return with excess range and precision
1952 [18981] math: i386 scalb*, ldexp return with excess range and precision
1953 [18982] stdio: va_list and vprintf
1954 [18985] time: Passing out of range data to strftime() causes a segfault
1956 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1957 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1958 [19012] locale: iconv_open leaks memory on error path
1959 [19016] math: clog, clog10 inaccuracy
1960 [19018] nptl: Mangle function pointers in tls_dtor_list
1961 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1962 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1963 [19048] malloc: malloc: arena free list can become cyclic, increasing
1965 [19049] math: [powerpc] erfc incorrect zero sign
1966 [19050] math: [powerpc] log* incorrect zero sign
1967 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1968 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1969 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1970 [19074] network: Data race in _res_hconf_reorder_addrs
1971 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1972 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1973 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1975 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1977 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1978 [19086] manual: posix_fallocate64 documented argument order is wrong.
1979 [19088] math: lround, llround missing exceptions close to overflow
1981 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1982 [19095] math: dbl-64 lrint incorrect for 64-bit long
1983 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1984 [19124] dynamic-link: ld.so failed to build with older assmebler
1985 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1986 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1987 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1988 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1989 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1990 pthread_setaffinity_np
1991 [19156] math: [ldbl-128] j0l spurious underflows
1992 [19164] nptl: tst-getcpu fails with many possible CPUs
1993 [19168] math: math/test-ildoubl and math/test-ldouble failure
1994 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1995 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1997 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1998 (FE_NOMASK_ENV) do not clear SSE exceptions
1999 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2000 _int_new_arena/reused_arena
2001 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2002 [19201] math: dbl-64 remainder incorrect sign of zero result
2003 [19205] math: bits/math-finite.h conditions do not match math.h and
2005 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2006 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2007 for C99-based standards
2008 [19212] libc: features.h not -Wundef clean
2009 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2011 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2012 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2013 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2015 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2016 "inexact" exceptions
2017 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2019 [19242] libc: strtol incorrect in Turkish locales
2020 [19243] malloc: reused_arena can pick an arena on the free list, leading
2021 to an assertion failure and reference count corruption
2022 [19253] time: tzset() ineffective when temporary TZ did not include DST
2024 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2025 [19270] math: [hppa] Shared libm missing __isnanl
2026 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2027 MAP_HUGETLB and MAP_STACK defines
2028 [19313] nptl: Wrong __cpu_mask for x32
2029 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2030 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2031 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2032 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2033 [19363] time: x32: times() return value wrongly truncates/sign extends
2035 [19367] dynamic-link: Improve branch prediction on Silvermont
2036 [19369] network: Default domain name not reset by res_ninit when "search"
2037 / "domain" entry is removed from resolv.conf
2038 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2040 [19385] localedata: bg_BG: time separator should be colon, not comma
2041 [19408] libc: linux personality syscall wrapper may erroneously return an
2042 error on 32-bit architectures
2043 [19415] libc: dladdr returns wrong names on hppa
2044 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2045 IBM905, IBM907, and IBM909
2046 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2047 [19443] build: build failures with -DDEBUG
2048 [19451] build: Make check fails on test-double-vlen2
2049 [19462] libc: Glibc failed to build with -Os
2050 [19465] math: Wrong code with -Os
2051 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2053 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2055 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2056 [19486] math: S390: Math tests fail with "Exception Inexact set".
2057 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2058 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2060 [19590] math: Fail to build shared objects that use libmvec.so functions.
2064 * The following bugs are resolved with this release:
2066 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2067 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2068 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2069 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2070 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2071 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2072 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2073 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2074 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2075 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2076 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2077 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2078 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2079 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2080 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2081 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2082 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2083 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2084 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2085 18657, 18676, 18694, 18696, 18887.
2087 * Cache information can be queried via sysconf() function on s390 e.g. with
2088 _SC_LEVEL1_ICACHE_SIZE as argument.
2090 * A buffer overflow in gethostbyname_r and related functions performing DNS
2091 requests has been fixed. If the NSS functions were called with a
2092 misaligned buffer, the buffer length change due to pointer alignment was
2093 not taken into account. This could result in application crashes or,
2094 potentially arbitrary code execution, using crafted, but syntactically
2095 valid DNS responses. (CVE-2015-1781)
2097 * The time zone file parser has been made more robust against crafted time
2098 zone files, avoiding heap buffer overflows related to the processing of
2099 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2100 large time zone data files. Overly long time zone specifiers in the TZ
2101 variable no longer result in stack overflows and crashes.
2103 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2104 for LD and GD on x86 and x86-64, has been implemented. You will need
2105 binutils-2.24 or later to enable this optimization.
2107 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2108 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2109 Hat). These updates cause user visible changes, such as the fix for bug
2112 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2113 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2114 condition in some applications.
2116 * Added vector math library named libmvec with the following vectorized x86_64
2117 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2119 The library can be disabled with --disable-mathvec. Use of the functions is
2120 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2121 Shared library libmvec.so is linked in as needed when using -lm (no need to
2122 specify -lmvec explicitly for not static builds).
2123 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2125 * A new fmemopen implementation has been added with the goal of POSIX
2126 compliance. The new implementation fixes the following long-standing
2127 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2128 old implementation is still present for use be by existing binaries.
2130 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2131 releases. It has been fixed to match 2.19 and older, but binaries built
2132 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2134 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2135 Contributed by Roland McGrath (Google).
2137 * The header <regexp.h> is deprecated, and will be removed in a future
2138 release. Use of this header will trigger a deprecation warning.
2139 Application developers should update their code to use <regex.h> instead.
2141 This header was formerly part of SUS, but was deprecated in 1994 and
2142 removed from the standard in 2001. Also, the glibc implementation
2143 leaks memory. See BZ#18681 for more details.
2147 * The following bugs are resolved with this release:
2149 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2150 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2151 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2152 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2153 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2154 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2155 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2156 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2157 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2158 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2161 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2162 too little memory for the to-be-scanned arguments and overflow the
2163 allocated buffer. The implementation now correctly computes the required
2164 buffer size when using malloc, and switches to malloc from alloca as
2167 * A new semaphore algorithm has been implemented in generic C code for all
2168 machines. Previous custom assembly implementations of semaphore were
2169 difficult to reason about or ensure that they were safe. The new version
2170 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2171 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2172 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2174 * Port to Altera Nios II has been contributed by Mentor Graphics.
2176 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2177 implementations for powerpc64/powerpc64le.
2178 Implemented by Adhemerval Zanella (IBM).
2180 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2181 and powerpc64le. This may improve lock scaling of existing programs on
2182 HTM capable systems. The lock elision code is only enabled with
2183 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2184 powerpc will issue a transaction abort on every syscall to avoid side
2185 effects being visible outside transactions.
2187 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2188 AArch64. Contributed by ARM Ltd.
2190 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2192 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2193 under certain input conditions resulting in the execution of a shell for
2194 command substitution when the application did not request it. The
2195 implementation now checks WRDE_NOCMD immediately before executing the
2196 shell and returns the error WRDE_CMDSUB as expected.
2198 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2199 processing format strings with a large number of format specifiers.
2201 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2202 infinite loop if the DNS response contained a PTR record of an unexpected
2205 * The minimum GCC version that can be used to build this version of the GNU
2206 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2207 still be used to compile programs using the GNU C Library.
2209 * The GNU C Library is now built with -Werror by default. This can be
2210 disabled by configuring with --disable-werror.
2212 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2214 * The obsolete sigvec function has been removed. This was the original
2215 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2216 programs have been using instead for about 25 years. Of course, ABI
2217 compatibility for old binaries using sigvec remains intact.
2219 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2220 with newer versions of bison.
2222 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2223 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2224 registers overlay two consecutive single-precision registers. MIPS32R2
2225 introduced a new FPU mode (FR=1) where double-precision registers extend the
2226 corresponding single-precision registers which is incompatible with the
2227 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2228 require the use of FR=1 making a transition necessary. New o32 ABI
2229 extensions enable users to migrate over time from the original o32 ABI
2230 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2231 tracks the ABI of any loaded object and verifies that new objects are
2232 compatible. Mode transitions will also be requested as required and
2233 unsupportable objects will be rejected. The ABI checks include both soft and
2234 hard float ABIs for o32, n32 and n64.
2236 GCC 5 with GNU binutils 2.25 onwards:
2237 It is strongly recommended that all o32 system libraries are built using the
2238 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2239 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2240 compiler using --with-fp-32=xx to set this by default.
2244 * The following bugs are resolved with this release:
2246 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2247 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2248 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2249 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2250 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2251 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2252 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2253 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2254 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2255 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2256 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2257 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2258 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2259 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2260 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2262 * Reverted change of ABI data structures for s390 and s390x:
2263 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2264 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2265 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2266 are preserved pointing straight to the same implementation as the old ones.
2267 Given that, new callers will simply provide a too-big buffer to these
2268 functions. Any applications/libraries out there that embed jmp_buf or
2269 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2270 against 2.19 headers will have to rebuilt again. This is necessary in any
2271 case to revert the breakage in their ABI caused by the glibc change.
2273 * Support for file description locks is added to systems running the
2274 Linux kernel. The standard file locking interfaces are extended to
2275 operate on file descriptions, not file descriptors, via the use of
2276 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2277 are associated with an open file instead of a process.
2279 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2281 * The minimum Linux kernel version that this version of the GNU C Library
2282 can be used with is 2.6.32.
2284 * Running the testsuite no longer terminates as soon as a test fails.
2285 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2286 with PASS or FAIL lines for individual tests. A summary of the results is
2287 printed, including a list of failing lists, and "make check" exits with
2288 error status if there were any unexpected failures. "make check
2289 stop-on-test-failure=y" may be used to keep the old behavior.
2291 * The am33 port, which had not worked for several years, has been removed
2294 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2295 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2296 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2297 conflicted with POSIX (support for which was removed in 2.19), the
2298 interfaces those macros enabled remain available when compiling with
2299 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2300 test macros defined.
2302 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2304 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2305 This may improve lock scaling of existing programs on TX capable systems.
2306 The lock elision code is only built with --enable-lock-elision=yes and
2307 then requires a GCC version supporting the TX builtins. With lock elision
2308 default mutexes are elided via __builtin_tbegin, if the cpu supports
2309 transactions. By default lock elision is not enabled and the elision code
2312 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2313 copy the path argument. This allowed programs to cause posix_spawn to
2314 deference a dangling pointer, or use an unexpected pathname argument if
2315 the string was modified after the posix_spawn_file_actions_addopen
2318 * All supported architectures now use the main glibc sysdeps directory
2319 instead of some being in a separate "ports" directory (which was
2320 distributed separately before glibc 2.17).
2322 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2323 On configurations that support it (all Linux configurations), it's now
2324 used regardless of the --enable-add-ons switch to configure. It is no
2325 longer possible to build such configurations without pthreads support.
2327 * Locale names, including those obtained from environment variables (LANG
2328 and the LC_* variables), are more tightly checked for proper syntax.
2329 setlocale will now fail (with EINVAL) for locale names that are overly
2330 long, contain slashes without starting with a slash, or contain ".." path
2331 components. (CVE-2014-0475) Previously, some valid locale names were
2332 silently replaced with the "C" locale when running in AT_SECURE mode
2333 (e.g., in a SUID program). This is no longer necessary because of the
2336 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2337 with application code using Intel MPX instructions. (With all previous
2338 versions, the MPX register state could be clobbered when making calls
2339 into or out of a shared library.) Note that while the new dynamic
2340 linker is compatible with all known x86 hardware whether or not it
2341 supports Intel MPX, some x86 instruction-set emulators might fail to
2342 handle the new instruction encodings. This is known to affect Valgrind
2343 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2344 release), and might affect other tools that do instruction emulation.
2346 * Support for loadable gconv transliteration modules has been removed.
2347 The support for transliteration modules has been non-functional for
2348 over a decade, and the removal is prompted by security defects. The
2349 normal gconv conversion modules are still supported. Transliteration
2350 with //TRANSLIT is still possible, and the //IGNORE specifier
2351 continues to be supported. (CVE-2014-5119)
2353 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2354 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2355 resulting a denial-of-service security vulnerability in applications which
2356 use functions related to iconv. (CVE-2014-6040)
2360 * The following bugs are resolved with this release:
2362 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2363 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2364 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2365 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2366 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2367 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2368 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2369 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2370 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2371 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2372 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2373 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2374 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2375 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2376 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2377 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2378 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2379 16431, 16453, 16474, 16506, 16510, 16529
2381 * Slovenian translations for glibc messages have been contributed by the
2382 Translation Project's Slovenian team of translators.
2384 * The public headers no longer use __unused nor __block. This change is to
2385 support compiling programs that are derived from BSD sources and use
2386 __unused internally, and to support compiling with Clang's -fblock
2387 extension which uses __block.
2389 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2390 large collation sequences to optimize multiple passes. This cache
2391 computation may overflow for large collation sequences and may cause a
2392 stack or buffer overflow. This is now fixed to use a slower algorithm
2393 which does not use a cache if there is an integer overflow.
2395 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2396 rules for large collation sequences to optimize multiple passes and falls
2397 back to alloca if malloc fails, resulting in a possible stack overflow.
2398 The implementation now falls back to an uncached collation sequence lookup
2401 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2402 initialized for static applications resulting in the security feature
2403 being disabled. The pointer guard is now correctly initialized to a
2404 random value for static applications. Existing static applications need
2405 to be recompiled to take advantage of the fix (bug 15754).
2407 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2408 to the d_name member of struct dirent, or omit the terminating NUL
2409 character. (Bugzilla #14699).
2411 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2412 aligned_alloc functions could allocate too few bytes or corrupt the
2413 heap when passed very large allocation size values (Bugzilla #15855,
2416 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2417 for AF_INET6 has been fixed (Bugzilla #16072).
2419 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2420 pap_CW, quz_PE, the_NP.
2422 * Substantially revised locales: gd_GB, ht_HT
2424 * The LC_ADDRESS field was updated to support country_car for almost all
2427 * ISO 1427 definitions were updated.
2429 * ISO 3166 definitions were updated.
2431 * The localedef utility now supports --big-endian and --little-endian
2432 command-line options to generate locales for a different system from that
2433 for which the C library was built.
2435 * Binary locale files now only depend on the endianness of the system for
2436 which they are generated and not on other properties of that system. As a
2437 consequence, binary files generated with new localedef may be incompatible
2438 with old versions of the GNU C Library, and binary files generated with
2439 old localedef may be incompatible with this version of the GNU C Library,
2440 in the following circumstances:
2442 + Locale files may be incompatible on m68k systems.
2444 + Locale archive files (but not separate files for individual locales) may
2445 be incompatible on systems where plain "char" is signed.
2447 * The configure option --disable-versioning has been removed. Builds with
2448 --disable-versioning had not worked for several years.
2450 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2451 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2453 * SystemTap probes for malloc have been introduced.
2455 * SystemTap probes for slow multiple precision fallback paths of
2456 transcendental functions have been introduced.
2458 * Support for powerpc64le has been added.
2460 * The soft-float powerpc port now supports e500 processors.
2462 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2464 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2465 set of header declarations that are enabled by default, even when other
2466 feature test macros or compiler options such as -std=c99 would otherwise
2467 disable some of those declarations.
2469 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2470 conflict with POSIX. The libbsd-compat library (which was a dummy library
2471 that did nothing) has also been removed.
2473 * Preliminary documentation about Multi-Thread, Async-Signal and
2474 Async-Cancel Safety has been added.
2476 * Change of ABI data structures for s390 and s390x:
2477 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2478 allow for future hardware extensions. All library functions that accept or
2479 return these structures were versioned in order to provide backwards
2480 compatibility with existing code. However, not all cases can be handled
2481 with interface versioning. If an application embeds either structure into
2482 another structure and that structure is passed to another compilation unit
2483 which expects the newer larger-sized object then any access to the new
2484 fields will result in undefined behaviour. Similarly any access to fields
2485 that were moved by the enlarging of the structures will cause undefined
2486 behaviour. To fix the undefined behaviour all objects that are part of the
2487 final in-memory image must agree on the size of structures, and this may
2488 require recompilation.
2492 * The following bugs are resolved with this release:
2494 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2495 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2496 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2497 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2498 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2499 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2500 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2501 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2502 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2503 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2504 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2505 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2506 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2509 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2510 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2511 Distributions can re-enable building and using pt_chown via the new configure
2512 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2513 considerable security risks and should only be used if the distribution
2514 understands and accepts the risks.
2516 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2519 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2520 fixed (Bugzilla #15330).
2522 * Add support for calling C++11 thread_local object destructors on thread
2523 and program exit. This needs compiler support for offloading C++11
2524 destructor calls to glibc.
2526 * Improved worst case performance of libm functions with double inputs and
2529 * Support for priority inherited mutexes in pthread condition variables on
2530 non-x86 architectures.
2532 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2534 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2536 * Optimized string functions for ARM. Implemented by Will Newton and
2539 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2541 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2544 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2545 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2547 * Added a benchmark framework to track performance of functions in glibc.
2549 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2550 It is based on draft TS 18661 and currently enabled as a GNU extension.
2552 * On Linux, the clock function now uses the clock_gettime system call
2553 for improved precision, rather than old times system call.
2555 * Added support for version-3 tz format files. This is needed when using
2556 the tz database release 2013e or later, and affects a few unusual cases --
2557 currently only TZ='America/Godthab' for time stamps after 2037.
2559 * Added new API functions pthread_getattr_default_np and
2560 pthread_setattr_default_np to get and set the default pthread
2561 attributes of a process.
2563 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2564 This may improve lock scaling of existing programs on TSX capable systems.
2565 When the --enable-lock-elision=yes parameter is specified at configure
2566 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2569 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2570 availability). Implemented by Ryan S. Arnold.
2572 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2576 * The following bugs are resolved with this release:
2578 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2579 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2580 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2581 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2582 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2583 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2584 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2585 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2586 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2587 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2588 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2589 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2590 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2591 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2592 14879, 14889, 14893, 14898, 14914.
2594 * Optimization of memcpy for MIPS.
2596 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2597 EMFILE has been fixed (Bugzilla #14889).
2599 * The add-on ports collection is now distributed in the "ports" subdirectory
2600 of the main GNU C Library distribution, rather than separately.
2602 * Port to ARM AArch64 contributed by Linaro.
2604 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2605 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2607 Implemented by Andreas Krebbel.
2609 * The new function secure_getenv allows secure access to the environment,
2610 returning NULL if running in a SUID/SGID process. This function replaces
2611 the internal function __secure_getenv.
2613 * SystemTap static probes have been added into the dynamic linker.
2614 Implemented by Gary Benson.
2616 * Optimizations of string functions strstr, strcasestr and memmem.
2617 Implemented by Maxim Kuvyrkov.
2619 * The minimum Linux kernel version that this version of the GNU C Library
2620 can be used with is 2.6.16.
2622 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2623 powerpc POWER7. Implemented by Will Schmidt.
2625 * New configure option --disable-nscd builds the C library such that it
2626 never attempts to contact the Name Service Caching Daemon (nscd).
2627 New configure option --disable-build-nscd avoids building nscd itself;
2628 this is the default if --disable-nscd is used.
2630 * Improved support for cross-compilation, including cross-testing and
2631 bootstrap builds without a previously built glibc.
2633 * Several testsuite tests are now able to test multiple IFUNC variants of an
2634 interface, rather than just testing the one that would be chooen by
2637 * New configure options --with-bugurl and --with-pkgversion, for
2638 distributors to use to embed their bug-reporting and package version
2639 information in --help and --version output.
2641 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2642 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2643 allows creation of chroots without the procfs mounted on /proc.
2645 * The `crypt' function now fails if passed salt bytes that violate the
2646 specification for those values. On Linux, the `crypt' function will
2647 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2648 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2649 when the mode is enabled.
2651 * The `clock_*' suite of functions (declared in <time.h>) is now available
2652 directly in the main C library. Previously it was necessary to link with
2653 -lrt to use these functions. This change has the effect that a
2654 single-threaded program that uses a function such as `clock_gettime' (and
2655 is not linked with -lrt) will no longer implicitly load the pthreads
2656 library at runtime and so will not suffer the overheads associated with
2657 multi-thread support in other code such as the C++ runtime library.
2659 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2664 * The following bugs are resolved with this release:
2666 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2667 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2668 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2669 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2670 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2671 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2672 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2673 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2674 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2675 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2676 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2677 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2678 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2679 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2680 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2681 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2682 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2683 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2684 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2685 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2686 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2687 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2690 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2691 configuring glibc with:
2692 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2693 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2694 Implemented by H.J. Lu.
2698 + define static_assert
2700 + do not declare gets
2702 + declare at_quick_exit and quick_exit also for ISO C11
2704 + aligned_alloc. NB: The code is deliberately allows the size parameter
2705 to not be a multiple of the alignment. This is a moronic requirement
2706 in the standard but it is only a requirement on the caller, not the
2709 + timespec_get added
2711 + uchar.h support added
2713 + CMPLX, CMPLXF, CMPLXL added
2715 Implemented by Ulrich Drepper.
2717 * Support for the IA-64 has been moved to ports.
2719 * Remove support for anything but ELF binary format
2721 * Checking versions of poll, ppoll added.
2722 Implemented by Ulrich Drepper.
2724 * More generic and 64-bit performance optimizations to math functions.
2725 Implemented by Ulrich Drepper.
2727 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2728 headers and functions available at compile time as they were before
2729 version 2.14. This option will be removed at some time in the future
2730 after the TI-RPC library becomes fully sufficient for the needs of
2731 existing applications.
2733 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2734 Note that glibc is not expected to work with any Linux kernel version
2737 * New header <sys/auxv.h> and function getauxval allowing easy access to
2738 the AT_* key-value pairs passed from the Linux kernel. The header also
2739 defines the HWCAP_* bits associated with the AT_HWCAP key.
2741 * New locales: mag_IN
2743 * New configure option --enable-systemtap builds SystemTap static probes
2744 into libc for setjmp and longjmp and into libpthread for various operations.
2745 So far the setjmp/longjmp probes and some of the libpthread probes are
2746 provided only for i*86 and x86_64.
2747 Implemented by Roland McGrath and Rayson Ho.
2749 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2751 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2754 * More optimized functions for SPARC. Implemented by David S. Miller.
2756 * Improved support for cross-compilation, especially bootstrap builds
2757 without a previously built glibc.
2759 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2760 Chris Metcalf from Tilera.
2762 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2763 now supported for ARM processors.
2765 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2766 as the name of the dynamic linker, to distinguish it from the
2767 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2769 * Support for CRIS has been removed from ports.
2771 * A new class of installed header has been documented for low-level
2772 platform-specific functionality. PowerPC added the first instance with a
2773 function to provide time base register access. Contributed by Tulio
2774 Magno Quites Machado Filho.
2776 * ix86 configurations now install header files that are consistent with
2777 what x86-64 configurations install. These same header files can be used
2778 for -m32, -m64, or -mx32 builds.
2779 Contributed by H.J. Lu.
2781 * Math library bug fixes. A thorough audit of all open math library bugs was
2782 conducted by Joseph Myers. Significant progress was made on many math
2783 library bugs resulting in more accurate exceptions and function results.
2784 Many thanks to all those that contributed including Andreas Jaeger for his
2785 patch review and work on the x87 trigonometric instruction issues.
2787 * Timezone data is no longer installed. Timezone-related binaries and scripts
2788 will continue to be installed. Users should obtain their timezone data from
2789 their distribution provider or from the tzdata package at
2790 <ftp://munnari.oz.au/pub/>.
2794 * The following bugs are resolved with this release:
2796 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2797 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2798 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2799 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2800 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2801 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2802 13484, 13506, 13515, 13523, 13524, 13538, 13540
2804 * New program pldd to list loaded object of a process
2805 Implemented by Ulrich Drepper.
2807 * Add nss_db support back to glibc. No more dependency on Berkeley db
2808 and support for initgroups lookups.
2809 Implemented by Ulrich Drepper.
2811 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2812 Contributed by HJ Lu.
2814 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2815 Contributed by HJ Lu.
2817 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2818 on x86-32 and x86-64.
2819 Contributed by Liubov Dmitrieva.
2821 * Optimized strchr and strrchr for SSE on x86-32.
2822 Contributed by Liubov Dmitrieva.
2824 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2825 for x86-64 and x86-32.
2826 Contributed by Liubov Dmitrieva.
2828 * New interfaces: scandirat, scandirat64
2829 Implemented by Ulrich Drepper.
2831 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2832 Implemented by Ulrich Drepper.
2834 * nscd now also caches the netgroup database.
2835 Implemented by Ulrich Drepper.
2837 * Integrate libm with gcc's -ffinite-math-only option.
2838 Implemented by Ulrich Drepper.
2840 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2841 to math functions. Implemented by Ulrich Drepper.
2843 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2844 Implemented by Ulrich Drepper.
2846 * New Linux interfaces: process_vm_readv, process_vm_writev
2848 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2849 Implemented by Ulrich Drepper.
2851 * Optimized nearbyint and strcasecmp for PPC.
2852 Implemented by Adhemerval Zanella.
2854 * New locales: bho_IN, unm_US, es_CU, ta_LK
2858 * The following bugs are resolved with this release:
2860 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2861 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2862 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2863 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2864 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2865 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2866 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2867 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2868 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2869 12795, 12811, 12813, 12814, 12841
2871 * The RPC implementation in libc is obsoleted. Old programs keep working
2872 but new programs cannot be linked with the routines in libc anymore.
2873 Programs in need of RPC functionality must be linked against TI-RPC.
2874 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2876 Visible changes of this change include (obviously) the inability to link
2877 programs using RPC functions without referencing the TI-RPC library and the
2878 removal of the RPC headers from the glibc headers.
2879 Implemented by Ulrich Drepper.
2881 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2882 syncfs, setns, sendmmsg
2884 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2885 yue_HK, lij_IT, mhr_RU
2887 * New encodings: CP770, CP771, CP772, CP773, CP774
2889 * New program sotruss to trace calls through PLTs
2890 Implemented by Ulrich Drepper.
2892 * The malloc hook implementation is marked deprecated and will be removed
2893 from the default implementation in the next version. The design never
2894 worked ever since the introduction of threads. Even programs which do
2895 not create threads themselves can use multiple threads created internally.
2899 * The following bugs are resolved with this release:
2901 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2902 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2903 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2904 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2907 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2909 * POWER7 optimizations: memset, memcmp, strncmp
2911 * New optimized string functions for x86-64: strnlen (SSE2),
2912 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2913 Implemented by Ulrich Drepper.
2917 * The following bugs are resolved with this release:
2919 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2920 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2921 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2922 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2923 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2924 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2925 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2926 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2928 * New interfaces: pthread_getname_np, pthread_setname_np
2930 * New Linux interface: recvmmsg
2932 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2934 * The dynamic linker now recognizes supported ABI versions from the
2935 EI_ABIVERSION field in the ELF header.
2936 Implemented by Ulrich Drepper.
2938 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2939 will not be used to fill in password fields in the passwd.byname replies.
2940 Instead it is used to synthesize the shadow.byname table, should it be
2941 missing. This is a useful mode in some installations involving Solaris.
2942 Implemented by Ulrich Drepper.
2944 * New locales: kok_IN, sq_MK, cv_RU
2948 * The following bugs are resolved with this release:
2950 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2951 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2952 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2953 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2954 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2955 10789, 10817, 10823, 10840, 10847
2957 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2959 Implemented by Ulrich Drepper.
2961 * Checking version of longjmp added that fails if an uninitialized stack
2962 frame would be created. Implemented by Ulrich Drepper.
2964 * STT_GNU_IFUNC is now supported in static executables.
2965 Implemented by H.J. Lu.
2967 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2969 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2970 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2971 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2972 Contributed by H.J. Lu.
2974 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2975 Implemented by Ulrich Drepper.
2977 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2979 Contributed by H.J. Lu.
2981 * Support for fma instruction in AVX on x86-64.
2982 Implemented by H.J. Lu and Ulrich Drepper.
2984 * AVX support in x86-64 auditing support in ld.so.
2985 Implemented by H.J. Lu.
2987 * STB_GNU_UNIQUE support added.
2988 Implemented by Ulrich Drepper.
2990 * Implement second fallback mode for optimized DNS lookup for even more
2991 broken environments. If two requests from the same port are not
2992 handled correctly close the socket and open a new one before sending
2993 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2994 can be used to select this mode right away, instead of rediscovering the
2995 necessity is every process again.
2996 Implemented by Ulrich Drepper.
2998 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2999 Implemented by Adam Tkac.
3001 * Optimized iconv conversions for S390x.
3002 Implemented by Andreas Krebbel.
3004 * Using condvars with PI mutexes is now more efficient due to kernel
3005 support for requeueing to PI futexes. NPTL support added for x86-64.
3006 Implemented by Ulrich Drepper.
3008 * New locales: ps_AF, my_MM
3012 * The following bugs are resolved with this release:
3014 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3015 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3016 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3017 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3018 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3020 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3021 Implemented by Ulrich Drepper.
3023 * New Linux interfaces: accept4, fallocate, fallocate64.
3024 Implemented by Ulrich Drepper.
3026 * Correct declarations of string function when used in C++ code. This
3027 could lead to compile errors for invalid C++ code.
3029 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3032 * New POSIX 2008 interface: psiginfo
3033 Implemented by Ulrich Drepper.
3035 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3036 Implemented by Ulrich Drepper.
3038 * Support for selecting between multiple function definitions at runtime
3039 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3041 * The libcrypt library can now use the hash function implementations in
3042 NSS. Implemented by Ulrich Drepper.
3044 * The malloc implementation can be compiled to be less memory efficient
3045 but higher performing in multi-threaded programs.
3046 Implemented by Ulrich Drepper.
3048 * New locales: nan_TW@latin, ks_IN
3050 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3051 Implemented by Ulrich Drepper.
3053 * Extended printf hook support. It is possible to use user-defined types
3054 and extend existing format specifiers.
3055 Implemented by Ulrich Drepper.
3057 * Handling for group shadow files has been added.
3058 Implemented by Ulrich Drepper.
3060 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3061 servers (the case, e.g., for some people using the built-in DNS
3062 server in ADSL modems/routers). There is a once-per-process timeout
3063 in case of a broken server. To avoid it, users can run nscd or put
3064 'options single-request' in /etc/resolv.conf.
3065 Implemented by Ulrich Drepper.
3069 * The following bugs are resolved with this release:
3071 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3072 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3073 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3074 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3075 6968, 6974, 6980, 6995, 7008, 7009, 7029
3077 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3078 up at the same time. Implemented by Ulrich Drepper.
3080 * TLS descriptors for LD and GD on x86 and x86-64.
3081 Implemented by Alexandre Oliva.
3083 * getaddrinfo now handles DCCP and UDPlite.
3084 Implemented by Ulrich Drepper.
3086 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3087 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3088 Implemented by Ulrich Drepper.
3090 * New implementation of memmem, strstr, and strcasestr which is O(n).
3091 Implemented by Eric Blake.
3093 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3095 * Implement "e" option for popen to open file descriptor with the
3096 close-on-exec flag set. Implemented by Ulrich Drepper.
3098 * Implement "b" mode for fmemopen. In this mode writes writes don't
3099 implicitly add a NUL byte and seeks from the end of the buffer really
3100 use the buffer end, not the string length as the basis.
3101 Implemented by Ulrich Drepper.
3103 * Many functions, exported and internal, now atomically set the close-on-exec
3104 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3106 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3108 Implemented by Pravin Satpute.
3110 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3114 * The following bugs are resolved with this release:
3116 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3117 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3118 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3119 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3120 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3121 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3122 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3124 * New locales: bo_CN, bo_IN, shs_CA.
3126 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3128 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3129 Implemented by Pravin Satpute.
3131 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3133 * nscd caches now all timeouts for DNS entries
3134 Implemented by Ulrich Drepper.
3136 * nscd is more efficient and wakes up less often.
3137 Implemented by Ulrich Drepper.
3139 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3140 vdprintf, and obstack_vprintf.
3141 Implemented by Jakub Jelinek.
3143 * Faster memset for x86-64.
3144 Implemented by Harsha Jagasia and H.J. Lu.
3146 * Faster memcpy on x86.
3147 Implemented by Ulrich Drepper.
3149 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3150 Implemented by Ulrich Drepper.
3152 * Faster sqrt and sqrtf implemention for some PPC variants.
3153 Implemented by Steven Munroe.
3157 * The following bugs are resolved with this release:
3159 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3160 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3161 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3162 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3163 5071, 5103, 5104, 5112, 5113, 5184, 5186
3165 * More checking functions: fread, fread_unlocked, open*, mq_open.
3166 Implemented by Jakub Jelinek and Ulrich Drepper.
3168 * Extend fortification to C++. Implemented by Jakub Jelinek.
3170 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3171 by not recognizing 'a' as a modifier when those specs are requested.
3172 Implemented by Jakub Jelinek.
3174 * PPC optimizations to math and string functions.
3175 Implemented by Steven Munroe.
3177 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3178 options to be passed. Implemented by Ulrich Drepper.
3180 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3182 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3183 Implemented by Ulrich Drepper.
3185 * Handle private futexes in the NPTL implementation.
3186 Implemented by Jakub Jelinek and Ulrich Drepper.
3188 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3189 Implemented by Roland McGrath and Ulrich Drepper.
3191 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3193 * SHA-256 and SHA-512 based password encryption.
3194 Implemented by Ulrich Drepper.
3196 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3197 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3200 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3201 Implemented by Ulrich Drepper.
3205 * The following bugs are resolved with this release:
3207 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3208 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3209 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3210 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3211 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3212 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3213 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3216 * New Linux interfaces: epoll_pwait, sched_getcpu.
3218 * New generic interfaces: strerror_l.
3220 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3224 * The following bugs are resolved with this release:
3226 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3227 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3228 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3229 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3230 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3231 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3232 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3233 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3234 3225, 3251, 3252, 3253, 3273
3236 * For Linux, the sorting of addresses returned by getaddrinfo now also
3237 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3238 Implemented by Ulrich Drepper.
3240 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3241 Implemented by Ulrich Drepper.
3243 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3245 * New iconv module for MIK. Contributed by Alexander Shopov.
3247 * For sites with broken group and/or passwd database, the auto-propagate
3248 option of nscd can prevent creating ID lookup entries from the results
3249 of a name lookup and vice versa. This usually is no problem but some
3250 site might have problems with the default behavior.
3251 Implemented by Ulrich Drepper.
3253 * Iterating over entire database in NIS can be slow. With the
3254 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3255 to trade time for memory. The entire database will be read at once.
3256 Implemented by Ulrich Drepper.
3258 * The interfaces introduced in RFC 3542 have been implemented by
3261 * Support for the new ELF hash table format was added by Ulrich Drepper.
3263 * Support for priority inheritance mutexes added by Jakub Jelinek and
3266 * Support for priority protected mutexes added by Jakub Jelinek.
3268 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3269 or_IN, csb_PL, fy_NL, sr_ME.
3273 * More overflow detection functions.
3275 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3276 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3278 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3279 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3280 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3281 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3282 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3283 by Masahide Washizawa.
3285 * It is now possible to install audit libraries for the dynamic linker using
3286 LD_AUDIT. Implemented by Ulrich Drepper.
3288 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3289 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3290 For a libc and libpthread that works well on Linux 2.4 kernels, we
3291 recommend using the stable 2.3 branch.
3293 * The new function open_wmemstream acts like open_memstream,
3294 but uses a wchar_t wide character buffer.
3296 * The new function ppoll is an improved version of poll, similar to pselect.
3298 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3299 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3300 renameat, unlinkat, linkat, symlinkat, readlinkat.
3302 * New Linux kernel system calls: unshare,
3303 inotify_init, inotify_add_watch, inotify_rm_watch.
3305 * The euidaccess function is now also known by the alias eaccess,
3306 for compatibility with some other systems.
3308 * Timezone data updated to 2006b version.
3312 * The following bugs are resolved with this release:
3314 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3315 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3316 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3317 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3318 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3319 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3321 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3323 * As of this release, GCC 4 can be used to compile the C Library.
3325 * Timezone data updated to 2005m version.
3329 * The following bugs are resolved with this release:
3331 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3332 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3333 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3334 777, 787, 821, 822, 823, 825
3336 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3340 * Support for RFC 3678. Real implementations exist only for Linux so far.
3341 Implemented by Ulrich Drepper.
3343 * nscd can now cache entries persistently. Expiring entries are reloaded.
3344 For speedups the cache can be shared in memory with client processes.
3345 Implemented by Ulrich Drepper.
3347 * nscd can now perform SELinux checks.
3348 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3350 * getaddrinfo queries are now cached. Canonical name lookup is performed
3352 Implemented by Ulrich Drepper.
3354 * The nothrow function attribute is used when headers are used by gcc when
3355 compiling C code. This can avoid significant amounts of exception
3358 * The malloc functions perform more error checking and are stricter when
3359 it comes to reacting on errors. The default action is to terminate
3360 the process after showing an error message. Implemented by Ulrich Drepper.
3362 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3363 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3365 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3366 namespaces using the new function dlmopen(). This feature is of course,
3367 like most other dynamic loading functionality, not available in statically
3368 linked applications. Implemented by Ulrich Drepper.
3370 * Low-overhead boundary checking variants of string and some stdio functions
3371 were added. These are to be used in conjunction with a gcc patch by
3372 Jakub Jelinek which adds calls to these functions if possible.
3373 Implemented by Jakub Jelinek and Ulrich Drepper.
3375 * Old code for several operating systems and machine architectures that
3376 have not been in working condition in a long time have been removed from
3377 the main source tree maintained by the GNU C Library's maintainers.
3378 These files are now reside in the separate `ports' source module
3379 that is usable as an add-on when building the library.
3383 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3384 interrogate the dynamic linker, compatible with the Solaris interface.
3386 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3387 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3389 * getifaddrs now uses the netlink interface on Linux to get its information.
3390 Implemented by Thorsten Kukuk.
3392 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3393 Implemented by Ulrich Drepper.
3395 * support for non-executable stacks on x86 has been added. Changes mostly
3398 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3401 * getaddrinfo now performs destination address selection according to
3406 * Thread-safe interfaces for many functions that access locale data
3407 were added in version 2.3, but these features were omitted from NEWS.
3408 Many functions have variants with an `_l' suffix that take a `locale_t'
3409 object as a parameter rather than consulting the current locale.
3410 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3411 create and maintain `locale_t' objects. Additionally, the new function
3412 `uselocale' sets "the current locale" (as used by functions not so
3413 parameterized) set for an individual thread. These features were added
3414 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3416 * The functions getresuid, getresgid, setresuid, and setresgid, which
3417 have long been available on Linux, are now declared in <unistd.h>
3418 and are now also available on the Hurd.
3420 * ELF thread-local storage support (TLS) now works on x86-64.
3422 * The new dynamic string token $LIB is expanded in shared library names.
3423 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3425 * Aldy Hernandez contributed complete software floating point support for
3426 PowerPC machines with no FPU.
3428 * fexecve is implemented on Linux.
3430 * The `btowc' function should work at least twice as fast due to
3431 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3433 * With approriate thread add-ons cancelable functions are now implemented
3434 in libc.so as well. No need to call the function in libpthread. This
3435 change allowed to finally disable the incorrect and expensive handling
3436 of weak definition in ld.so.
3438 * Yet more PLT entries in libc.so have been removed. We finally arrived
3439 at the bare minimum. Startup times improved appropriately.
3441 * Support for the new Linux/x86 system call interface was added. The
3442 AT_SYSINFO auxiliary vector entry is recognized and handled.
3446 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3449 * iconv (the program and the interface) now accepts empty names (excluding
3450 options like //TRANSLIT) to mean "use charset of current locale".
3452 * localedef can now transliterate characters in strings which are not in
3453 the provided charmap. The information from the input locale is used.
3455 * Prelinking support was added for ELF targets. This requires additional
3456 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3458 * Read-only stdio streams now use mmap to speed up operation by eliminating
3459 copying and buffer underflows. To use add 'm' to the mode string of
3460 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3462 * The malloc functions were completely rewritten by Wolfram Gloger based
3463 on Doug Lea's malloc-2.7.0.c.
3465 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3466 implementation of regex.
3468 * Bruno Haible upgraded the iconv and locale implementation to support
3471 * Contents of the LC_* and LANG environment variables in the CEN style are
3472 not recognized anymore. It never was used. Change by Ulrich Drepper.
3474 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3475 thread-local storage (TLS) ABI on some platforms.
3476 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3478 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3479 EUC-JISX0213, and TSCII.
3481 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3482 BSD-compatible interface for getting all network interface addresses.
3483 Implementation for IPv4 by Roland McGrath.
3485 * Loading of locale data is faster due to the introduction of a locale
3486 archive. Implemented by Roland McGrath and Ulrich Drepper.
3488 * Startup times are significantly reduced by not using exported functions
3489 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3492 * Steven Munroe contributed a port to PowerPC64/Linux.
3496 * The Hurd now uses the GNU libio implementation of stdio rather than the
3497 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3499 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3500 with the same behavior as the Linux system call of the same name.
3504 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3505 128-bit long double format.
3507 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3508 IBM1161, and IBM1162 charsets.
3510 * Andreas Jaeger contributed a port to x86-64/Linux.
3512 * Peter Bruin contributed a port to PowerPC/Hurd.
3514 * libc functions using I/O streams now can handle wide-oriented streams
3517 * optimizations in the dynamic linker. Binaries created by recent binutils
3518 versions start up quicker due to reduced time spent on relocations.
3520 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3524 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3525 asinh, atanh, j0 for the 128-bit long double format.
3527 * Bruno Haible updated all the code handling Unicode in some form to
3528 support Unicode 3.1.
3530 * Speed of regex for single-byte locales is back to previous levels.
3531 Patch by Isamu Hasegawa.
3533 * Alpha, SPARC, and IA-64 now also using floating stacks.
3535 * Startup time of internationalized applications greatly improved through
3536 iconv cache. Use iconvconfig to generate the cache file.
3537 Contributed by Ulrich Drepper.
3539 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3540 severe performance problems. Patch by David Mosberger.
3542 * The Hurd port got a lot more functionality like AIO, various stdio
3543 extensions, etc. Mainly done by Roland McGrath.
3545 * mtrace can now lookup symbols in shared libraries.
3549 * Intel's IA-64 math library is largely integrated. It provides fast and
3550 accurate implementatations for most basic and standard math functions
3551 in float, double, and long double format.
3553 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3554 and asin for the 96-bit long double format and asin, log, tan for the
3555 128-bit long double format.
3557 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3558 The basic double functions exist today. Contributed by Abraham Ziv
3559 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3560 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3562 * An asynchronous name lookup library was added. The interface is designed
3563 after POSIX AIO. The proposal was circulated beforehand to get comments.
3564 No negative ones came in. Implemented by Ulrich Drepper.
3566 * Port to S390/64bit contributed by Martin Schwidefsky
3567 <schwidefsky@de.ibm.com>.
3569 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3570 of functions for Linux/IA-64.
3572 * The RPC code is now thread safe. Threads can now use the same service
3573 of different services at the same time. Patch by Eric Norum
3574 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3576 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3577 family of functions for Linux/S390.
3579 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3580 of functions for Linux/x86.
3582 * Port to Linux/CRIS contributed by Axis Communications.
3586 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3587 we can now check for namespace violations and missing declarations. The
3588 result is that almost all headers are now Unix-compliant (as defined in
3589 the upcoming XPG6). The negative side is that some programs might need
3590 corrections, too, if they depend on the incorrect form of the headers in
3591 previous versions which defined too many symbols and included too many
3594 * regex now handles multibyte character sets correctly.
3595 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3597 * iconv (the program) does now conform to the upcoming XPG6 and handles
3598 charmaps. Instead of the charset names the path of charmaps can be
3599 provided and the conversion happens based on this data.
3600 Contributed by Ulrich Drepper.
3602 * The locale program now provides detailed information about the installed
3607 only lists the names of the supported locales
3611 provides details such as country, language, and codeset name.
3612 Contributed by Ulrich Drepper.
3616 * The gencat program now parses the input file according to the charset
3617 selected by the LC_CTYPE category. This is important for stateful
3618 character sets. To make generating catalogs easier there is a way
3619 to overwrite the charset selected by the locale: before the first
3620 message or $ quote line the catalog can contain a line like
3622 $ codeset=ISO-8859-2
3624 to select the charset (ISO-8859-2 in this case).
3626 Implemented by Shinya Hanataka and Ulrich Drepper.
3628 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3629 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3630 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3632 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3633 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3635 * The LANGUAGE environment variable is now ignored unless the locale is
3636 changed from the default "C" locale.
3638 * The usual bug fixes.
3642 * Greg McGary added runtime support for bounds checking using gcc's
3643 new -fbounded-pointers option. ix86 target is complete. PowerPC
3646 * Thorsten Kukuk added secure mode support to nscd.
3648 * The Berkeley db libraries have been removed.
3650 Related, the nss_db module is now in a separate package since it
3651 obviously requires a database library being available.
3653 * Wide character I/O streams implemented by Ulrich Drepper.
3655 * Functions from the extended socket API added by Ulrich Drepper.
3657 * Functions feenableexcept and fedisableexcept to control the
3658 behaviour of individual exceptions have been added by Andreas Jaeger.
3660 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3662 * The resolver code has been updated from bind 8.2.3-T5B which supports
3663 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3666 This change could in some situations effect backward compatibility. Since
3667 now `_res' is a thread-local instead of a global variable, modifying it
3668 in one thread does not have any effect in other threads.
3670 The resolver library was also extended to allow IPv6 as the transport
3671 protocol for the requests. This work was done by Stig Venaas.
3673 * Compatibility code for K&R C compilers has been removed from the
3674 header files. A ISO C compiler is needed to use the library
3675 (conforming to either C89 or C99 standard).
3677 * Complete rewrite of the localedef program to support multibyte character
3678 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3679 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3680 Implemented by Ulrich Drepper.
3682 Bruno Haible significantly improved the generation and use of the data
3683 structures for the wide character tables.
3685 * Plural handling in gettext implemented by Ulrich Drepper.
3687 * The utmp daemon has been removed.
3689 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3691 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3694 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3696 * POSIX barriers implemented by Kaz Kylheku.
3698 * POSIX spawn function family implemented by Ulrich Drepper.
3700 * POSIX spinlocks are now available.
3702 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3704 * the configure option --enable-kernel=X.Y.Z allows to strip out
3705 compatibility for kernel versions before X.Y.Z. This is currently only
3706 implemented for Linux.
3708 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3709 field sin6_scope_id. This means that all programs using IPv6 should be
3710 recompiled. Don't expect binary compatibility with previous glibc
3713 * various conversion modules for IBM character sets contributed by
3716 * IA-64 port by Jes Sorensen and HJ Lu.
3730 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3732 * New xdr functions are added; some rpc functions are now 64bit clean.
3734 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3736 * Fixed known incompatibilities with glibc 2.0.
3738 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3740 * Optimized a number of functions (especially the ELF dynamic loader).
3742 * Update timezone data files.
3744 * lots of charmaps corrections
3746 * some new locale definitions and charmaps
3751 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3752 conform to POSIX member type requirements. Symbol versions have been
3753 adjusted as needed within the library, and for direct use by applications,
3754 but there is potential for problems if third-party libraries use
3755 struct timeval as part of their interface. This does not present
3756 a problem for X and other "essential" system libraries.
3758 * An additional locale model to support C++ Standard Library locale
3759 model and probably more was implemented by Ulrich Drepper.
3761 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3764 * Miles Bader provided the `argp' function family to support hierachical
3765 command line argument parsing, layered on top of getopt.
3767 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3769 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3772 * scanf recognizes the %a and %A format for scanning floating point numbers.
3774 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3775 information and interfaces for the available integer types.
3777 * about 130 new math functions were added to implement the ISO C9x math
3780 * the new header <complex.h> contains definitions of the complex math
3781 functions from ISO C 9X.
3783 * the new header <tgmath.h> defines generic macros to use complex or
3784 real valued functions.
3786 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3788 * Andreas Jaeger provided a test suite for the math library.
3790 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3792 * Ulrich Drepper added character set conversion functions (iconv).
3794 * Optimized string functions have been added.
3796 * The localedata addon is now part of glibc.
3798 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3800 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3801 daemon for NSS (nscd).
3803 Missing a better place here are some numbers on improvements. Under
3804 Linux 2.1.125 un-tar-ing the kernel sources takes
3808 using local files 12.19s 6.88s 22.91s
3810 using NIS 13.92s 8.91s 26.34s
3812 using NIS & nscd 10.37s 7.34s 25.30s
3814 using NIS+ 27.57s 30.37s 640.46s
3816 using NIS+ & nscd 10.25s 7.83s 26.51s
3818 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3820 Keep in mind that non-namelookup related operations dominate above times.
3821 It was just a common complain that using NIS+ unpacking the kernel is
3824 [1] The old nscd implementation is not available anymore since it was
3825 distributed with glibc up to version 2.0.98 and thus is now replaced.
3827 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3829 * Mark Kettenis provided a Hesiod NSS module.
3831 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3832 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3834 * rcmd can now handle netgroups (Dick Streefland).
3836 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3839 * Support for the IPv6 protocol has been added to the socket API, as per the
3840 latest draft standards.
3842 * Support for Linux 2.2 has been added.
3844 * Interface changes relative to the latest 2.0.x release:
3845 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3846 addseverity NEW: Unix98
3847 alphasort64 NEW: LFS
3848 argp_err_exit_status NEW: argp, GNU ext
3849 argp_error NEW: argp, GNU ext
3850 argp_failure NEW: argp, GNU ext
3851 argp_help NEW: argp, GNU ext
3852 argp_parse NEW: argp, GNU ext
3853 argp_program_bug_address NEW: argp, GNU ext
3854 argp_program_version NEW: argp, GNU ext
3855 argp_program_version_hook NEW: argp, GNU ext
3856 argp_state_help NEW: argp, GNU ext
3857 argp_usage NEW: argp, GNU ext
3858 authdes_create NEW: Secure RPC
3859 authdes_getucred NEW: Secure RPC
3860 authdes_pk_create NEW: Secure RPC
3861 backtrace NEW: GNU ext.
3862 backtrace_symbols NEW: GNU ext.
3863 backtrace_symbols_fd NEW: GNU ext.
3865 cacosf NEW: ISO C 9x
3866 cacosh NEW: ISO C 9x
3867 cacoshf NEW: ISO C 9x
3868 cacoshl NEW: ISO C 9x
3869 cacosl NEW: ISO C 9x
3876 casinf NEW: ISO C 9x
3877 casinh NEW: ISO C 9x
3878 casinhf NEW: ISO C 9x
3879 casinhl NEW: ISO C 9x
3880 casinl NEW: ISO C 9x
3882 catanf NEW: ISO C 9x
3883 catanh NEW: ISO C 9x
3884 catanhf NEW: ISO C 9x
3885 catanhl NEW: ISO C 9x
3886 catanl NEW: ISO C 9x
3887 cbc_crypt NEW: Secure RPC
3891 ccoshf NEW: ISO C 9x
3892 ccoshl NEW: ISO C 9x
3898 cimagf NEW: ISO C 9x
3899 cimagl NEW: ISO C 9x
3900 clearerr_locked REMOVED
3901 clntunix_create NEW: sunrpc ext
3903 clog10 NEW: ISO C 9x
3904 clog10f NEW: ISO C 9x
3905 clog10l NEW: ISO C 9x
3915 cprojf NEW: ISO C 9x
3916 cprojl NEW: ISO C 9x
3918 crealf NEW: ISO C 9x
3919 creall NEW: ISO C 9x
3924 csinhf NEW: ISO C 9x
3925 csinhl NEW: ISO C 9x
3928 csqrtf NEW: ISO C 9x
3929 csqrtl NEW: ISO C 9x
3933 ctanhf NEW: ISO C 9x
3934 ctanhl NEW: ISO C 9x
3936 des_setparity NEW: Secure RPC
3937 ecb_crypt NEW: Secure RPC
3938 endutxent NEW: Unix98
3940 exp10f NEW: ISO C 9x
3941 exp10l NEW: ISO C 9x
3945 fattach NEW: STREAMS
3946 fdetach NEW: STREAMS
3950 feclearexcept NEW: ISO C 9x
3951 fegetenv NEW: ISO C 9x
3952 fegetexceptflag NEW: ISO C 9x
3953 fegetround NEW: ISO C 9x
3954 feholdexcept NEW: ISO C 9x
3956 feraiseexcept NEW: ISO C 9x
3957 ferror_locked REMOVED
3958 fesetenv NEW: ISO C 9x
3959 fesetexceptflag NEW: ISO C 9x
3960 fesetround NEW: ISO C 9x
3961 fetestexcept NEW: ISO C 9x
3962 feupdateenv NEW: ISO C 9x
3963 fflush_locked REMOVED
3967 fgets_unlocked NEW: GNU ext.
3968 fileno_locked REMOVED
3980 fputc_locked REMOVED
3981 fputs_unlocked NEW: GNU ext.
3982 fread_unlocked NEW: GNU ext.
3987 fstatvfs NEW: Unix98
3991 ftruncate64 NEW: LFS
3993 fwrite_unlocked NEW: GNU ext.
3994 gai_strerror NEW: IPv6
3998 getchar_locked REMOVED
4000 getdate_err NEW: Unix98
4001 getdate_r NEW: GNU ext.
4003 getnameinfo NEW: IPv6
4004 getnetname NEW: Secure RPC
4005 getpmsg NEW: STREAMS
4006 getpt NEW: Unix98 PTY
4007 getrlimit64 NEW: LFS
4008 getutxent NEW: Unix98
4009 getutxid NEW: Unix98
4010 getutxline NEW: Unix98
4011 glob64 NEW: GNU ext.
4012 globfree64 NEW: GNU ext.
4013 gnu_get_libc_release NEW: GNU ext.
4014 gnu_get_libc_version NEW: GNU ext.
4015 grantpt NEW: Unix98 PTY
4016 host2netname NEW: Secure RPC
4018 iconv_close NEW: iconv
4019 iconv_open NEW: iconv
4020 if_freenameindex NEW: IPv6
4021 if_indextoname NEW: IPv6
4022 if_nameindex NEW: IPv6
4023 if_nametoindex NEW: IPv6
4024 in6addr_any NEW: IPv6
4025 in6addr_loopback NEW: IPv6
4026 inet6_isipv4mapped NEW: IPv6
4027 isastream NEW: STREAMS
4028 iswblank NEW: GNU ext.
4029 key_decryptsession NEW: Secure RPC
4030 key_decryptsession_pk NEW: Secure RPC
4031 key_encryptsession NEW: Secure RPC
4032 key_encryptsession_pk NEW: Secure RPC
4033 key_gendes NEW: Secure RPC
4034 key_get_conv NEW: Secure RPC
4035 key_secretkey_is_set NEW: Secure RPC
4036 key_setnet NEW: Secure RPC
4037 key_setsecret NEW: Secure RPC
4038 llrint NEW: ISO C 9x
4039 llrintf NEW: ISO C 9x
4040 llrintl NEW: ISO C 9x
4041 llround NEW: ISO C 9x
4042 llroundf NEW: ISO C 9x
4043 llroundl NEW: ISO C 9x
4048 lrintf NEW: ISO C 9x
4049 lrintl NEW: ISO C 9x
4050 lround NEW: ISO C 9x
4051 lroundf NEW: ISO C 9x
4052 lroundl NEW: ISO C 9x
4054 makecontext NEW: Unix98
4055 mempcpy NEW: GNU ext.
4058 modify_ldt NEW: kernel
4062 nearbyint NEW: ISO C 9x
4063 nearbyintf NEW: ISO C 9x
4064 nearbyintl NEW: ISO C 9x
4065 netname2host NEW: Secure RPC
4066 netname2user NEW: Secure RPC
4067 nexttoward NEW: ISO C 9x
4068 nexttowardf NEW: ISO C 9x
4069 nexttowardl NEW: ISO C 9x
4073 passwd2des NEW: Secure RPC
4075 pow10f NEW: GNU ext.
4076 pow10l NEW: GNU ext.
4079 printf_size NEW: GNU ext.
4080 printf_size_info NEW: GNU ext.
4081 profil_counter REMOVED
4082 pthread_mutexattr_getkind_np REPLACED
4083 pthread_mutexattr_setkind_np REPLACED
4084 ptsname NEW: Unix98 PTY
4085 ptsname_r NEW: Unix98 PTY
4087 putchar_locked REMOVED
4088 putgrent NEW: GNU ext.
4090 putpmsg NEW: STREAMS
4091 pututxline NEW: Unix98
4095 readdir64_r NEW: LFS
4096 remquo NEW: ISO C 9x
4097 remquof NEW: ISO C 9x
4098 remquol NEW: ISO C 9x
4100 roundf NEW: ISO C 9x
4101 roundl NEW: ISO C 9x
4103 scalbln NEW: ISO C 9x
4104 scalblnf NEW: ISO C 9x
4105 scalblnl NEW: ISO C 9x
4107 sendfile NEW: kernel
4108 setcontext NEW: Unix98
4109 setrlimit64 NEW: LFS
4110 setutxent NEW: Unix98
4112 sigignore NEW: Unix98
4113 sigqueue NEW: POSIX.1b
4114 sigrelse NEW: Unix98
4115 sigset NEW: POSIX.1b
4116 sigtimedwait NEW: POSIX.1b
4117 sigwaitinfo NEW: POSIX.1b
4118 sincos NEW: GNU ext.
4119 sincosf NEW: GNU ext.
4120 sincosl NEW: GNU ext.
4124 strcasestr NEW: GNU ext.
4125 strtoimax NEW: ISO C 9x
4126 strtoumax NEW: ISO C 9x
4127 strverscmp NEW: GNU ext.
4128 svcauthdes_stats NEW: Secure RPC
4129 svcunix_create NEW: sunrpc etx
4130 svcunixfd_create NEW: sunrpc ext
4131 swapcontext NEW: Unix98
4132 tcgetsid NEW: Unix98 PTY
4133 tdestroy NEW: GNU ext.
4134 tgamma NEW: ISO C 9x
4135 tgammaf NEW: ISO C 9x
4136 tgammal NEW: ISO C 9x
4140 truncf NEW: ISO C 9x
4141 truncl NEW: ISO C 9x
4143 unlockpt NEW: Unix98 PTY
4144 updwtmpx NEW: Unix98
4145 user2netname NEW: Secure RPC
4146 utmpxname NEW: Unix98
4147 versionsort NEW: GNU ext.
4148 versionsort64 NEW: GNU ext.
4150 wcscasecmp NEW: GNU ext.
4151 wcsncasecmp NEW: GNU ext.
4152 wcsnlen NEW: GNU ext.
4153 wcstoimax NEW: ISO C 9x
4154 wcstoll NEW: ISO C 9x
4155 wcstoull NEW: ISO C 9x
4156 wcstoumax NEW: ISO C 9x
4158 wordexp NEW: POSIX.2
4159 wordfree NEW: POSIX.2
4160 write_profiling REMOVED
4161 xdecrypt NEW: Secure RPC
4162 xdr_authdes_cred NEW: Secure RPC
4163 xdr_authdes_verf NEW: Secure RPC
4164 xdr_cryptkeyarg NEW: Secure RPC
4165 xdr_cryptkeyarg2 NEW: Secure RPC
4166 xdr_cryptkeyres NEW: Secure RPC
4167 xdr_getcredres NEW: Secure RPC
4168 xdr_key_netstarg NEW: Secure RPC
4169 xdr_key_netstres NEW: Secure RPC
4170 xdr_keybuf NEW: Secure RPC
4171 xdr_keystatus NEW: Secure RPC
4172 xdr_netnamestr NEW: Secure RPC
4173 xdr_sizeof NEW: Secure RPC
4174 xdr_unixcred NEW: sunrpc ext
4175 xencrypt NEW: Secure RPC
4176 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4187 * inet_ntoa is thread-safe
4189 * updwtmp is moved from libutil to libc
4191 * rewrite of cbrt function
4193 * update of timezone data
4207 * add atoll function
4209 * fix complex problems in Berkeley DB code
4211 * fix math functions
4215 * fixed lots of header problems (especially Linux/GNU specific)
4217 * dynamic loader preserves all registers
4219 * Roland McGrath provided support for handling of auxiliary objects in
4220 the ELF dynamic loader.
4222 * support for parallel builds is improved
4226 * GNU extensions are no longer declared by default. To enable them you
4227 must define the macro `_GNU_SOURCE' in your program or compile with
4230 * The library has changed from using GNU ld symbol aliases to using weak
4231 symbols where available. The ELF object file format supports weak
4232 symbols; GNU ld also supports weak symbols in the a.out format. (There
4233 is also now support for other GNU ld extensions in ELF. Use the
4234 `--with-elf' option to configure to indicate you have ELF, and
4235 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4236 of many files which contained only symbol aliases, reducing the size of
4237 the source and the compiled library; many other files were renamed to
4238 less cryptic names previously occupied by the symbol alias files.
4239 There is a new header file <elf.h> for programs which operate on
4240 files in the ELF format.
4242 * Converted to Autoconf version 2, so `configure' has more options.
4243 Run `configure --help' to see the details.
4245 * The library can now be configured to build profiling, highly-optimized
4246 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4247 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4248 `configure' enable building these extra libraries. The shared library is
4249 built by default when using both ELF and GNU ld. When shared libraries
4250 are enabled, the new library `-ldl' is available for arbitrary run-time
4251 loading of shared objects; its interface is defined in <dlfcn.h>. The
4252 new header file <link.h> gives access to the internals of the run-time
4253 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4254 application of same name on other systems and it provides information
4255 about dynamically linked binaries.
4257 * The C library now provides the run-time support code for profiling
4258 executables compiled with `-pg'. Programs can control the profiling code
4259 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4260 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4261 the support for this file format was contributed by David Mosberger-Tang.
4263 * The math code has been replaced with a math library based on fdlibm from
4264 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4265 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4266 functions. The math functions now reside in a separate library, so
4267 programs using them will need to use `-lm' their linking commands.
4269 * John C. Bowman contributed optimized ix87 assembler inline functions.
4271 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4272 mechanism similar to that found in Solaris 2. This is now used for the
4273 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4274 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4275 file controls what services are used for each individual database. This
4276 works by loading shared libraries with names specified in `nsswitch.conf',
4277 so service modules can be changed or added at any time without even
4278 relinking any program. Currently there are the file, db, and NIS based
4279 NSS services available.
4281 * The new functions `strtoq' and `strtouq' parse integer values from
4282 strings, like `strtol' and `strtoul', but they return `long long int' and
4283 `unsigned long long int' values, respectively (64-bit quantities).
4285 * The new functions `strtof' and `strtold' parse floating-point values from
4286 strings, like `strtod', but they return `float' and `long double' values,
4287 respectively (on some machines `double' and `long double' are the same).
4289 * Ulrich Drepper has contributed new implementations of the floating-point
4290 printing and reading code used in the `printf' family of functions and
4291 `strtod', `strtof', and `strtold'. These new functions are perfectly
4292 accurate, and much faster than the old ones.
4294 * The implementation of the POSIX locale model was completely rewritten by
4295 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4296 to compile the POSIX locale definition.
4298 * The former dummy implementations of the strcoll and strxfrm function are
4299 now replaced by fully functional code contributed by Ulrich Drepper. The
4300 collation information comes from the POSIX locale definitions.
4302 * The new header <langinfo.h> defines an interface for accessing
4303 various locale-dependent data (using the locale chosen with `setlocale').
4305 * Ulrich Drepper has contributed a new suite of functions for operation on
4306 wide-character and multibyte-character strings, in <wchar.h>;
4307 and classification and case conversion of wide characters, in <wctype.h>.
4308 These new functions are conforming to the ISO C, Amendement 1 specification.
4310 * There is now a second implementation of the standard I/O library available.
4311 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4312 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4314 * You can now use positional parameter specifications in format strings
4315 for the `printf' and `scanf' families of functions. For example,
4316 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4317 ``Number 6, Mr Jones''. This is mainly useful when providing different
4318 format strings for different languages, whose grammars may dictate
4319 different orderings of the values being printed. To support this
4320 feature, the interface for `register_printf_handler' has changed; see
4321 the header file <printf.h> for details.
4323 * The `printf' and `scanf' families of functions now understand a new
4324 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4325 to group numbers as indicated by the locale; for `scanf' and friends, this
4326 says to accept as valid only a number with all the proper grouping
4327 separators in the right places. In the default "C" locale, numbers are
4328 not grouped; but locales for specific countries will define the usual
4329 conventions (i.e. separate thousands with `,' in the US locale).
4331 * The pgrp functions have been regularized, slightly incompatibly but much
4332 less confusingly. The core functions are now `getpgid' and `setpgid',
4333 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4334 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4335 provided for compatibility. There is no longer an incompatible `getpgrp'
4336 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4338 * The new header file <fts.h> and suite of functions simplify programs that
4339 operate on directory trees. This code comes from 4.4 BSD.
4341 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4342 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4343 NSS scheme used in glibc.
4345 * There is a new malloc debugging hook `__memalign_hook'.
4347 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4348 `unsigned int' in <sys/types.h>. These are for compatibility only and
4349 their use is discouraged.
4351 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4352 done differently, so that it works even without GNU ld.
4354 * New function `euidaccess' checks allowed access to a file like `access',
4355 but using the effective IDs instead of the real IDs.
4357 * The time zone data files have been updated for the latest and greatest
4358 local time conventions of the countries of the world.
4360 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4363 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4364 for formatting floating-point numbers. They are provided only for
4365 compatibility; new programs should use `sprintf' instead. There are
4366 also equivalent function for the `long double' floating-point type and
4367 all functions also exist in a reentrant form.
4369 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4370 functions for maintaining the login-record files (primarily of use to
4371 system programs such as `login'), and convenient functions for
4372 allocating and initializing a pseudo-terminal (pty) device.
4374 * Ulrich Drepper has contributed new support for System V style
4375 shared memory and IPC on systems that support it.
4377 * Ulrich Drepper has contributed several miscellaneous new functions found
4378 in System V: The `hsearch' family of functions provide an effective
4379 implementation of hash tables; `a64l' and `l64a' provide a very simple
4380 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4383 * Ulrich Drepper has contributed new reentrant counterparts for the
4384 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4386 * Ulrich Drepper has contributed new, highly-optimized versions of several
4387 string functions for the i486/Pentium family of processors.
4389 * Ulrich Drepper has updated the Linux-specific code, based largely
4390 on work done in Hongjiu Lu's version of GNU libc for Linux.
4391 The GNU library now supports Linux versions 2.0.10 and later,
4392 using the ELF object file format (i[3456]86-*-linux).
4394 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4396 * David Mosberger-Tang and Richard Henderson have ported the C library
4397 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4398 dynamic linking support for ELF/Alpha.
4400 * Richard Henderson contributed several Alpha optimized assembler function
4401 for arithmetic and string handling.
4403 * Ulrich Drepper has contributed a new set of message catalog functions to
4404 support multiple languages using the <libintl.h> interface, for use with
4405 his new package GNU gettext. Translation volunteers have contributed
4406 catalogs of the library's messages in Spanish, German, and Korean.
4408 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4409 program and the `catgets' function for reading the catalog files it
4410 creates. (The <libintl.h> interface is preferred; we include the
4411 <nl_types.h> interface using `catgets' only for source compatibility with
4412 programs already written to use it.)
4414 * New header file <values.h> gives SVID-compatible names for <limits.h>
4417 * Various new macros, declarations, and small header files for compatibility
4420 * New function `group_member' is a convenient way to check if a process has
4421 a given effective group ID.
4423 * When using GCC 2.7 and later, the socket functions are now declared in a
4424 special way so that passing an argument of type `struct sockaddr_in *',
4425 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4426 `struct sockaddr *' type, does not generate a type-clash warning.
4428 * New function `error' declared in header file <error.h> is a convenient
4429 function for printing error messages and optionally exiting; this is the
4430 canonical function used in GNU programs. The new functions `err', `warn',
4431 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4432 doing the same thing.
4434 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4435 POSIX.2 `glob' function to do ~ and {...} expansion.
4437 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4438 `clearenv' which is used in POSIX.9 is also available.
4440 * New function `getsid' returns session ID number on systems that support it.
4442 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4443 files <db.h> and <mpool.h> provide a rich set of functions for several
4444 types of simple databases stored in memory and in files, and <ndbm.h> is
4445 an old `ndbm'-compatible interface using the `db' functions. Link with
4446 `-ldb' to get these functions.
4448 * New macro `strdupa' copies a string like `strdup', but uses local stack
4449 space from `alloca' instead of dynamic heap space from `malloc'.
4451 * New function `strnlen' is like `strlen' but searches only a given maximum
4452 number of characters for the null terminator. `stpncpy', `strndup' and
4453 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4456 * New function `statfs' in header <sys/statfs.h>.
4458 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4459 provide convenient functions for operating on blocks of null-terminated
4462 * A new suite of functions in <utmp.h> handle all the details of reading
4463 and writing the utmp file.
4465 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4468 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4469 completely POSIX compliant and also implemented the extended functionality
4470 to handle alternate digit representation and alternate era date formats.
4472 * Ulrich Drepper provided an implementation of the `strptime' function
4473 defined in XPG4.2 which transforms a string into a `struct tm' value.
4475 * Paul Eggert provided the tzselect shell script as part of the timezone
4476 code. The shell script makes it easy to select the correct timezone
4479 * The implementation of the malloc family of functions is completely replaced
4480 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4481 The implementation uses the mmap function (if available) and it is
4482 optimized for the use in multi threaded programs.
4484 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4485 functions. This new functionality is usable by specifying a special
4486 salt string and it is compatible with implementation on *BSD systems.
4488 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4489 `getsubopt' to handle second level command line options, `bsd_signal'
4490 to access BSD style `signal' functionality, the obsolete `regexp' style
4493 * the `lchown' function is available on system which support this
4496 * The implementation of the shadow password handling function was contributed
4499 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4501 * POSIX.1g support was added. The <sys/select.h> header is available,
4502 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4503 implementation of `getaddrinfo'.
4507 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4509 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4510 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4512 * The new function `daemon' from 4.4 BSD is useful for server programs that
4513 want to put themselves in the background.
4515 * Joel Sherrill has contributed support for several standalone boards that
4516 run without an operating system.
4518 * `printf', `scanf' and friends now accept a `q' type modifier for long
4519 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4521 * All of the code taken from BSD (notably most of the math and networking
4522 routines) has been updated from the BSD 4.4-Lite release.
4524 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4526 * The new functions `getdomainname' and `setdomainname' fetch or change the
4527 YP/NIS domain name. These are system calls which exist on systems which
4530 * The time zone data files have been updated for the latest international
4533 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4534 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4538 * The C library now includes support for Sun RPC, from Sun's free
4539 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4540 are included. (There is still no support for YP.)
4542 * Tom Quinn has contributed a port of the C library to SGI machines running
4543 Irix 4 (mips-sgi-irix4).
4545 * The new `lockf' function is a simplified interface to the locking
4546 facilities of `fcntl', included for compatibility.
4548 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4550 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4552 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4555 * The error message printed by `assert' for a failed assertion now includes
4556 the name of the program (if using GNU ld) and the name of the calling
4557 function (with versions of GCC that support this).
4559 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4561 * The library now includes the <sys/mman.h> header file and memory
4562 management functions `mmap', `munmap', `mprotect', `msync', and
4563 `madvise', on systems that support those facilities.
4565 * The interface for `mcheck' has changed slightly: the function called to
4566 abort the program when an allocation inconsistency is detected now takes
4567 an argument that indicates the type of failure. The new function
4568 `mprobe' lets you request a consistency check for a particular block at
4569 any time (checks are normally done only when you call `free' or `realloc'
4572 * It is now possible to easily cross-compile the C library, building on one
4573 system a library to run on another machine and/or operating system. All
4574 you need to do is set the variable `HOST_CC' in `configparms' to the
4575 native compiler for programs to run on the machine you are building on (a
4576 few generator programs are used on Unix systems); set `CC' to the
4579 * The new function `fexecve' (only implemented on the GNU system) executes
4580 a program file given a file descriptor already open on the file.
4584 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4585 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4587 * You can set the variable `libprefix' in `configparms' to specify a prefix
4588 to be prepended to installed library files; this makes it easy to install
4589 the GNU C library to be linked as `-lgnuc' or whatever.
4591 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4592 copies a limited number of characters from a string, and returns the
4593 address of the last character written.
4595 * You no longer need to check for whether the installed `stddef.h' is
4596 compatible with the GNU C library. configure now checks for you.
4598 * You can now define a per-stream `fileno' function to convert the
4599 stream's cookie into an integral file descriptor.
4601 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4602 allocates zero bytes of storage, and returns a unique pointer which
4603 you can pass to `realloc' or `free'. The behavior is undefined if
4604 you dereference this pointer.
4606 * The C library now runs on Sony NEWS m68k machines running either
4607 NewsOS 3 or NewsOS 4.
4609 * The new `syscall' function is a system-dependent primitive function
4610 for invoking system calls. It has the canonical behavior on Unix
4611 systems, including unreliable return values for some calls (such as
4612 `pipe', `fork' and `getppid').
4614 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4615 to `EAGAIN', which is the preferred name. On systems whose kernels
4616 use two distinct codes, the C library now translates EWOULDBLOCK to
4617 EAGAIN in every system call function.
4621 * The GNU C Library Reference Manual is now distributed with the library.
4622 `make dvi' will produce a DVI file of the printed manual.
4623 `make info' will produce Info files that you can read on line using C-h i
4624 in Emacs or the `info' program.
4625 Please send comments on the manual to bug-glibc-manual@gnu.org.
4627 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4629 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4631 * Jason Merrill has contributed a port to the Sequent Symmetry running
4632 Dynix version 3 (i386-sequent-dynix).
4634 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4635 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4637 * New function `memory_warnings' lets you arrange to get warnings when
4638 malloc is running out of memory to allocate, like Emacs gives you.
4640 * The C library now contains the relocating allocator used in Emacs 19 for
4641 its editing buffers. This allocator (ralloc) minimizes allocation
4642 overhead and fragmentation by moving allocated regions around whenever it
4643 needs to. You always refer to a ralloc'd region with a "handle" (a
4644 pointer to a pointer--an object of type `void **').
4646 * There is a new `printf' format: `%m' gives you the string corresponding
4647 to the error code in `errno'.
4649 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4650 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4651 pass, the `a' modifier says to fill in a `char **' you pass with a
4654 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4655 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4656 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4658 * `mkstemp' is a traditional Unix function to atomically create and open a
4659 uniquely-named temporary file.
4663 * The standard location for the file that says what the local timezone is
4664 has changed again. It is now `/usr/local/etc/localtime' (or more
4665 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4667 * The distribution no longer contains any files with names longer than 14
4670 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4671 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4673 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4677 * The configuration process has changed quite a bit. The `configure'
4678 script is now used just like the configuration scripts for other GNU
4679 packages. The `sysdeps' directory hierarchy is much rearranged.
4680 The file `INSTALL' explains the new scheme in detail.
4682 * The header files no longer need to be processed into ANSI C and
4683 traditional C versions. There is just one set of files to install, and
4684 it will work with ANSI or old C compilers (including `gcc -traditional').
4686 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4687 MIPS DECStation running Ultrix 4.
4689 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4690 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4691 made itself into a shared library.
4693 * Yet further improved support for the i386, running 4.3 BSD-like systems
4694 (such as Mach 3 with the Unix single-server), or System V.
4696 * New function `strncasecmp' to do case-insensitive string comparison
4697 with limited length.
4699 * New function `strsep' is a reentrant alternative to `strtok'.
4701 * New functions `scandir' and `alphasort' for searching directories.
4703 * New function `setenv' is a better interface to `putenv'.
4705 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4706 function for traversing a directory tree.
4708 * The GNU obstack package is now also part of the C library.
4709 The new function `open_obstack_stream' creates a stdio stream that
4710 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4711 formatted output directly to an obstack.
4713 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4714 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4716 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4718 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4719 things to your strings.
4721 * There are some new test programs: test-fseek, testmb, and testrand.
4723 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4724 These ports are not finished, but are a good starting place for really
4725 supporting those systems.
4727 * `/etc/localtime' is now the standard location for the file that says what
4728 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4729 This follows the general principle that `/etc' is the place for all local
4730 configuration files.
4732 * The C library header files now use `extern "C"' when used by the C++
4733 compiler, so the C library should now work with C++ code.
4735 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4736 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4739 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4740 function which uses the `merge sort' algorithm, and is said to be
4741 significantly faster than the old GNU `qsort' function. Merge sort is
4742 now the standard `qsort' function. The new algorithm can require a lot
4743 of temporary storage; so, the old sorting function is called when the
4744 required storage is not available.
4746 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4747 provides the Unix `crypt' function, plus some other entry points.
4749 * The code and header files taken from 4.4 BSD have been updated with the
4750 latest files released from Berkeley.
4752 ----------------------------------------------------------------------
4753 Copyright information:
4755 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4757 Permission is granted to anyone to make or distribute verbatim copies
4758 of this document as received, in any medium, provided that the
4759 copyright notice and this permission notice are preserved,
4760 thus giving the recipient permission to redistribute in turn.
4762 Permission is granted to distribute modified versions
4763 of this document, or of portions of it,
4764 under the above conditions, provided also that they
4765 carry prominent notices stating who last changed them.
4768 version-control: never