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 * <math.h> functions that round their results to a narrower type are added
13 from TS 18661-1:2014 and TS 18661-3:2015:
15 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
18 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
21 * Nominative and genitive month names are now supported for the Catalan and
22 Czech languages. The Catalan and Greek languages now support abbreviated
23 alternative month names.
25 Deprecated and removed features, and other changes affecting compatibility:
27 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
28 installed. Software that was using either header should be updated to
29 use standard <stdio.h> interfaces instead.
31 * The stdio.h functions 'getc' and 'putc' are no longer defined as macros.
32 This was never required by the C standard, and the macros just expanded
33 to call alternative names for the same functions. If you hoped getc and
34 putc would provide performance improvements over fgetc and fputc, instead
35 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
36 necessary, flockfile and funlockfile.
38 * All stdio functions now treat end-of-file as a sticky condition. If you
39 read from a file until EOF, and then the file is enlarged by another
40 process, you must call clearerr or another function with the same effect
41 (e.g. fseek, rewind) before you can read the additional data. This
42 corrects a longstanding C99 conformance bug. It is most likely to affect
43 programs that use stdio to read interactive input from a terminal.
46 * The macros 'major', 'minor', and 'makedev' are now only available from
47 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
48 headers that happen to include <sys/types.h>. These macros are rarely
49 used, not part of POSIX nor XSI, and their names frequently collide with
50 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
53 <sys/sysmacros.h> is a GNU extension. Portable programs that require
54 these macros should first include <sys/types.h>, and then include
55 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
57 Changes to build and runtime requirements:
59 [Add changes to build and runtime requirements here]
61 Security related changes:
63 [Add security related changes here]
65 The following bugs are resolved with this release:
67 [The release manager will add the list generated by
68 scripts/list-fixed-bugs.py just before the release.]
75 * The GNU C Library can now be compiled with support for building static
76 PIE executables (See --enable-static-pie in INSTALL). These static PIE
77 executables are like static executables but can be loaded at any address
78 and provide additional security hardening benefits at the cost of some
79 memory and performance. When the library is built with --enable-static-pie
80 the resulting libc.a is usable with GCC 8 and above to create static PIE
81 executables using the GCC option '-static-pie'. This feature is currently
82 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
83 aarch64 with binutils 2.30 or later.
85 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
86 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
89 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
91 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
93 * In order to support faster and safer process termination the malloc API
94 family of functions will no longer print a failure address and stack
95 backtrace after detecting heap corruption. The goal is to minimize the
96 amount of work done after corruption is detected and to avoid potential
97 security issues in continued process execution. Reducing shutdown time
98 leads to lower overall process restart latency, so there is benefit both
99 from a security and performance perspective.
101 * The abort function terminates the process immediately, without flushing
102 stdio streams. Previous glibc versions used to flush streams, resulting
103 in deadlocks and further data corruption. This change also affects
104 process aborts as the result of assertion failures.
106 * On platforms where long double has the IEEE binary128 format (aarch64,
107 alpha, mips64, riscv, s390 and sparc), the math library now implements
108 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
109 These are the same interfaces added in version 2.26 for some platforms where
110 this format is supported but is not the format of long double.
112 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
113 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
114 implements interfaces for that type, as defined by ISO/IEC TS
115 18661-3:2015. These are corresponding interfaces to those supported for
118 * The math library now implements interfaces for the _Float32, _Float64 and
119 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
120 corresponding interfaces to those supported for _Float128.
122 * glibc now implements the memfd_create and mlock2 functions on Linux.
124 * Support for memory protection keys was added. The <sys/mman.h> header now
125 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
128 * The copy_file_range function was added.
130 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
132 * The ldconfig utility now processes `include' directives using the C/POSIX
133 collation ordering. Previous glibc versions used locale-specific
134 ordering, the change might break systems that relied on that.
136 * Support for two grammatical forms of month names has been added.
137 In a call to strftime, the "%B" and "%b" format specifiers will now
138 produce the grammatical form required when the month is used as part
139 of a complete date. New "%OB" and "%Ob" specifiers produce the form
140 required when the month is named by itself. For instance, in Greek
141 and in many Slavic and Baltic languages, "%B" will produce the month
142 in genitive case, and "%OB" will produce the month in nominative case.
144 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
145 are all valid and will all accept any known form of month
146 name---standalone or complete, abbreviated or full. In a call to
147 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
148 the strings used by "%B" and "%b", respectively. New query
149 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
150 used by "%OB" and "%Ob", respectively.
152 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
153 define the strings for %OB and %Ob, respectively; these have the
154 same syntax as "mon" and "abmon". These arrays are optional; if they
155 are not provided then they have the same content as "mon" and "abmon",
158 These features are provided for locales which define "alt_mon" and/or
159 "ab_alt_mon" in their locale source data. This release includes such
160 alternative month name data for the following languages: Belarusian,
161 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
163 This feature is currently a GNU extension, but it is expected to
164 be added to the next revision of POSIX, and it is also already
165 available on some BSD-derived operating systems.
167 This feature will cause existing statically compiled applications
168 to fail to load locales and fall back to the builtin C/POSIX locales.
169 See notes below for other changes affecting compatibility.
171 * Support for the RISC-V ISA running on Linux has been added. This port
172 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
173 for the following ISA and ABI pairs:
179 Deprecated and removed features, and other changes affecting compatibility:
181 * Statically compiled applications attempting to load locales compiled for the
182 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
183 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
184 support for two grammatical forms of the month names, also extends the locale
185 data binary format. Static applications needing locale support must be
186 recompiled to match the runtime and data they are deployed with. In some
187 distributions there is an upgrade window where dynamically linked applications
188 may use a new library but the old locale data and also fall back to the
189 builtin C/POSIX locales; restarting the application process is sufficient to
192 * Support for statically linked applications which call dlopen is deprecated
193 and will be removed in a future version of glibc. Applications which call
194 dlopen need to be linked dynamically instead.
196 * Support for old programs which use internal stdio data structures and
197 functions is deprecated. This includes programs which use the C++ streams
198 provided by libstdc++ in GCC 2.95. Programs which use the internal
199 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
200 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
201 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
202 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
203 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
204 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
205 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
206 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
207 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
208 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
209 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
210 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
211 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
212 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
213 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
214 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
215 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
216 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
217 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
218 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
219 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
220 future version of glibc. Unlike other symbol removals, these old
221 applications will not be supported using compatibility symbols.
223 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
224 defined by <sys/ptrace.h>.
226 * libm no longer supports SVID error handling (calling a user-provided
227 matherr function on error) or the _LIB_VERSION variable to control error
228 handling. (SVID error handling and the _LIB_VERSION variable still work
229 for binaries linked against older versions of the GNU C Library.) The
230 libieee.a library is no longer provided. math.h no longer defines struct
231 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
232 TLOSS, PLOSS and HUGE.
234 * The libm functions pow10, pow10f and pow10l are no longer supported for
235 new programs. Programs should use the standard names exp10, exp10f and
236 exp10l for these functions instead.
238 * The mcontext_t type is no longer the same as struct sigcontext. On
239 platforms where it was previously the same, this changes the C++ name
240 mangling for interfaces involving this type.
242 * The add-ons mechanism for building additional packages at the same time as
243 glibc has been removed. The --enable-add-ons configure option is now
246 * The --without-fp configure option is now ignored. Whether hardware
247 floating-point instructions are used is now configured based on whether
248 the compiler used at configure time (without any options implied by a
249 --with-cpu= configure option) uses such instructions.
251 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
252 the specified string can be parsed as a domain name.
254 * In the malloc_info output, the <heap> element may contain another <aspace>
255 element, "subheaps", which contains the number of sub-heaps.
257 * The libresolv function p_secstodate is no longer supported for new
260 * The tilepro-*-linux-gnu configuration is no longer supported.
262 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
263 and will be removed in a future release. Software that is still using
264 either header should be updated to use standard <stdio.h> interfaces
267 libio.h was originally the header for a set of supported GNU extensions,
268 but they have not been maintained as such in many years, they are now
269 standing in the way of improvements to stdio, and we don't think there are
270 any remaining external users. _G_config.h was never intended for public
271 use, but predates the bits convention.
273 Changes to build and runtime requirements:
275 * bison version 2.7 or later is required to generate code in the 'intl'
278 Security related changes:
280 CVE-2009-5064: The ldd script would sometimes run the program under
281 examination directly, without preventing code execution through the
282 dynamic linker. (The glibc project disputes that this is a security
283 vulnerability; only trusted binaries must be examined using the ldd
286 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
287 suffered from a one-byte overflow during ~ operator processing (either
288 on the stack or the heap, depending on the length of the user name).
289 Reported by Tim Rühsen.
291 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
292 would sometimes fail to free memory allocated during ~ operator
293 processing, leading to a memory leak and, potentially, to a denial
296 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
297 without GLOB_NOESCAPE, could write past the end of a buffer while
298 unescaping user names. Reported by Tim Rühsen.
300 CVE-2017-17426: The malloc function, when called with an object size near
301 the value SIZE_MAX, would return a pointer to a buffer which is too small,
302 instead of NULL. This was a regression introduced with the new malloc
303 thread cache in glibc 2.26. Reported by Iain Buclaw.
305 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
306 to the allocation of too much memory. (This is not a security bug per se,
307 it is mentioned here only because of the CVE assignment.) Reported by
310 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
311 of the number of search path components. (This is not a security
312 vulnerability per se because no trust boundary is crossed if the fix for
313 CVE-2017-1000366 has been applied, but it is mentioned here only because
314 of the CVE assignment.) Reported by Qualys.
316 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
317 for AT_SECURE or SUID binaries could be used to load libraries from the
320 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
321 succeeds without returning an absolute path due to unexpected behaviour
322 of the Linux kernel getcwd syscall. Reported by halfdog.
324 CVE-2018-6485: The posix_memalign and memalign functions, when called with
325 an object size near the value of SIZE_MAX, would return a pointer to a
326 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
328 CVE-2018-6551: The malloc function, when called with an object size near
329 the value of SIZE_MAX, would return a pointer to a buffer which is too
330 small, instead of NULL.
332 The following bugs are resolved with this release:
334 [866] glob: glob should match dangling symlinks
335 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
336 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
338 [5997] math: Very slow execution of sinf function
339 [10580] localedata: hr_HR: updated locale
340 [10871] locale: 'mon' array should contain both nominative and genitive
342 [12349] localedata: eu_ES: incorrect thousands separator
343 [13605] localedata: shn_MM: new Shan locale
344 [13805] localedata: ru_RU: currency should use ',' as radix point
345 [13953] localedata: km_KH: locale update
346 [13994] localedata: mjw_IN: new locale
347 [14121] build: make writes .mo files in po directory
348 [14333] libc: Fix the race between atexit() and exit()
349 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
350 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
352 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
353 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
354 width Latin characters
355 [15332] localedata: es_CU: locale update
356 [15436] stdio: Don't close or flush stdio streams on abort
357 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
359 [16148] localedata: ca_ES: incorrect thousands separator
360 [16750] dynamic-link: ldd should not try to execute the binaries
362 [16777] localedata: pl_PL: incorrect thousands separator in locale
363 [16905] localedata: hanzi: new collation
364 [17563] localedata: cmn_TW: add hanzi collation
365 [17750] localedata: wrong collation order of diacritics in most locales
366 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
368 [17956] build: Build fails on missing definitions from header file
369 nss/nss.h when Mozilla NSS is used for cryptography
370 [18203] libc: realpath() does not handle unreachable paths correctly
371 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
372 [18812] localedata: kab_DZ: new Kabyle Algeria locale
373 [18822] libc: Internal functions are called via PLT
374 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
375 [19170] libc: __gmon_start__ defined in hppa in crtn.S
376 [19574] libc: glibc should support building static PIE binaries
377 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
379 [19971] glob: glob: Do not skip entries with zero d_ino values
380 [19982] localedata: fr.po: spelling mistake for error code EXDEV
381 [20008] localedata: km_KH: convert to translit_neutral
382 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
383 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
384 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
385 [20482] localedata: de_CH: abbreviated weekdays should be two letters
386 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
388 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
389 they're set, wrong errors returned by gaih_inet when lookup functions
391 [20756] localedata: [PATCH] Use Unicode wise thousands separator
392 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
394 [20952] localedata: yuw_PG: new locale
395 [21084] localedata: charmaps/IBM858: new codepage
396 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
397 [21242] libc: assert gives pedantic warning in old gcc versions
398 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
399 __regcall calling convention
400 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
401 [21326] libc: C99 functions are not declared for C++11 and later if
402 _GNU_SOURCE is not predefined by g++
403 [21457] libc: sys/ucontext.h namespace
404 [21530] libc: tmpfile() should be implemented using O_TMPFILE
405 [21660] math: GCC fails to compile a formula with tgmath.h
406 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
407 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
408 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
409 [21684] math: tgmath.h handling of complex integers
410 [21685] math: tgmath.h handling of bit-fields
411 [21686] math: tgmath.h handling of __int128
412 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
414 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
416 [21750] localedata: column width of characters incompatible with classical
418 [21754] malloc: malloc: Perform as little work as possible after heap
419 consistency check failures
420 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
421 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
422 [21791] string: Unused XXX_chk_XXX functions in libc.a
423 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
425 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
427 [21845] localedata: Added new Locale bho_NP
428 [21853] localedata: Fix abday Which looks same as day in zh_SG
429 [21854] localedata: Added New Locale en_SC
430 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
431 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
432 _dl_runtime_resolve_avx_slow
433 [21885] network: getaddrinfo: gethosts does not release resolver context
434 on memory allocation failure
435 [21899] libc: XPG4.2 sigaction namespace
436 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
438 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
440 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
441 errno value, causing getaddrinfo to fail
442 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
443 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
444 instead of EAI_NODATA
445 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
446 constant PTRACE_SEIZE_DEVEL
447 [21930] math: C-only gcc builtins used in <math.h> isinf
448 [21932] network: Unpaired __resolv_context_get in generic get*_r
450 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
451 [21944] libc: sigval namespace
452 [21951] localedata: Update hanzi collation by stroke
453 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
454 sysdeps/x86_64/fpu/e_expf.S
455 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
456 allocation in syscall loops)
457 [21959] localedata: Fix Country name for xh_ZA
458 [21960] localedata: Fix abmon for bem_ZM
459 [21966] math: AVX2 mathvec functions use FMA without checking
460 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
461 [21971] localedata: Added New Locale for mfe_MU
462 [21972] libc: assert macro requires operator== (int) for its argument type
463 [21973] math: [sparc] libm missing sqrtl compat symbol
464 [21974] libc: Remove __bb_init_func and __bb_exit_func
465 [21982] string: stratcliff.c: error: assuming signed overflow does not
467 [21986] stdio: __guess_grouping is called incorrectly
468 [21987] math: [sparc32] wrong bits/long-double.h installed
469 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
471 [22022] localedata: Missing country_name for mni_IN
472 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
474 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
476 [22026] locale: iconv_open: heap overflow on gconv_init failure
477 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
478 [22035] math: [m68k] bits/math-inline.h macro namespace
479 [22038] localedata: Fix abbreviated weeks and months for Somali
480 [22044] localedata: Remove redundant data for Limburgish Language
481 [22050] malloc: Linking with -lmcheck does not hook
482 __malloc_initialize_hook correctly
483 [22051] libc: zero terminator in the middle of glibc's .eh_frame
484 [22052] malloc: malloc failed to compile with GCC 7 and -O3
485 [22070] localedata: charmaps/UTF-8: wcwidth for
486 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
487 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
488 Jungseong and Jongseong) should be 0
489 [22078] nss: nss_files performance issue in multi mode
490 [22082] math: bits/math-finite.h exp10 condition
491 [22086] libc: pcprofiledump incorrect cross-endian condition
492 [22093] dynamic-link: ld.so no longer searches in .../x86_64
493 [22095] network: Name server address allocation memory leak in resolv.conf
495 [22096] network: __resolv_conf_attach can incorrectly free passed conf
497 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
498 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
500 [22111] malloc: malloc: per thread cache is not returned when thread exits
501 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
502 [22134] libc: [linux] implement fexecve with execveat
503 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
505 [22145] libc: ttyname() gives up too early in the face of namespaces
506 [22146] math: C++ build issue with float128 on x86_64
507 [22153] nptl: nptl: save error code before process termination
508 [22156] libc: [hppa,ia64,microblaze] Executable stack default
509 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
510 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
511 [22165] libc: [hppa] Text relocations in libc.so
512 [22180] libc: destructor registered via __cxa_atexit is called twice
513 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
514 symlinks") cause "make" segfaults
515 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
517 [22207] libc: FAIL: stdlib/test-atexit-race
518 [22225] math: nearbyint arithmetic moved before feholdexcept
519 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
520 [22235] math: iscanonical in C++ and float128
521 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
522 the svid compat wrapper
523 [22244] math: ynf and yn are wrong without the svid compat wrapper
524 [22273] libc: Improper assert in Linux posix_spawn implementation
525 [22284] libc: -pg -pie doesn't work
526 [22292] locale: localedef exits with error 4 when it should be error 1
527 [22294] locale: Allow "" for int_currency_symbol definition in locales.
528 [22295] locale: Don't warn on non-symbolic characters in locale sources in
530 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
531 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
533 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
534 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
535 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
536 [22322] libc: [mips64] wrong bits/long-double.h installed
537 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
538 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
540 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
541 [22343] malloc: Integer overflow in posix_memalign
542 [22347] libc: getrandom() returns the number of bytes that were copied to
543 the buffer even though the comments say "Return 0 on success and -1 on
545 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
546 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
548 [22370] dynamic-link: Incorrect note padding check
549 [22375] libc: malloc returns pointer from tcache_get when should return
550 NULL (CVE-2017-17426)
551 [22377] math: iseqsig, float128 and C++
552 [22382] localedata: Error in tpi_PG locale
553 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
554 inside the ASCII printable range
555 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
557 [22403] localedata: Slash needs escaping in some locales
558 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
559 [22409] network: res_hnok does not accept some host names used on the
561 [22412] network: res_dnok, res_hnok should perform syntax checks
562 [22413] network: ns_name_pton ignores syntactically invalid trailing
564 [22415] stdio: setvbuf can lead to invalid free/segfault
565 [22432] build: Non-deterministic build
566 [22439] malloc: malloc_info should compute summary statistics for all sub-
568 [22442] network: if_nametoindex could report index for the wrong
570 [22446] build: aliasing violation calling readlink in handle_request
571 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
572 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
573 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
574 __nscd_hash/__nss_hash
575 [22463] network: p_secstodate overflow handling
576 [22469] localedata: pl_PL LC_COLLATE does not use i18n
577 [22478] libc: sigwait can fail with EINTR
578 [22505] libc: ldconfig processes include directive in locale-specific
580 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
581 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
582 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
583 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
584 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
585 [22534] localedata: Collation rules for Serbian and Bosnian should be the
587 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
588 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
589 [22577] libc: missing newline after "cannot allocate TLS data structures
591 [22588] manual: manual/conf.texi: missing underscore in front of
593 [22593] math: nextafter and nexttoward are declared with const attribute
594 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
595 [22603] string: ia64 memchr overflows internal pointer check
596 [22605] libc: SH clone does not set the exit code correctly
597 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
599 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
600 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
601 modified in case of success
602 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
603 [22615] manual: manual: ambiguous wording about errno value in case of
605 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
607 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
608 binaries or if /proc is not mounted (CVE-2017-16997)
609 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
610 [22630] build: $(no-pie-ldflag) is no longer effective
611 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
612 [22635] nptl: pthread_self returns NULL before libpthread is loaded
613 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
614 [22637] nptl: guard size is subtracted from thread stack size instead of
616 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
617 [22657] localedata: hu_HU: Avoid double space in date
618 [22660] math: fmax, fmin sNaN handling on alpha
619 [22664] libc: New warning of GCC8
620 [22665] math: alpha: ceil and floor raise inexact exceptions
621 [22666] math: alpha: trunc raise inexact exceptions
622 [22667] libc: makecontext lacks stack alignment on i386
623 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
624 [22679] libc: getcwd(3) can succeed without returning an absolute path
626 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
628 [22687] math: [powerpc-nofpu] complex long double functions spurious
630 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
631 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
632 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
634 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
635 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
637 [22701] nis: Incomplete removal of libnsl
638 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
639 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
640 [22715] dynamic-link: FAIL: elf/tst-audit10
641 [22719] libc: Backtrace tests fail on hppa
642 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
643 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
644 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
645 before the first call to crypt_r () in crypt/badsalttest.c
652 * A per-thread cache has been added to malloc. Access to the cache requires
653 no locks and therefore significantly accelerates the fast path to allocate
654 and free small amounts of memory. Refilling an empty cache requires locking
655 the underlying arena. Performance measurements show significant gains in a
656 wide variety of user workloads. Workloads were captured using a special
657 instrumented malloc and analyzed with a malloc simulator. Contributed by
658 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
660 * Unicode 10.0.0 Support: Character encoding, character type info, and
661 transliteration tables are all updated to Unicode 10.0.0, using
662 generator scripts contributed by Mike FABIAN (Red Hat).
663 These updates cause user visible changes, especially the changes in
664 wcwidth for many emoji characters cause problems when emoji sequences
665 are rendered with pango, see for example:
666 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
668 * Collation of Hungarian has been overhauled and is now consistent with "The
669 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
672 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
674 - The GNU C Library will now detect when /etc/resolv.conf has been
675 modified and reload the changed configuration. The new resolver option
676 “no-reload” (RES_NORELOAD) disables this behavior.
678 - The GNU C Library now supports an arbitrary number of search domains
679 (configured using the “search” directive in /etc/resolv.conf);
680 previously, there was a hard limit of six domains. For backward
681 compatibility, applications that directly modify the ‘_res’ global
682 object are still limited to six search domains.
684 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
685 Library will now randomly pick a name server from the configuration as a
686 starting point. (Previously, the second name server was always used.)
688 * The tunables feature is now enabled by default. This allows users to tweak
689 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
691 * New function reallocarray, which resizes an allocated block (like realloc)
692 to the product of two sizes, with a guaranteed clean failure upon integer
693 overflow in the multiplication. Originally from OpenBSD, contributed by
694 Dennis Wölfing and Rüdiger Sonderfeld.
696 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
697 These are extended versions of preadv and pwritev, respectively, taking an
698 additional flags argument. The set of supported flags depends on the
699 running kernel; full support currently requires kernel 4.7 or later.
701 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
702 create a new session ID for the spawned process. This feature is
703 scheduled to be added to the next major revision of POSIX; for the time
704 being, it is available under _GNU_SOURCE.
706 * errno.h is now safe to use from C-preprocessed assembly language on all
707 supported operating systems. In this context, it will only define the
708 Exxxx constants, as preprocessor macros expanding to integer literals.
710 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
711 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
712 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
713 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
715 To compile programs that use this feature, the compiler must support
716 128-bit floating point with the type name _Float128 (as defined by TS
717 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
718 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
719 must be defined to make the new interfaces visible.
721 The new functions and macros correspond to those present for other
722 floating-point types (except for a few obsolescent interfaces not
723 supported for the new type), with F128 or f128 suffixes; for example,
724 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
725 printf or scanf formats for the new type; the strfromf128 and strtof128
726 interfaces should be used instead.
728 Deprecated and removed features, and other changes affecting compatibility:
730 * The synchronization that pthread_spin_unlock performs has been changed to
731 now be equivalent to a C11 atomic store with release memory order to the
732 spin lock's memory location. Previously, several (but not all)
733 architectures used stronger synchronization (e.g., containing what is
734 often called a full barrier). This change can improve performance, but
735 may affect odd fringe uses of spin locks that depend on the previous
736 behavior (e.g., using spin locks as atomic variables to try to implement
737 Dekker's mutual exclusion algorithm).
739 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
742 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
743 will only be built and installed when the GNU C Library is configured with
744 --enable-obsolete-rpc. This allows alternative RPC implementations, such
745 as TIRPC or rpcsvc-proto, to be used.
747 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
748 libnss_compat, are deprecated, and will not be built or installed by
751 The NIS(+) support library, libnsl, is also deprecated. By default, a
752 compatibility shared library will be built and installed, but not headers
753 or development libraries. Only a few NIS-related programs require this
754 library. (In particular, the GNU C Library has never required programs
755 that use 'gethostbyname' to be linked with libnsl.)
757 Replacement implementations based on TIRPC, which additionally support
758 IPv6, are available from <https://github.com/thkukuk/>. The configure
759 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
760 name service modules, to be built and installed.
762 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
763 support is enabled, the configured recursive resolver must support EDNS.
764 (Responding to EDNS-enabled queries with responses which are not
765 EDNS-enabled is fine, but FORMERR responses are not.)
767 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
768 servers have not supported this opcode for a long time.
770 * The _res_opcodes variable has been removed from libresolv. It had been
771 exported by accident.
773 * <string.h> no longer includes inline versions of any string functions,
774 as this kind of optimization is better done by the compiler. The macros
775 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
777 * The nonstandard header <xlocale.h> has been removed. Most programs should
778 use <locale.h> instead. If you have a specific need for the definition of
779 locale_t with no other declarations, please contact
780 libc-alpha@sourceware.org and explain.
782 * The obsolete header <sys/ultrasound.h> has been removed.
784 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
786 * The obsolete function cfree has been removed. Applications should use
789 * The stack_t type no longer has the name struct sigaltstack. This changes
790 the C++ name mangling for interfaces involving this type.
792 * The ucontext_t type no longer has the name struct ucontext. This changes
793 the C++ name mangling for interfaces involving this type.
795 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
796 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
797 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
798 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
799 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
800 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
801 fpregset_t type no longer has the name struct fpu. This changes the C++
802 name mangling for interfaces involving those types.
804 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
805 synced with the kernel:
807 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
808 are not supported on this architecture and have been removed.
810 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
811 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
812 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
814 Programs that assume the GET/SETREGS ptrace requests are universally
815 available will now fail to build, instead of malfunctioning at runtime.
817 Changes to build and runtime requirements:
819 * Linux kernel 3.2 or later is required at runtime, on all architectures
820 supported by that kernel. (This is a change from version 2.25 only for
823 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
825 * On most architectures, GCC 4.9 or later is required to build the GNU C
826 Library. On powerpc64le, GCC 6.2 or later is required.
828 Older GCC versions and non-GNU compilers are still supported when
829 compiling programs that use the GNU C Library. (We do not know exactly
830 how old, and some GNU extensions to C may be _de facto_ required. If you
831 are interested in helping us make this statement less vague, please
832 contact libc-alpha@sourceware.org.)
834 Security related changes:
836 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
837 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
839 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
840 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
842 * Avoid printing a backtrace from the __stack_chk_fail function since it is
843 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
846 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
847 fixed (CVE-2017-12133).
849 The following bugs are resolved with this release:
851 [984] network: Respond to changed resolv.conf in gethostbyname
852 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
853 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
855 [12189] libc: __stack_chk_fail should not attempt a backtrace
857 [14096] time: Race condition on timezone/tst-timezone.out
858 [14172] localedata: az_IR: new locale
859 [14995] build: glibc fails to build if gold is the default linker, even if
861 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
862 [16637] network: inet_pton function is accepting IPv6 with bad format
863 [16640] string: string/strtok.c: undefined behaviour inconsistent between
864 x86 and other generic code
865 [16875] localedata: ko_KR: fix lang_name
866 [17225] localedata: ar_SY: localized month names for May and June are
868 [17297] localedata: da_DK: wrong date_fmt string
869 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
870 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
871 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
872 [19066] localedata: ar_SA abbreviated day and month names are in English
873 [19569] network: resolv: Support an arbitrary number of search domains
874 [19570] network: Implement random DNS server selection in the stub
876 [19838] locale: localedef fails on PA-RISC
877 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
878 order of 0D36 and 0D37
879 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
881 [20098] libc: FAIL: debug/backtrace-tst on hppa
882 [20257] network: sunrpc: clntudp_call does not enforce timeout when
884 [20275] localedata: locale day/abday/mon/abmon should not have trailing
886 [20313] localedata: Update locale data to Unicode 9.0
887 [20424] manual: Document how to provide a malloc replacement
888 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
890 [20686] locale: Add el_GR@euro to SUPPORTED.
891 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
892 failures consistently
893 [21015] dynamic-link: Document and fix --enable-bind-now
894 [21016] nptl: pthread_cond support is broken on hppa
895 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
896 frame-pointer on i386
897 [21049] libc: segfault in longjmp_chk() due to clobbered processor
899 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
900 [21088] libc: Build fails with --enable-static-nss
901 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
902 generic c code is used
903 [21109] libc: Tunables broken on big-endian
904 [21112] math: powf has large ulp errors with base close to 1 and exponent
906 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
908 [21120] malloc: glibc malloc is incompatible with GCC 7
909 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
911 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
912 (-0.0) when linking with -lieee
913 [21171] math: log10, log2 and lgamma return incorrect results
914 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
915 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
916 [21207] localedata: ce_RU: update weekdays from CLDR
917 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
918 [21217] localedata: Update months from CLDR-31
919 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
920 [21243] libc: support_delete_temp_file should issue warning for failed
922 [21244] libc: support resolv_test_start() socket fd close should be
924 [21253] libc: localedef randomly segfaults when using -fstack-check due to
925 new posix_spawn implementation
926 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
927 leads to lower CPU frequency
928 [21259] libc: [alpha] termios.h missing IXANY for POSIX
929 [21261] libc: [sparc64] bits/setjmp.h namespace
930 [21267] network: [mips] bits/socket.h IOC* namespace
931 [21268] libc: [alpha] termios.h NL2, NL3 namespace
932 [21270] libc: mmap64 silently truncates large offset values
933 [21275] libc: posix_spawn always crashes on ia64 now
934 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
935 [21280] math: [powerpc] logbl for POWER7 return incorrect results
936 [21289] libc: Incorrect declaration for 32-bit platforms with
937 _FILE_OFFSET_BITS=64 causes build error
938 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
939 not support gethostbyname4_r
940 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
942 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
944 [21340] libc: Support POSIX_SPAWN_SETSID
945 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
946 [21359] network: ns_name_pack needs additional byte in destination buffer
947 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
948 against fragmentation attacks (CVE-2017-12132)
949 [21369] network: resolv: Remove EDNS fallback
950 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
952 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
953 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
954 [21393] stdio: Missing dup3 error check in freopen, freopen64
955 [21396] libc: Use AVX2 memcpy/memset on Skylake server
956 [21399] localedata: Bad description for U00EC in
957 localedata/charmaps/CP1254
958 [21411] malloc: realloc documentation error
959 [21426] network: sys/socket.h uio.h namespace
960 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
961 [21445] libc: signal.h bsd_signal namespace
962 [21455] network: Network headers stdint.h namespace
963 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
965 [21475] network: resolv: Overlong search path is truncated mid-label
966 [21511] libc: sigstack namespace
967 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
968 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
969 bits/syscall.h: No such file or directory
970 [21517] libc: struct sigaltstack namespace
971 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
972 [21533] localedata: Update locale data to Unicode 10.0
974 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
975 junk at end of line, first unrecognized character is `@'
976 [21538] libc: SIG_HOLD missing for XPG4
977 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
978 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
979 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
980 [21543] libc: sigevent namespace
981 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
982 size for general purpose registers in mcontext_t structure
983 [21550] libc: sigwait namespace
984 [21552] libc: XPG4 bsd_signal namespace
985 [21554] libc: sigpause namespace
986 [21560] libc: sys/wait.h signal.h namespace
987 [21561] libc: waitid namespace
988 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
989 '/usr/include/stdlib.h': Permission denied
990 [21575] libc: sys/wait.h missing struct rusage definition
991 [21584] libc: sigaltstack etc namespace
992 [21597] libc: siginterrupt namespace
993 [21607] math: hppa: FAIL: math/test-tgmath
994 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
996 [21622] libc: [tile] missing SA_* for POSIX.1:2008
997 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
998 stack and heap (CVE-2017-1000366)
999 [21625] libc: wait3 namespace
1000 [21654] nss: Incorrect pointer alignment in NSS group merge result
1002 [21657] network: Parse interface zone id for node-local multicast
1003 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1005 [21666] libc: .symver is used on common symbol
1006 [21668] network: resolv: res_init cross-thread broadcast introduces race
1008 [21687] math: tgmath.h totalorder, totalordermag return type
1009 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1011 [21696] libc: Incorrect assumption of of __cpu_mask in
1012 posix/sched_cpucount.c
1013 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1014 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1016 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1017 leading to relocation crash
1018 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1019 `resolv_conf_matches (resp, conf)' failed.
1020 [21710] localedata: Added Samoan language locale for Samoa
1021 [21711] localedata: Pashto yesstr/nostr locale are missing
1022 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1023 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1024 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1025 [21723] localedata: yesstr/nostr missing for Chinese language locale
1026 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1028 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1030 [21728] localedata: New Locale for Tongan language
1031 [21729] localedata: incorrect LC_NAME fields for hi_IN
1032 [21733] localedata: yesstr and nostr are missing for zh_HK
1033 [21734] localedata: Missing yesstr and nostr are for kw_GB
1034 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1035 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1036 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1037 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1039 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1041 [21749] localedata: Wrong abbreviated day name (“abday”) for
1043 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1044 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1045 [21759] localedata: missing yesstr and nostr for Tigrinya
1046 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1047 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1048 [21767] localedata: Missing Bislama locales
1049 [21768] localedata: Missing yesstr and nostr for aa_ET
1050 [21770] localedata: Missing Field in li_NL
1051 [21778] nptl: Robust mutex may deadlock
1052 [21779] libc: MicroBlaze segfaults when loading libpthread
1053 [21783] localedata: Fix int_select international_call_prefixes
1054 [21784] localedata: Inconsistency in country_isbn
1055 [21788] localedata: Missing Country Postal Abbreviations
1056 [21794] localedata: Added-country_isbn-for-Italy
1057 [21795] localedata: Add/Fix country_isbn for France
1058 [21796] localedata: Added country_isbn for Republic of Korea
1059 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1060 [21799] localedata: Added int_select international_call_prefixes
1061 [21801] localedata: Added int_select international_call_prefixes
1062 [21804] nptl: Double semicolon in thread-shared-types.h
1063 [21807] localedata: LC_ADDRESS fix for pap_CW
1064 [21808] localedata: Fix LC_ADDRESS for pap_AW
1065 [21821] localedata: Added country_name in mai_IN
1066 [21822] localedata: Fix LC_TIME for mai_IN
1067 [21823] localedata: missing yesstr, nostr for sa_IN
1068 [21825] localedata: Fix name_mrs for mag_IN
1069 [21828] localedata: 2.26 changelog should mention user visible changes
1071 [21835] localedata: Added Maithili language locale for Nepal
1072 [21838] localedata: Removed redundant data for the_NP
1073 [21839] localedata: Fix LC_MONETARY for ta_LK
1074 [21844] localedata: Fix Latin characters and Months Sequence.
1075 [21848] localedata: Fix mai_NP Title Name
1080 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1081 24731-2:2010, is supported to enable declarations of functions from that
1082 TR. Note that not all functions from that TR are supported by the GNU C
1085 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1086 18661-1:2014, is supported to enable declarations of functions and macros
1087 from that TS. Note that not all features from that TS are supported by
1090 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1091 18661-4:2015, is supported to enable declarations of functions and macros
1092 from that TS. Note that most features from that TS are not supported by
1095 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1096 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1097 Since the GNU C Library defaults to a much newer revision of POSIX, this
1098 will only affect programs that specifically request an old conformance
1099 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1100 see a change in the visible declarations, but a program compiled with
1101 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1104 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1105 defined by all multithreaded code, but glibc has not required this for
1108 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1109 means that in a future release, the macros “major”, “minor”, and “makedev”
1110 will only be available from <sys/sysmacros.h>.
1112 These macros are not part of POSIX nor XSI, and their names frequently
1113 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1114 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1115 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1118 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1119 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1120 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1122 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1123 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1124 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1125 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1126 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1127 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1128 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1129 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1130 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1131 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1132 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1135 * New <math.h> features are added from TS 18661-1:2014:
1137 - Signaling NaN macros: SNANF, SNAN, SNANL.
1139 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1140 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1141 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1143 - llogb functions: the llogb, llogbf and llogbl functions, and the
1144 FP_LLOGB0 and FP_LLOGBNAN macros.
1146 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1149 - Comparison macros: iseqsig.
1151 - Classification macros: iscanonical, issubnormal, iszero.
1153 - Total order functions: totalorder, totalorderf, totalorderl,
1154 totalordermag, totalordermagf, totalordermagl.
1156 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1158 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1159 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1161 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1162 are added to libc. They convert a floating-point number into string.
1164 * Most of glibc can now be built with the stack smashing protector enabled.
1165 It is recommended to build glibc with --enable-stack-protector=strong.
1166 Implemented by Nick Alcock (Oracle).
1168 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1169 intended to be used instead of memset() to erase sensitive data after use;
1170 the compiler will not optimize out calls to explicit_bzero even if they
1171 are "unnecessary" (in the sense that no _correct_ program can observe the
1172 effects of the memory clear).
1174 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1175 to float instead of double. This does not affect the ABI of any libraries
1176 that are part of the GNU C Library, but may affect the ABI of other
1177 libraries that use this type in their interfaces.
1179 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1180 float_t and double_t types are now defined to long double instead of float
1181 and double. These options are not the default, and this does not affect
1182 the ABI of any libraries that are part of the GNU C Library, but it may
1183 affect the ABI of other libraries that use this type in their interfaces,
1184 if they are compiled or used with those options.
1186 * The getentropy and getrandom functions, and the <sys/random.h> header file
1189 * The buffer size for byte-oriented stdio streams is now limited to 8192
1190 bytes by default. Previously, on Linux, the default buffer size on most
1191 file systems was 4096 bytes (and thus remains unchanged), except on
1192 network file systems, where the buffer size was unpredictable and could be
1193 as large as several megabytes.
1195 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1196 for the Linux quota interface which predates kernel version 2.4.22 has
1199 * The malloc_get_state and malloc_set_state functions have been removed.
1200 Already-existing binaries that dynamically link to these functions will
1201 get a hidden implementation in which malloc_get_state is a stub. As far
1202 as we know, these functions are used only by GNU Emacs and this change
1203 will not adversely affect already-built Emacs executables. Any undumped
1204 Emacs executables, which normally exist only during an Emacs build, should
1205 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1207 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1208 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1209 “no-ip6-dotint” had already been the default, and support for the
1210 “ip6-dotint” option was removed from the Internet in 2006.
1212 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1213 flag from <resolv.h> have been removed. The option relied on a
1214 backwards-incompatible DNS extension which was never deployed on the
1217 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1218 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1219 They were already unimplemented.
1221 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1222 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1223 removed again from the IETF name lookup interface specification in RFC
1224 2553. Applications should use getaddrinfo instead.
1226 * DNSSEC-related declarations and definitions have been removed from the
1227 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1228 decode the data part of DNSSEC record types. Previous versions of glibc
1229 only implemented minimal support for the previous version of DNSSEC, which
1230 is incompatible with the currently deployed version.
1232 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1233 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1234 <arpa/nameser.h> header file because the distinction between RR types and
1235 meta-RR types is not officially standardized, subject to revision, and
1236 thus not suitable for encoding in a macro.
1238 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1239 and rhook members of the res_state type in <resolv.h> have been removed.
1240 The glibc stub resolver did not support these hooks, but the header file
1241 did not reflect that.
1243 * For multi-arch support it is recommended to use a GCC which has
1244 been built with support for GNU indirect functions. This ensures
1245 that correct debugging information is generated for functions
1246 selected by IFUNC resolvers. This support can either be enabled by
1247 configuring GCC with '--enable-gnu-indirect-function', or by
1248 enabling it by default by setting 'default_gnu_indirect_function'
1249 variable for a particular architecture in the GCC source file
1252 * GDB pretty printers have been added for mutex and condition variable
1253 structures in POSIX Threads. When installed and loaded in gdb these pretty
1254 printers show various pthread variables in human-readable form when read
1255 using the 'print' or 'display' commands in gdb.
1257 * Tunables feature added to allow tweaking of the runtime for an application
1258 program. This feature can be enabled with the '--enable-tunables' configure
1259 flag. The GNU C Library manual has details on usage and README.tunables has
1260 instructions on adding new tunables to the library.
1262 * A new version of condition variables functions have been implemented in
1263 the NPTL implementation of POSIX Threads to provide stronger ordering
1266 * A new version of pthread_rwlock functions have been implemented to use a more
1267 scalable algorithm primarily through not using a critical section anymore to
1270 Security related changes:
1272 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1273 have been created with makecontext could fail to terminate due to a
1274 missing .cantunwind annotation. This has been observed to lead to a hang
1275 (denial of service) in some Go applications compiled with gccgo. Reported
1276 by Andreas Schwab. (CVE-2016-6323)
1278 * The DNS stub resolver functions would crash due to a NULL pointer
1279 dereference when processing a query with a valid DNS question type which
1280 was used internally in the implementation. The stub resolver now uses a
1281 question type which is outside the range of valid question type values.
1284 The following bugs are resolved with this release:
1286 [4099] stdio: Overly agressive caching by stream i/o functions.
1287 [7065] build: Support building glibc with -fstack-protector or -fstack-
1289 [9842] localedata: en_CA: incorrect date format
1290 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1291 before it started waiting
1292 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1293 [15765] nptl: sem_open is wrongly a cancellation point
1294 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1295 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1297 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1298 library linked with pthread
1299 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1300 "FIXME: Ingo" issue)
1301 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1302 [17252] libc: getrandom and getentropy syscall
1303 [17730] malloc: thread-local storage is sometimes improperly free()'d
1304 after being __libc_memalign()'d
1305 [18241] stdio: failed fseek on memstream does not set errno and can fail
1307 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1309 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1310 PTHREAD_PRIO_INHERIT mutex on ARM
1311 [18784] network: res_query and related function crash for special record
1312 type queries (CVE-2015-5180)
1313 [19380] math: strtod does not raise "inexact"
1314 [19387] string: Integer overflow in memchr
1315 [19390] string: Integer overflow in strncat
1316 [19398] build: linknamespace tests fail with massively parallel build
1317 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1319 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1320 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1321 [19514] libc: [PATCH] Fix spelling errors in spelling
1322 "implement"/"implementation" in several places
1323 [19582] network: Deprecate RES_USE_INET6
1324 [19673] manual: clog10 docs appear to be erroneous
1325 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1326 causes a segmentation fault
1327 [19826] libc: invalid pointer returned from __tls_get_addr with static
1329 [20016] network: resolv: Remove hooks support from the API
1330 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1332 [20033] math: [x86_64] vectorized math function don't call the __finite
1334 [20116] nptl: use after free in pthread_create
1335 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1336 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1338 [20311] nptl: please install proc_service.h
1339 [20366] libc: Compilation errors in installed headers in strict-compliance
1341 [20370] malloc: malloc: Arena free list management is still racy
1342 (incorrect fix in bug 19243)
1343 [20386] libc: assert (X = 0) does not result in GCC warning
1344 [20432] malloc: malloc: Minimize interface required for interposition
1345 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1346 in _Unwind_Backtrace (CVE-2016-6323)
1347 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1348 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1350 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1351 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1352 [20477] network: resolv: incorrect double-checked locking related to
1354 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1356 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1358 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1359 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1360 cause transition penalty
1361 [20517] math: sparcv9 missing fdiml compat symbol
1362 [20524] manual: strverscmp is inconsistent
1363 [20525] libc: <sys/quota.h> should be based on kernel headers
1364 [20539] math: GCC 7 -static -lm fails to link at -O3
1365 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1367 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1369 [20591] network: Remove obsolete DNSSEC support
1370 [20592] network: DNS resource record type classification macros in
1371 <arpa/nameser.h> are incorrect
1372 [20593] network: Update DNS RR type definitions
1373 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1374 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1376 [20629] network: libresolv: Remove support for bitstring labels
1378 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1379 assertion: "maxidx >= 2"
1380 [20660] math: [arm] Use VSQRT
1381 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1382 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1383 [20689] libc: Test for FMA should also check for AVX.
1384 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1385 [20715] math: iszero macro breaks existing code
1386 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1387 [20728] libc: powerpc: Missing TOC stub in clone
1388 [20729] build: glibc-2.24 fails to build for i486 with -Os
1389 [20750] build: Build fails with default PIE enabled gcc-6.x
1390 [20768] math: [alpha] sqrt fegetenv namespace
1391 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1393 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1394 __FLT_EVAL_METHOD__ is 2
1395 [20790] network: rpcgen buffer overrun in get_prog_declaration
1396 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1397 [20829] libc: crypt snprintf namespace
1398 [20847] libc: tst-vfork3 failure
1399 [20855] math: Default bits/mathdef.h has inappropriate float_t
1400 [20859] math: [sh4] FP_ILOGB0 invalid
1401 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1402 U+20AC), not same as GBK
1403 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1404 [20916] math: pow handling of sNaN arguments
1405 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1406 [20919] math: Bad pow (qNaN, 0) result with -lieee
1407 [20940] math: hypot sNaN handling
1408 [20947] math: fmax, fmin sNaN handling
1409 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1410 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1412 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1413 [20973] nptl: robust mutexes: Lost wake-ups
1414 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1415 [20978] nis: strlen on null pointer in nss_nisplus
1416 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1418 [21014] string: i686 memchr overflows internal pointer check
1419 [21019] libc: [mips] n32 lseek incorrect on overflow
1420 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1421 [21026] libc: [MIPS] readahead syscall is broken on n64
1422 [21028] math: Fallback fesetexceptflag should always succeed
1423 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1424 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1426 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1427 [21061] librt: [microblaze] librt lost clock_* exports
1428 [21073] libc: tunables: insecure environment variables passed to
1429 subprocesses with AT_SECURE
1430 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1434 * The minimum Linux kernel version that this version of the GNU C Library
1435 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1436 kernel version 2.6.32 or later suffices (on architectures that already
1437 required kernel versions more recent than 3.2, those requirements remain
1438 unchanged). Linux 3.2 or later kernel headers are required on all
1441 * The pap_AN locale has been deleted. This has been deprecated for a long
1442 time. It has been replaced by pap_AW & pap_CW, both of which have long
1443 been included in previous releases.
1445 * The readdir_r and readdir64_r functions have been deprecated. It is
1446 recommended to use readdir and readdir64 instead.
1448 * The type “union wait” has been removed. It was deprecated in the early
1449 1990s and never part of POSIX. Application code should use the int type
1450 instead of “union wait”.
1452 * A new NSS action is added to facilitate large distributed system
1453 administration. The action, MERGE, allows remote user stores like LDAP
1454 to be merged into local user stores like /etc/groups in order to provide
1455 easy to use, updated, and managed sets of merged credentials. The new
1456 action can be used by configuring it in /etc/nsswitch.conf:
1457 group: files [SUCCESS=merge] nis
1458 Implemented by Stephen Gallagher (Red Hat).
1460 * The deprecated __malloc_initialize_hook variable has been removed from the
1463 * The long unused localedef --old-style option has been removed. It hasn't
1464 done anything in over 16 years. Scripts using this option can safely
1467 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1468 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1469 return the next representable value in the direction of positive infinity
1470 and the nextdown functions return the next representable value in the
1471 direction of negative infinity. These are currently enabled as GNU
1474 Security related changes:
1476 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1477 could result in a stack overflow when getnetbyname was called with an
1478 overly long name. (CVE-2016-3075)
1480 * Previously, getaddrinfo copied large amounts of address data to the stack,
1481 even after the fix for CVE-2013-4458 has been applied, potentially
1482 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1483 instead. Reported by Michael Petlan. (CVE-2016-3706)
1485 * The glob function suffered from a stack-based buffer overflow when it was
1486 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1487 Reported by Alexander Cherepanov. (CVE-2016-1234)
1489 * The Sun RPC UDP client could exhaust all available stack space when
1490 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1491 alloca plugin for GCC. (CVE-2016-4429)
1493 * The IPv6 name server management code in libresolv could result in a memory
1494 leak for each thread which is created, performs a failing naming lookup,
1495 and exits. Over time, this could result in a denial of service due to
1496 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1498 The following bugs are resolved with this release:
1500 [1170] localedata: ne_NP: update Nepali locale definition file
1501 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1503 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1504 [6796] math: fdim() does not set errno on overflow
1505 [10354] libc: posix_spawn should use vfork() in more cases than presently
1506 [11213] localedata: localedata: add copyright disclaimer to locale files
1507 [12143] localedata: chr_US: new Cherokee locale
1508 [12450] localedata: sgs_LT: new locale
1509 [12676] localedata: ln_CD: new locale
1510 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1512 [13304] math: fma, fmaf, fmal produce wrong results
1513 [14259] build: --localedir arg to configure is ignored
1514 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1515 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1516 [14934] localedata: es_CL: wrong first weekday chilean locale
1517 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1519 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1521 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1522 [15368] nptl: raise() is not async-signal-safe
1523 [15479] math: ceil, floor, round and trunc raise inexact exception
1524 [15578] localedata: kk_KZ: various updates
1525 [16003] localedata: pap_AN: punt old locale
1526 [16137] localedata: iw_IL: punt old locale
1527 [16190] localedata: eo: new esperanto locale
1528 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1529 [16742] malloc: race condition: pthread_atfork() called before first
1530 malloc() results in unexpected locking behaviour/deadlocks
1531 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1533 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1534 [17565] localedata: pt_PT: wrong (work-)week start
1535 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1536 [17950] build: Build fails with -msse
1537 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1538 [18433] libc: posix_spawn does not return correctly upon failure to
1540 [18453] localedata: charmaps/IBM875: incorrect codes
1541 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1543 [18896] localedata: he_IL: improvements for currency
1544 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1545 Romanian locale data
1546 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1548 [19056] libc: Deprecate readdir_r
1549 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1551 [19198] localedata: nl_NL: small improvements for Dutch locales
1552 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1553 nameservers (CVE-2016-5417)
1554 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1556 [19400] locale: Language missing in "iso-639.def", trivial fix in
1558 [19431] malloc: Deadlock between fflush, getdelim, and fork
1559 [19505] libc: Incorrect file descriptor validity checks in
1560 posix_spawn_file_actions_add{open,close,dup2}
1561 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1562 when using RTLD_NEXT
1563 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1564 `intl/{gettextP,loadinfo}.h'
1565 [19534] libc: execle, execlp may use malloc
1566 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1567 [19573] network: res_nclose and __res_maybe_init disagree about name
1568 server initialization, breaking Hesiod
1569 [19575] localedata: Status of GB18030 tables
1570 [19581] localedata: sr_* date_fmt string contains additional newline
1571 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1573 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1574 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1575 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1576 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1578 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1580 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1582 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1583 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1584 [19633] locale: strfmon_l applies global locale to number formatting
1585 [19642] network: Memory leak in getnameinfo
1586 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1587 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1589 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1590 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1592 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1593 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1595 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1597 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1599 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1600 pointers and lengths in error-case.
1601 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1602 report errors on UTF-16 surrogates values.
1603 [19755] nscd: nscd assertion failure in gc
1604 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1605 [19759] libc: mempcpy shouldn't be inlined
1606 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1607 [19765] libc: s390 needs an optimized mempcpy
1608 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1609 NAME_MAX limit assumption (CVE-2016-1234)
1610 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1611 [19787] network: Missing and incorrect truncation checks in getnameinfo
1612 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1614 [19791] network: Assertion failure in res_query.c with un-connectable name
1616 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1617 [19822] math: libm.so install clobbers old version
1618 [19825] network: resolv: send_vc can return uninitialized data in second
1619 response to getaddrinfo
1620 [19830] network: nss_dns: should check RDATA length against buffer length
1621 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1622 confronted with A/AAAA records of invalid size
1623 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1624 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1625 [19853] stdio: Printing IBM long double in decimal with high precision is
1627 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1628 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1629 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1631 [19865] network: Assertion failure or memory leak in
1632 _nss_dns_getcanonname_r
1633 [19868] network: nss_dns: netent code does not skip over non-PTR records
1634 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1636 [19881] string: Improve x86-64 memset
1637 [19907] string: Incorrect memcpy tests
1638 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1640 [19925] libc: termios.h XCASE namespace
1641 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1642 [19929] libc: limits.h NL_NMAX namespace
1643 [19931] stdio: Memory leak in vfprintf
1644 [19957] libc: clone(CLONE_VM) access invalid parent memory
1645 [19963] localedata: en_IL: New locale
1646 [19989] stdio: stdio.h cuserid namespace
1647 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1649 [19996] locale: langinfo.h nl_langinfo_l namespace
1650 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1651 [20010] network: getaddrinfo: Stack overflow in hostent translation
1653 [20012] stdio: libio: fmemopen append mode failure
1654 [20014] stdio: stdio.h namespace for pre-threads POSIX
1655 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1656 [20023] libc: fcntl.h timespec namespace
1657 [20024] math: [x86_64] vectorized sincos trashes the stack
1658 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1659 [20041] time: sys/time.h timespec namespace
1660 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1661 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1662 [20051] libc: ttyslot in wrong header under wrong conditions
1663 [20054] libc: gethostname not declared for XPG4
1664 [20055] libc: termios.h missing tcgetsid for XPG4
1665 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1667 [20073] libc: sys/stat.h fchmod namespace
1668 [20074] libc: stdlib.h rand_r namespace
1669 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1670 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1672 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1673 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1675 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1676 [20119] libc: Wrong mask for processors level type from CPUID
1677 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1678 AS not supporting AVX512
1679 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1681 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1682 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1684 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1685 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1686 [20178] libc: posix_spawn{p} should not call exit
1687 [20191] stdio: libio: vtables hardening
1688 [20195] string: FMA4 detection requires CPUID execution with register
1690 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1691 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1693 [20212] math: acos (sNaN) returns sNaN
1694 [20213] math: asin (sNaN) returns sNaN
1695 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1696 [20218] math: [i386] asinhl (sNaN) returns sNaN
1697 [20219] math: [i386] atanhl (sNaN) returns sNaN
1698 [20222] stdio: fopencookie: Mangle function pointers
1699 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1700 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1701 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1703 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1704 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1705 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1706 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1707 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1708 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1709 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1710 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1711 [20240] math: modf (sNaN) returns sNaN
1712 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1713 [20250] math: frexp (sNaN) returns sNaN
1714 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1715 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1717 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1718 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1719 subscript is below array bounds [-Werror=array-bounds]
1720 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1721 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1723 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1724 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1726 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1727 "invalid" exceptions
1728 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1729 [20316] localedata: id_ID: Februari instead of Pebruari
1730 [20327] string: POWER8 strcasecmp returns incorrect result
1731 [20347] math: Failure: Test: j0_downward (0xap+0)
1732 [20348] libc: FAIL: misc/tst-preadvwritev64
1733 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1734 [20350] libc: There is no test for p{read,write}64
1735 [20357] math: Incorrect cos result for 1.5174239687223976
1736 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1740 * Unicode 8.0.0 Support: Character encoding, character type info, and
1741 transliteration tables are all updated to Unicode 8.0.0, using new
1742 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1743 These updates cause user visible changes, such as the fixes for bugs
1744 89, 16061, and 18568.
1746 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1747 kernel-internal CPU set size. This means that requests that change the
1748 CPU affinity which failed before (for example, an all-ones CPU mask) will
1749 now succeed. Applications that need to determine the effective CPU
1750 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1751 setting it because the kernel can adjust it (and the previous size check
1752 would not detect this in the majority of cases).
1754 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1755 the following new symbols are used: fts64_children, fts64_close,
1756 fts64_open, fts64_read and fts64_set.
1758 * getaddrinfo now detects certain invalid responses on an internal netlink
1759 socket. If such responses are received, an affected process will
1760 terminate with an error message of "Unexpected error <number> on netlink
1761 descriptor <number>" or "Unexpected netlink response of size <number> on
1762 descriptor <number>". The most likely cause for these errors is a
1763 multi-threaded application which erroneously closes and reuses the netlink
1764 file descriptor while it is used by getaddrinfo.
1766 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1767 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1768 unnecessary serialization of memory allocation requests across threads.
1769 The defect is now corrected. Users should see a substantial increase in
1770 the concurent throughput of allocation requests for applications which
1771 trigger this bug. Affected applications typically create and
1772 destroy threads frequently. (Bug 19048 was reported and analyzed by
1775 * There is now a --disable-timezone-tools configure option for disabling the
1776 building and installing of the timezone related utilities (zic, zdump, and
1777 tzselect). This is useful for people who build the timezone data and code
1778 independent of the GNU C Library.
1780 * The obsolete header <regexp.h> has been removed. Programs that require
1781 this header must be updated to use <regex.h> instead.
1783 * The obsolete functions bdflush, create_module, get_kernel_syms,
1784 query_module and uselib are no longer available to newly linked binaries;
1785 the header <sys/kdaemon.h> has been removed. These functions and header
1786 were specific to systems using the Linux kernel and could not usefully be
1787 used with the GNU C Library on systems with version 2.6 or later of the
1790 * Optimized string, wcsmbs and memory functions for IBM z13.
1791 Implemented by Stefan Liebler.
1793 * Newly linked programs that define a variable called signgam will no longer
1794 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1795 require signgam to be set by those functions must ensure that they use the
1796 variable provided by the GNU C Library and declared in <math.h>, without
1797 defining their own copy.
1799 * The minimum GCC version that can be used to build this version of the GNU
1800 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1801 still be used to compile programs using the GNU C Library.
1803 Security related changes:
1805 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1806 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1808 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1809 the pointer guard feature. It is always enabled. Previously,
1810 LD_POINTER_GUARD could be used to disable security hardening in binaries
1811 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1814 * An integer overflow in hcreate and hcreate_r could lead to an
1815 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1817 * The catopen function no longer has unbounded stack usage. Reported by
1818 Max. (CVE-2015-8779)
1820 * The nan, nanf and nanl functions no longer have unbounded stack usage
1821 depending on the length of the string passed as an argument to the
1822 functions. Reported by Joseph Myers. (CVE-2014-9761)
1824 * A stack-based buffer overflow was found in libresolv when invoked from
1825 libnss_dns, allowing specially crafted DNS responses to seize control
1826 of execution flow in the DNS client. The buffer overflow occurs in
1827 the functions send_dg (send datagram) and send_vc (send TCP) for the
1828 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1829 family. The use of AF_UNSPEC triggers the low-level resolver code to
1830 send out two parallel queries for A and AAAA. A mismanagement of the
1831 buffers used for those queries could result in the response of a query
1832 writing beyond the alloca allocated buffer created by
1833 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1834 the overflow. Thanks to the Google Security Team and Red Hat for
1835 reporting the security impact of this issue, and Robert Holiday of
1836 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1838 The following bugs are resolved with this release:
1840 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1841 [887] math: Math library function "logb" and "nextafter" inconsistent
1842 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1843 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1844 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1845 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1847 [4404] localedata: German translation of "Alarm clock" is misleading
1848 [6799] math: nextafter() and nexttoward() doen't set errno on
1849 overflow/underflow errors
1850 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1852 [10432] nis: _nss_nis_setnetgrent assertion failure
1853 [11460] libc: fts has no LFS support
1854 [12926] network: getaddrinfo()/make_request() may spin forever
1855 [13065] nptl: Race condition in pthread barriers
1856 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1857 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1859 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1860 [14912] libc: Rename non-installed bits/*.h headers
1861 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1862 [15367] math: Let gcc use __builtin_isinf
1863 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1864 [15421] math: lgamma wrongly sets signgam for ISO C
1865 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1866 FE_INVALID with argument out of range
1867 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1869 [15786] dynamic-link: ifunc resolver functions can smash function
1871 [15918] math: Unnecessary check for equality in hypotf()
1872 [16061] localedata: Review / update transliteration data
1873 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1875 [16141] time: strptime %z offset restriction
1876 [16171] math: drem should be alias of remainder
1877 [16296] math: fegetround is pure?
1878 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1879 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1881 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1882 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1884 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1885 "Exception "Inexact" set" on ppc32
1886 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1887 [16517] math: Missing underflow exception from tanf/tan/tanl
1888 [16519] math: Missing underflow exception from sinhf
1889 [16520] math: Missing underflow exception from tanhf
1890 [16521] math: Missing underflow exception from exp2
1891 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1893 [16734] stdio: fopen calls mmap to allocate its buffer
1894 [16961] math: nan function incorrect handling of bad sequences
1895 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1896 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1897 [16985] locale: localedef: confusing error message when opening output
1899 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1900 [17197] locale: Redundant shift character in iconv conversion output at
1902 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1903 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1904 [17250] dynamic-link: static linking breaks nss loading
1905 (getaddrinfo/getpwnam/etc...)
1906 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1908 [17441] math: isnan() should use __builtin_isnan() in GCC
1909 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1910 (related to lock elision)
1911 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1912 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1913 [17887] time: strptime should be able to parse "+01:00" style timezones
1914 [17905] libc: catopen() Multiple unbounded stack allocations
1916 [18084] libc: backtrace (..., 0) dumps core on x86
1917 [18086] libc: nice() sets errno to 0 on success
1918 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1919 is too large (CVE-2015-8778)
1920 [18251] dynamic-link: SONAME missing when audit modules provides path
1921 [18265] libc: add attributes for wchar string and memory functions
1922 [18370] math: csqrt missing underflows
1923 [18421] libc: [hppa] read-only segment has dynamic relocations
1924 [18472] libc: Obsolete syscall wrappers should be compat symbols
1925 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1926 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1927 [18525] localedata: Remove locale timezone information
1928 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1929 [18568] localedata: Update locale data to Unicode 8.0
1930 [18589] locale: sort-test.sh fails at random
1931 [18595] math: ctan, ctanh missing underflows
1932 [18604] libc: assert macro-expands its argument
1933 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1934 contains a vector instruction exception.
1935 [18611] math: j1, jn missing errno setting on underflow
1936 [18618] localedata: sync Chechen locale definitions with other *_RU
1938 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1939 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1940 [18665] network: In send_dg, the recvfrom function is NOT always using the
1941 buffer size of a newly created buffer (CVE-2015-7547)
1942 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1944 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1946 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1947 [18699] math: tilegx cproj() for various complex infinities does not yield
1949 [18724] libc: Harden put*ent functions against data injection
1950 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1951 [18755] build: build errors with -DNDEBUG
1952 [18757] stdio: fmemopen fails to set errno on failure
1953 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1955 [18781] libc: openat64 lacks O_LARGEFILE
1956 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1957 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1958 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1959 [18790] math: [ldbl-128ibm] tanhl inaccurate
1960 [18795] libc: stpncpy fortification misses buffer lengths that are
1961 statically too large
1962 [18796] build: build fails for --disable-mathvec
1963 [18803] math: hypot missing underflows
1964 [18820] stdio: fmemopen may leak memory on failure
1965 [18823] math: csqrt spurious underflows
1966 [18824] math: fma spurious underflows
1967 [18825] math: pow missing underflows
1968 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1969 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1970 [18870] build: sem_open.c fails to compile with missing symbol
1972 [18872] stdio: Fix memory leak in printf_positional
1973 [18873] libc: posix_fallocate overflow check ineffective
1974 [18875] math: Excess precision leads incorrect libm
1975 [18877] libc: arm: mmap offset regression
1976 [18887] libc: memory corruption when using getmntent on blank lines
1977 [18918] localedata: hu_HU: change time to HH:MM:SS format
1978 [18921] libc: Regression: extraneous stat() and fstat() performed by
1980 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1981 binaries (CVE-2015-8777)
1982 [18951] math: tgamma missing underflows
1983 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1985 [18953] localedata: lt_LT: change currency symbol to the euro
1986 [18956] math: powf inaccuracy
1987 [18961] math: [i386] exp missing underflows
1988 [18966] math: [i386] exp10 missing underflows
1989 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1990 [18969] build: multiple string test failures due to missing locale
1992 [18970] libc: Reference of pthread_setcancelstate in libc.a
1993 [18977] math: float / long double Bessel functions not in XSI POSIX
1994 [18980] math: i386 libm functions return with excess range and precision
1995 [18981] math: i386 scalb*, ldexp return with excess range and precision
1996 [18982] stdio: va_list and vprintf
1997 [18985] time: Passing out of range data to strftime() causes a segfault
1999 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2000 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2001 [19012] locale: iconv_open leaks memory on error path
2002 [19016] math: clog, clog10 inaccuracy
2003 [19018] nptl: Mangle function pointers in tls_dtor_list
2004 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2005 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2006 [19048] malloc: malloc: arena free list can become cyclic, increasing
2008 [19049] math: [powerpc] erfc incorrect zero sign
2009 [19050] math: [powerpc] log* incorrect zero sign
2010 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2011 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2012 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2013 [19074] network: Data race in _res_hconf_reorder_addrs
2014 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2015 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2016 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2018 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2020 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2021 [19086] manual: posix_fallocate64 documented argument order is wrong.
2022 [19088] math: lround, llround missing exceptions close to overflow
2024 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2025 [19095] math: dbl-64 lrint incorrect for 64-bit long
2026 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2027 [19124] dynamic-link: ld.so failed to build with older assmebler
2028 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2029 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2030 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2031 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2032 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2033 pthread_setaffinity_np
2034 [19156] math: [ldbl-128] j0l spurious underflows
2035 [19164] nptl: tst-getcpu fails with many possible CPUs
2036 [19168] math: math/test-ildoubl and math/test-ldouble failure
2037 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2038 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2040 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2041 (FE_NOMASK_ENV) do not clear SSE exceptions
2042 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2043 _int_new_arena/reused_arena
2044 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2045 [19201] math: dbl-64 remainder incorrect sign of zero result
2046 [19205] math: bits/math-finite.h conditions do not match math.h and
2048 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2049 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2050 for C99-based standards
2051 [19212] libc: features.h not -Wundef clean
2052 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2054 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2055 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2056 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2058 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2059 "inexact" exceptions
2060 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2062 [19242] libc: strtol incorrect in Turkish locales
2063 [19243] malloc: reused_arena can pick an arena on the free list, leading
2064 to an assertion failure and reference count corruption
2065 [19253] time: tzset() ineffective when temporary TZ did not include DST
2067 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2068 [19270] math: [hppa] Shared libm missing __isnanl
2069 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2070 MAP_HUGETLB and MAP_STACK defines
2071 [19313] nptl: Wrong __cpu_mask for x32
2072 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2073 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2074 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2075 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2076 [19363] time: x32: times() return value wrongly truncates/sign extends
2078 [19367] dynamic-link: Improve branch prediction on Silvermont
2079 [19369] network: Default domain name not reset by res_ninit when "search"
2080 / "domain" entry is removed from resolv.conf
2081 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2083 [19385] localedata: bg_BG: time separator should be colon, not comma
2084 [19408] libc: linux personality syscall wrapper may erroneously return an
2085 error on 32-bit architectures
2086 [19415] libc: dladdr returns wrong names on hppa
2087 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2088 IBM905, IBM907, and IBM909
2089 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2090 [19443] build: build failures with -DDEBUG
2091 [19451] build: Make check fails on test-double-vlen2
2092 [19462] libc: Glibc failed to build with -Os
2093 [19465] math: Wrong code with -Os
2094 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2096 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2098 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2099 [19486] math: S390: Math tests fail with "Exception Inexact set".
2100 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2101 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2103 [19590] math: Fail to build shared objects that use libmvec.so functions.
2107 * The following bugs are resolved with this release:
2109 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2110 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2111 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2112 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2113 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2114 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2115 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2116 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2117 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2118 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2119 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2120 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2121 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2122 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2123 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2124 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2125 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2126 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2127 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2128 18657, 18676, 18694, 18696, 18887.
2130 * Cache information can be queried via sysconf() function on s390 e.g. with
2131 _SC_LEVEL1_ICACHE_SIZE as argument.
2133 * A buffer overflow in gethostbyname_r and related functions performing DNS
2134 requests has been fixed. If the NSS functions were called with a
2135 misaligned buffer, the buffer length change due to pointer alignment was
2136 not taken into account. This could result in application crashes or,
2137 potentially arbitrary code execution, using crafted, but syntactically
2138 valid DNS responses. (CVE-2015-1781)
2140 * The time zone file parser has been made more robust against crafted time
2141 zone files, avoiding heap buffer overflows related to the processing of
2142 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2143 large time zone data files. Overly long time zone specifiers in the TZ
2144 variable no longer result in stack overflows and crashes.
2146 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2147 for LD and GD on x86 and x86-64, has been implemented. You will need
2148 binutils-2.24 or later to enable this optimization.
2150 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2151 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2152 Hat). These updates cause user visible changes, such as the fix for bug
2155 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2156 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2157 condition in some applications.
2159 * Added vector math library named libmvec with the following vectorized x86_64
2160 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2162 The library can be disabled with --disable-mathvec. Use of the functions is
2163 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2164 Shared library libmvec.so is linked in as needed when using -lm (no need to
2165 specify -lmvec explicitly for not static builds).
2166 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2168 * A new fmemopen implementation has been added with the goal of POSIX
2169 compliance. The new implementation fixes the following long-standing
2170 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2171 old implementation is still present for use be by existing binaries.
2173 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2174 releases. It has been fixed to match 2.19 and older, but binaries built
2175 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2177 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2178 Contributed by Roland McGrath (Google).
2180 * The header <regexp.h> is deprecated, and will be removed in a future
2181 release. Use of this header will trigger a deprecation warning.
2182 Application developers should update their code to use <regex.h> instead.
2184 This header was formerly part of SUS, but was deprecated in 1994 and
2185 removed from the standard in 2001. Also, the glibc implementation
2186 leaks memory. See BZ#18681 for more details.
2190 * The following bugs are resolved with this release:
2192 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2193 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2194 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2195 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2196 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2197 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2198 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2199 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2200 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2201 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2204 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2205 too little memory for the to-be-scanned arguments and overflow the
2206 allocated buffer. The implementation now correctly computes the required
2207 buffer size when using malloc, and switches to malloc from alloca as
2210 * A new semaphore algorithm has been implemented in generic C code for all
2211 machines. Previous custom assembly implementations of semaphore were
2212 difficult to reason about or ensure that they were safe. The new version
2213 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2214 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2215 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2217 * Port to Altera Nios II has been contributed by Mentor Graphics.
2219 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2220 implementations for powerpc64/powerpc64le.
2221 Implemented by Adhemerval Zanella (IBM).
2223 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2224 and powerpc64le. This may improve lock scaling of existing programs on
2225 HTM capable systems. The lock elision code is only enabled with
2226 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2227 powerpc will issue a transaction abort on every syscall to avoid side
2228 effects being visible outside transactions.
2230 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2231 AArch64. Contributed by ARM Ltd.
2233 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2235 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2236 under certain input conditions resulting in the execution of a shell for
2237 command substitution when the application did not request it. The
2238 implementation now checks WRDE_NOCMD immediately before executing the
2239 shell and returns the error WRDE_CMDSUB as expected.
2241 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2242 processing format strings with a large number of format specifiers.
2244 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2245 infinite loop if the DNS response contained a PTR record of an unexpected
2248 * The minimum GCC version that can be used to build this version of the GNU
2249 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2250 still be used to compile programs using the GNU C Library.
2252 * The GNU C Library is now built with -Werror by default. This can be
2253 disabled by configuring with --disable-werror.
2255 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2257 * The obsolete sigvec function has been removed. This was the original
2258 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2259 programs have been using instead for about 25 years. Of course, ABI
2260 compatibility for old binaries using sigvec remains intact.
2262 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2263 with newer versions of bison.
2265 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2266 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2267 registers overlay two consecutive single-precision registers. MIPS32R2
2268 introduced a new FPU mode (FR=1) where double-precision registers extend the
2269 corresponding single-precision registers which is incompatible with the
2270 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2271 require the use of FR=1 making a transition necessary. New o32 ABI
2272 extensions enable users to migrate over time from the original o32 ABI
2273 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2274 tracks the ABI of any loaded object and verifies that new objects are
2275 compatible. Mode transitions will also be requested as required and
2276 unsupportable objects will be rejected. The ABI checks include both soft and
2277 hard float ABIs for o32, n32 and n64.
2279 GCC 5 with GNU binutils 2.25 onwards:
2280 It is strongly recommended that all o32 system libraries are built using the
2281 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2282 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2283 compiler using --with-fp-32=xx to set this by default.
2287 * The following bugs are resolved with this release:
2289 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2290 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2291 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2292 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2293 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2294 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2295 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2296 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2297 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2298 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2299 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2300 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2301 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2302 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2303 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2305 * Reverted change of ABI data structures for s390 and s390x:
2306 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2307 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2308 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2309 are preserved pointing straight to the same implementation as the old ones.
2310 Given that, new callers will simply provide a too-big buffer to these
2311 functions. Any applications/libraries out there that embed jmp_buf or
2312 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2313 against 2.19 headers will have to rebuilt again. This is necessary in any
2314 case to revert the breakage in their ABI caused by the glibc change.
2316 * Support for file description locks is added to systems running the
2317 Linux kernel. The standard file locking interfaces are extended to
2318 operate on file descriptions, not file descriptors, via the use of
2319 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2320 are associated with an open file instead of a process.
2322 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2324 * The minimum Linux kernel version that this version of the GNU C Library
2325 can be used with is 2.6.32.
2327 * Running the testsuite no longer terminates as soon as a test fails.
2328 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2329 with PASS or FAIL lines for individual tests. A summary of the results is
2330 printed, including a list of failing lists, and "make check" exits with
2331 error status if there were any unexpected failures. "make check
2332 stop-on-test-failure=y" may be used to keep the old behavior.
2334 * The am33 port, which had not worked for several years, has been removed
2337 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2338 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2339 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2340 conflicted with POSIX (support for which was removed in 2.19), the
2341 interfaces those macros enabled remain available when compiling with
2342 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2343 test macros defined.
2345 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2347 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2348 This may improve lock scaling of existing programs on TX capable systems.
2349 The lock elision code is only built with --enable-lock-elision=yes and
2350 then requires a GCC version supporting the TX builtins. With lock elision
2351 default mutexes are elided via __builtin_tbegin, if the cpu supports
2352 transactions. By default lock elision is not enabled and the elision code
2355 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2356 copy the path argument. This allowed programs to cause posix_spawn to
2357 deference a dangling pointer, or use an unexpected pathname argument if
2358 the string was modified after the posix_spawn_file_actions_addopen
2361 * All supported architectures now use the main glibc sysdeps directory
2362 instead of some being in a separate "ports" directory (which was
2363 distributed separately before glibc 2.17).
2365 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2366 On configurations that support it (all Linux configurations), it's now
2367 used regardless of the --enable-add-ons switch to configure. It is no
2368 longer possible to build such configurations without pthreads support.
2370 * Locale names, including those obtained from environment variables (LANG
2371 and the LC_* variables), are more tightly checked for proper syntax.
2372 setlocale will now fail (with EINVAL) for locale names that are overly
2373 long, contain slashes without starting with a slash, or contain ".." path
2374 components. (CVE-2014-0475) Previously, some valid locale names were
2375 silently replaced with the "C" locale when running in AT_SECURE mode
2376 (e.g., in a SUID program). This is no longer necessary because of the
2379 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2380 with application code using Intel MPX instructions. (With all previous
2381 versions, the MPX register state could be clobbered when making calls
2382 into or out of a shared library.) Note that while the new dynamic
2383 linker is compatible with all known x86 hardware whether or not it
2384 supports Intel MPX, some x86 instruction-set emulators might fail to
2385 handle the new instruction encodings. This is known to affect Valgrind
2386 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2387 release), and might affect other tools that do instruction emulation.
2389 * Support for loadable gconv transliteration modules has been removed.
2390 The support for transliteration modules has been non-functional for
2391 over a decade, and the removal is prompted by security defects. The
2392 normal gconv conversion modules are still supported. Transliteration
2393 with //TRANSLIT is still possible, and the //IGNORE specifier
2394 continues to be supported. (CVE-2014-5119)
2396 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2397 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2398 resulting a denial-of-service security vulnerability in applications which
2399 use functions related to iconv. (CVE-2014-6040)
2403 * The following bugs are resolved with this release:
2405 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2406 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2407 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2408 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2409 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2410 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2411 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2412 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2413 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2414 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2415 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2416 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2417 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2418 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2419 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2420 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2421 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2422 16431, 16453, 16474, 16506, 16510, 16529
2424 * Slovenian translations for glibc messages have been contributed by the
2425 Translation Project's Slovenian team of translators.
2427 * The public headers no longer use __unused nor __block. This change is to
2428 support compiling programs that are derived from BSD sources and use
2429 __unused internally, and to support compiling with Clang's -fblock
2430 extension which uses __block.
2432 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2433 large collation sequences to optimize multiple passes. This cache
2434 computation may overflow for large collation sequences and may cause a
2435 stack or buffer overflow. This is now fixed to use a slower algorithm
2436 which does not use a cache if there is an integer overflow.
2438 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2439 rules for large collation sequences to optimize multiple passes and falls
2440 back to alloca if malloc fails, resulting in a possible stack overflow.
2441 The implementation now falls back to an uncached collation sequence lookup
2444 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2445 initialized for static applications resulting in the security feature
2446 being disabled. The pointer guard is now correctly initialized to a
2447 random value for static applications. Existing static applications need
2448 to be recompiled to take advantage of the fix (bug 15754).
2450 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2451 to the d_name member of struct dirent, or omit the terminating NUL
2452 character. (Bugzilla #14699).
2454 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2455 aligned_alloc functions could allocate too few bytes or corrupt the
2456 heap when passed very large allocation size values (Bugzilla #15855,
2459 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2460 for AF_INET6 has been fixed (Bugzilla #16072).
2462 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2463 pap_CW, quz_PE, the_NP.
2465 * Substantially revised locales: gd_GB, ht_HT
2467 * The LC_ADDRESS field was updated to support country_car for almost all
2470 * ISO 1427 definitions were updated.
2472 * ISO 3166 definitions were updated.
2474 * The localedef utility now supports --big-endian and --little-endian
2475 command-line options to generate locales for a different system from that
2476 for which the C library was built.
2478 * Binary locale files now only depend on the endianness of the system for
2479 which they are generated and not on other properties of that system. As a
2480 consequence, binary files generated with new localedef may be incompatible
2481 with old versions of the GNU C Library, and binary files generated with
2482 old localedef may be incompatible with this version of the GNU C Library,
2483 in the following circumstances:
2485 + Locale files may be incompatible on m68k systems.
2487 + Locale archive files (but not separate files for individual locales) may
2488 be incompatible on systems where plain "char" is signed.
2490 * The configure option --disable-versioning has been removed. Builds with
2491 --disable-versioning had not worked for several years.
2493 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2494 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2496 * SystemTap probes for malloc have been introduced.
2498 * SystemTap probes for slow multiple precision fallback paths of
2499 transcendental functions have been introduced.
2501 * Support for powerpc64le has been added.
2503 * The soft-float powerpc port now supports e500 processors.
2505 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2507 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2508 set of header declarations that are enabled by default, even when other
2509 feature test macros or compiler options such as -std=c99 would otherwise
2510 disable some of those declarations.
2512 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2513 conflict with POSIX. The libbsd-compat library (which was a dummy library
2514 that did nothing) has also been removed.
2516 * Preliminary documentation about Multi-Thread, Async-Signal and
2517 Async-Cancel Safety has been added.
2519 * Change of ABI data structures for s390 and s390x:
2520 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2521 allow for future hardware extensions. All library functions that accept or
2522 return these structures were versioned in order to provide backwards
2523 compatibility with existing code. However, not all cases can be handled
2524 with interface versioning. If an application embeds either structure into
2525 another structure and that structure is passed to another compilation unit
2526 which expects the newer larger-sized object then any access to the new
2527 fields will result in undefined behaviour. Similarly any access to fields
2528 that were moved by the enlarging of the structures will cause undefined
2529 behaviour. To fix the undefined behaviour all objects that are part of the
2530 final in-memory image must agree on the size of structures, and this may
2531 require recompilation.
2535 * The following bugs are resolved with this release:
2537 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2538 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2539 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2540 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2541 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2542 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2543 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2544 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2545 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2546 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2547 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2548 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2549 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2552 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2553 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2554 Distributions can re-enable building and using pt_chown via the new configure
2555 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2556 considerable security risks and should only be used if the distribution
2557 understands and accepts the risks.
2559 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2562 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2563 fixed (Bugzilla #15330).
2565 * Add support for calling C++11 thread_local object destructors on thread
2566 and program exit. This needs compiler support for offloading C++11
2567 destructor calls to glibc.
2569 * Improved worst case performance of libm functions with double inputs and
2572 * Support for priority inherited mutexes in pthread condition variables on
2573 non-x86 architectures.
2575 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2577 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2579 * Optimized string functions for ARM. Implemented by Will Newton and
2582 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2584 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2587 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2588 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2590 * Added a benchmark framework to track performance of functions in glibc.
2592 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2593 It is based on draft TS 18661 and currently enabled as a GNU extension.
2595 * On Linux, the clock function now uses the clock_gettime system call
2596 for improved precision, rather than old times system call.
2598 * Added support for version-3 tz format files. This is needed when using
2599 the tz database release 2013e or later, and affects a few unusual cases --
2600 currently only TZ='America/Godthab' for time stamps after 2037.
2602 * Added new API functions pthread_getattr_default_np and
2603 pthread_setattr_default_np to get and set the default pthread
2604 attributes of a process.
2606 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2607 This may improve lock scaling of existing programs on TSX capable systems.
2608 When the --enable-lock-elision=yes parameter is specified at configure
2609 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2612 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2613 availability). Implemented by Ryan S. Arnold.
2615 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2619 * The following bugs are resolved with this release:
2621 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2622 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2623 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2624 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2625 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2626 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2627 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2628 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2629 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2630 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2631 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2632 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2633 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2634 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2635 14879, 14889, 14893, 14898, 14914.
2637 * Optimization of memcpy for MIPS.
2639 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2640 EMFILE has been fixed (Bugzilla #14889).
2642 * The add-on ports collection is now distributed in the "ports" subdirectory
2643 of the main GNU C Library distribution, rather than separately.
2645 * Port to ARM AArch64 contributed by Linaro.
2647 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2648 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2650 Implemented by Andreas Krebbel.
2652 * The new function secure_getenv allows secure access to the environment,
2653 returning NULL if running in a SUID/SGID process. This function replaces
2654 the internal function __secure_getenv.
2656 * SystemTap static probes have been added into the dynamic linker.
2657 Implemented by Gary Benson.
2659 * Optimizations of string functions strstr, strcasestr and memmem.
2660 Implemented by Maxim Kuvyrkov.
2662 * The minimum Linux kernel version that this version of the GNU C Library
2663 can be used with is 2.6.16.
2665 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2666 powerpc POWER7. Implemented by Will Schmidt.
2668 * New configure option --disable-nscd builds the C library such that it
2669 never attempts to contact the Name Service Caching Daemon (nscd).
2670 New configure option --disable-build-nscd avoids building nscd itself;
2671 this is the default if --disable-nscd is used.
2673 * Improved support for cross-compilation, including cross-testing and
2674 bootstrap builds without a previously built glibc.
2676 * Several testsuite tests are now able to test multiple IFUNC variants of an
2677 interface, rather than just testing the one that would be chooen by
2680 * New configure options --with-bugurl and --with-pkgversion, for
2681 distributors to use to embed their bug-reporting and package version
2682 information in --help and --version output.
2684 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2685 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2686 allows creation of chroots without the procfs mounted on /proc.
2688 * The `crypt' function now fails if passed salt bytes that violate the
2689 specification for those values. On Linux, the `crypt' function will
2690 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2691 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2692 when the mode is enabled.
2694 * The `clock_*' suite of functions (declared in <time.h>) is now available
2695 directly in the main C library. Previously it was necessary to link with
2696 -lrt to use these functions. This change has the effect that a
2697 single-threaded program that uses a function such as `clock_gettime' (and
2698 is not linked with -lrt) will no longer implicitly load the pthreads
2699 library at runtime and so will not suffer the overheads associated with
2700 multi-thread support in other code such as the C++ runtime library.
2702 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2707 * The following bugs are resolved with this release:
2709 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2710 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2711 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2712 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2713 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2714 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2715 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2716 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2717 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2718 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2719 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2720 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2721 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2722 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2723 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2724 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2725 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2726 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2727 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2728 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2729 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2730 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2733 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2734 configuring glibc with:
2735 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2736 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2737 Implemented by H.J. Lu.
2741 + define static_assert
2743 + do not declare gets
2745 + declare at_quick_exit and quick_exit also for ISO C11
2747 + aligned_alloc. NB: The code is deliberately allows the size parameter
2748 to not be a multiple of the alignment. This is a moronic requirement
2749 in the standard but it is only a requirement on the caller, not the
2752 + timespec_get added
2754 + uchar.h support added
2756 + CMPLX, CMPLXF, CMPLXL added
2758 Implemented by Ulrich Drepper.
2760 * Support for the IA-64 has been moved to ports.
2762 * Remove support for anything but ELF binary format
2764 * Checking versions of poll, ppoll added.
2765 Implemented by Ulrich Drepper.
2767 * More generic and 64-bit performance optimizations to math functions.
2768 Implemented by Ulrich Drepper.
2770 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2771 headers and functions available at compile time as they were before
2772 version 2.14. This option will be removed at some time in the future
2773 after the TI-RPC library becomes fully sufficient for the needs of
2774 existing applications.
2776 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2777 Note that glibc is not expected to work with any Linux kernel version
2780 * New header <sys/auxv.h> and function getauxval allowing easy access to
2781 the AT_* key-value pairs passed from the Linux kernel. The header also
2782 defines the HWCAP_* bits associated with the AT_HWCAP key.
2784 * New locales: mag_IN
2786 * New configure option --enable-systemtap builds SystemTap static probes
2787 into libc for setjmp and longjmp and into libpthread for various operations.
2788 So far the setjmp/longjmp probes and some of the libpthread probes are
2789 provided only for i*86 and x86_64.
2790 Implemented by Roland McGrath and Rayson Ho.
2792 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2794 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2797 * More optimized functions for SPARC. Implemented by David S. Miller.
2799 * Improved support for cross-compilation, especially bootstrap builds
2800 without a previously built glibc.
2802 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2803 Chris Metcalf from Tilera.
2805 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2806 now supported for ARM processors.
2808 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2809 as the name of the dynamic linker, to distinguish it from the
2810 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2812 * Support for CRIS has been removed from ports.
2814 * A new class of installed header has been documented for low-level
2815 platform-specific functionality. PowerPC added the first instance with a
2816 function to provide time base register access. Contributed by Tulio
2817 Magno Quites Machado Filho.
2819 * ix86 configurations now install header files that are consistent with
2820 what x86-64 configurations install. These same header files can be used
2821 for -m32, -m64, or -mx32 builds.
2822 Contributed by H.J. Lu.
2824 * Math library bug fixes. A thorough audit of all open math library bugs was
2825 conducted by Joseph Myers. Significant progress was made on many math
2826 library bugs resulting in more accurate exceptions and function results.
2827 Many thanks to all those that contributed including Andreas Jaeger for his
2828 patch review and work on the x87 trigonometric instruction issues.
2830 * Timezone data is no longer installed. Timezone-related binaries and scripts
2831 will continue to be installed. Users should obtain their timezone data from
2832 their distribution provider or from the tzdata package at
2833 <ftp://munnari.oz.au/pub/>.
2837 * The following bugs are resolved with this release:
2839 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2840 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2841 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2842 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2843 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2844 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2845 13484, 13506, 13515, 13523, 13524, 13538, 13540
2847 * New program pldd to list loaded object of a process
2848 Implemented by Ulrich Drepper.
2850 * Add nss_db support back to glibc. No more dependency on Berkeley db
2851 and support for initgroups lookups.
2852 Implemented by Ulrich Drepper.
2854 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2855 Contributed by HJ Lu.
2857 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2858 Contributed by HJ Lu.
2860 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2861 on x86-32 and x86-64.
2862 Contributed by Liubov Dmitrieva.
2864 * Optimized strchr and strrchr for SSE on x86-32.
2865 Contributed by Liubov Dmitrieva.
2867 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2868 for x86-64 and x86-32.
2869 Contributed by Liubov Dmitrieva.
2871 * New interfaces: scandirat, scandirat64
2872 Implemented by Ulrich Drepper.
2874 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2875 Implemented by Ulrich Drepper.
2877 * nscd now also caches the netgroup database.
2878 Implemented by Ulrich Drepper.
2880 * Integrate libm with gcc's -ffinite-math-only option.
2881 Implemented by Ulrich Drepper.
2883 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2884 to math functions. Implemented by Ulrich Drepper.
2886 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2887 Implemented by Ulrich Drepper.
2889 * New Linux interfaces: process_vm_readv, process_vm_writev
2891 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2892 Implemented by Ulrich Drepper.
2894 * Optimized nearbyint and strcasecmp for PPC.
2895 Implemented by Adhemerval Zanella.
2897 * New locales: bho_IN, unm_US, es_CU, ta_LK
2901 * The following bugs are resolved with this release:
2903 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2904 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2905 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2906 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2907 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2908 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2909 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2910 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2911 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2912 12795, 12811, 12813, 12814, 12841
2914 * The RPC implementation in libc is obsoleted. Old programs keep working
2915 but new programs cannot be linked with the routines in libc anymore.
2916 Programs in need of RPC functionality must be linked against TI-RPC.
2917 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2919 Visible changes of this change include (obviously) the inability to link
2920 programs using RPC functions without referencing the TI-RPC library and the
2921 removal of the RPC headers from the glibc headers.
2922 Implemented by Ulrich Drepper.
2924 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2925 syncfs, setns, sendmmsg
2927 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2928 yue_HK, lij_IT, mhr_RU
2930 * New encodings: CP770, CP771, CP772, CP773, CP774
2932 * New program sotruss to trace calls through PLTs
2933 Implemented by Ulrich Drepper.
2935 * The malloc hook implementation is marked deprecated and will be removed
2936 from the default implementation in the next version. The design never
2937 worked ever since the introduction of threads. Even programs which do
2938 not create threads themselves can use multiple threads created internally.
2942 * The following bugs are resolved with this release:
2944 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2945 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2946 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2947 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2950 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2952 * POWER7 optimizations: memset, memcmp, strncmp
2954 * New optimized string functions for x86-64: strnlen (SSE2),
2955 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2956 Implemented by Ulrich Drepper.
2960 * The following bugs are resolved with this release:
2962 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2963 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2964 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2965 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2966 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2967 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2968 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2969 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2971 * New interfaces: pthread_getname_np, pthread_setname_np
2973 * New Linux interface: recvmmsg
2975 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2977 * The dynamic linker now recognizes supported ABI versions from the
2978 EI_ABIVERSION field in the ELF header.
2979 Implemented by Ulrich Drepper.
2981 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2982 will not be used to fill in password fields in the passwd.byname replies.
2983 Instead it is used to synthesize the shadow.byname table, should it be
2984 missing. This is a useful mode in some installations involving Solaris.
2985 Implemented by Ulrich Drepper.
2987 * New locales: kok_IN, sq_MK, cv_RU
2991 * The following bugs are resolved with this release:
2993 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2994 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2995 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2996 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2997 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2998 10789, 10817, 10823, 10840, 10847
3000 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3002 Implemented by Ulrich Drepper.
3004 * Checking version of longjmp added that fails if an uninitialized stack
3005 frame would be created. Implemented by Ulrich Drepper.
3007 * STT_GNU_IFUNC is now supported in static executables.
3008 Implemented by H.J. Lu.
3010 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3012 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3013 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3014 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3015 Contributed by H.J. Lu.
3017 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3018 Implemented by Ulrich Drepper.
3020 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3022 Contributed by H.J. Lu.
3024 * Support for fma instruction in AVX on x86-64.
3025 Implemented by H.J. Lu and Ulrich Drepper.
3027 * AVX support in x86-64 auditing support in ld.so.
3028 Implemented by H.J. Lu.
3030 * STB_GNU_UNIQUE support added.
3031 Implemented by Ulrich Drepper.
3033 * Implement second fallback mode for optimized DNS lookup for even more
3034 broken environments. If two requests from the same port are not
3035 handled correctly close the socket and open a new one before sending
3036 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3037 can be used to select this mode right away, instead of rediscovering the
3038 necessity is every process again.
3039 Implemented by Ulrich Drepper.
3041 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3042 Implemented by Adam Tkac.
3044 * Optimized iconv conversions for S390x.
3045 Implemented by Andreas Krebbel.
3047 * Using condvars with PI mutexes is now more efficient due to kernel
3048 support for requeueing to PI futexes. NPTL support added for x86-64.
3049 Implemented by Ulrich Drepper.
3051 * New locales: ps_AF, my_MM
3055 * The following bugs are resolved with this release:
3057 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3058 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3059 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3060 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3061 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3063 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3064 Implemented by Ulrich Drepper.
3066 * New Linux interfaces: accept4, fallocate, fallocate64.
3067 Implemented by Ulrich Drepper.
3069 * Correct declarations of string function when used in C++ code. This
3070 could lead to compile errors for invalid C++ code.
3072 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3075 * New POSIX 2008 interface: psiginfo
3076 Implemented by Ulrich Drepper.
3078 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3079 Implemented by Ulrich Drepper.
3081 * Support for selecting between multiple function definitions at runtime
3082 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3084 * The libcrypt library can now use the hash function implementations in
3085 NSS. Implemented by Ulrich Drepper.
3087 * The malloc implementation can be compiled to be less memory efficient
3088 but higher performing in multi-threaded programs.
3089 Implemented by Ulrich Drepper.
3091 * New locales: nan_TW@latin, ks_IN
3093 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3094 Implemented by Ulrich Drepper.
3096 * Extended printf hook support. It is possible to use user-defined types
3097 and extend existing format specifiers.
3098 Implemented by Ulrich Drepper.
3100 * Handling for group shadow files has been added.
3101 Implemented by Ulrich Drepper.
3103 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3104 servers (the case, e.g., for some people using the built-in DNS
3105 server in ADSL modems/routers). There is a once-per-process timeout
3106 in case of a broken server. To avoid it, users can run nscd or put
3107 'options single-request' in /etc/resolv.conf.
3108 Implemented by Ulrich Drepper.
3112 * The following bugs are resolved with this release:
3114 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3115 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3116 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3117 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3118 6968, 6974, 6980, 6995, 7008, 7009, 7029
3120 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3121 up at the same time. Implemented by Ulrich Drepper.
3123 * TLS descriptors for LD and GD on x86 and x86-64.
3124 Implemented by Alexandre Oliva.
3126 * getaddrinfo now handles DCCP and UDPlite.
3127 Implemented by Ulrich Drepper.
3129 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3130 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3131 Implemented by Ulrich Drepper.
3133 * New implementation of memmem, strstr, and strcasestr which is O(n).
3134 Implemented by Eric Blake.
3136 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3138 * Implement "e" option for popen to open file descriptor with the
3139 close-on-exec flag set. Implemented by Ulrich Drepper.
3141 * Implement "b" mode for fmemopen. In this mode writes writes don't
3142 implicitly add a NUL byte and seeks from the end of the buffer really
3143 use the buffer end, not the string length as the basis.
3144 Implemented by Ulrich Drepper.
3146 * Many functions, exported and internal, now atomically set the close-on-exec
3147 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3149 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3151 Implemented by Pravin Satpute.
3153 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3157 * The following bugs are resolved with this release:
3159 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3160 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3161 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3162 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3163 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3164 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3165 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3167 * New locales: bo_CN, bo_IN, shs_CA.
3169 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3171 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3172 Implemented by Pravin Satpute.
3174 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3176 * nscd caches now all timeouts for DNS entries
3177 Implemented by Ulrich Drepper.
3179 * nscd is more efficient and wakes up less often.
3180 Implemented by Ulrich Drepper.
3182 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3183 vdprintf, and obstack_vprintf.
3184 Implemented by Jakub Jelinek.
3186 * Faster memset for x86-64.
3187 Implemented by Harsha Jagasia and H.J. Lu.
3189 * Faster memcpy on x86.
3190 Implemented by Ulrich Drepper.
3192 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3193 Implemented by Ulrich Drepper.
3195 * Faster sqrt and sqrtf implemention for some PPC variants.
3196 Implemented by Steven Munroe.
3200 * The following bugs are resolved with this release:
3202 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3203 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3204 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3205 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3206 5071, 5103, 5104, 5112, 5113, 5184, 5186
3208 * More checking functions: fread, fread_unlocked, open*, mq_open.
3209 Implemented by Jakub Jelinek and Ulrich Drepper.
3211 * Extend fortification to C++. Implemented by Jakub Jelinek.
3213 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3214 by not recognizing 'a' as a modifier when those specs are requested.
3215 Implemented by Jakub Jelinek.
3217 * PPC optimizations to math and string functions.
3218 Implemented by Steven Munroe.
3220 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3221 options to be passed. Implemented by Ulrich Drepper.
3223 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3225 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3226 Implemented by Ulrich Drepper.
3228 * Handle private futexes in the NPTL implementation.
3229 Implemented by Jakub Jelinek and Ulrich Drepper.
3231 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3232 Implemented by Roland McGrath and Ulrich Drepper.
3234 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3236 * SHA-256 and SHA-512 based password encryption.
3237 Implemented by Ulrich Drepper.
3239 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3240 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3243 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3244 Implemented by Ulrich Drepper.
3248 * The following bugs are resolved with this release:
3250 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3251 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3252 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3253 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3254 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3255 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3256 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3259 * New Linux interfaces: epoll_pwait, sched_getcpu.
3261 * New generic interfaces: strerror_l.
3263 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3267 * The following bugs are resolved with this release:
3269 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3270 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3271 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3272 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3273 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3274 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3275 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3276 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3277 3225, 3251, 3252, 3253, 3273
3279 * For Linux, the sorting of addresses returned by getaddrinfo now also
3280 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3281 Implemented by Ulrich Drepper.
3283 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3284 Implemented by Ulrich Drepper.
3286 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3288 * New iconv module for MIK. Contributed by Alexander Shopov.
3290 * For sites with broken group and/or passwd database, the auto-propagate
3291 option of nscd can prevent creating ID lookup entries from the results
3292 of a name lookup and vice versa. This usually is no problem but some
3293 site might have problems with the default behavior.
3294 Implemented by Ulrich Drepper.
3296 * Iterating over entire database in NIS can be slow. With the
3297 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3298 to trade time for memory. The entire database will be read at once.
3299 Implemented by Ulrich Drepper.
3301 * The interfaces introduced in RFC 3542 have been implemented by
3304 * Support for the new ELF hash table format was added by Ulrich Drepper.
3306 * Support for priority inheritance mutexes added by Jakub Jelinek and
3309 * Support for priority protected mutexes added by Jakub Jelinek.
3311 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3312 or_IN, csb_PL, fy_NL, sr_ME.
3316 * More overflow detection functions.
3318 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3319 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3321 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3322 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3323 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3324 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3325 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3326 by Masahide Washizawa.
3328 * It is now possible to install audit libraries for the dynamic linker using
3329 LD_AUDIT. Implemented by Ulrich Drepper.
3331 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3332 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3333 For a libc and libpthread that works well on Linux 2.4 kernels, we
3334 recommend using the stable 2.3 branch.
3336 * The new function open_wmemstream acts like open_memstream,
3337 but uses a wchar_t wide character buffer.
3339 * The new function ppoll is an improved version of poll, similar to pselect.
3341 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3342 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3343 renameat, unlinkat, linkat, symlinkat, readlinkat.
3345 * New Linux kernel system calls: unshare,
3346 inotify_init, inotify_add_watch, inotify_rm_watch.
3348 * The euidaccess function is now also known by the alias eaccess,
3349 for compatibility with some other systems.
3351 * Timezone data updated to 2006b version.
3355 * The following bugs are resolved with this release:
3357 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3358 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3359 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3360 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3361 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3362 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3364 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3366 * As of this release, GCC 4 can be used to compile the C Library.
3368 * Timezone data updated to 2005m version.
3372 * The following bugs are resolved with this release:
3374 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3375 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3376 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3377 777, 787, 821, 822, 823, 825
3379 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3383 * Support for RFC 3678. Real implementations exist only for Linux so far.
3384 Implemented by Ulrich Drepper.
3386 * nscd can now cache entries persistently. Expiring entries are reloaded.
3387 For speedups the cache can be shared in memory with client processes.
3388 Implemented by Ulrich Drepper.
3390 * nscd can now perform SELinux checks.
3391 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3393 * getaddrinfo queries are now cached. Canonical name lookup is performed
3395 Implemented by Ulrich Drepper.
3397 * The nothrow function attribute is used when headers are used by gcc when
3398 compiling C code. This can avoid significant amounts of exception
3401 * The malloc functions perform more error checking and are stricter when
3402 it comes to reacting on errors. The default action is to terminate
3403 the process after showing an error message. Implemented by Ulrich Drepper.
3405 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3406 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3408 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3409 namespaces using the new function dlmopen(). This feature is of course,
3410 like most other dynamic loading functionality, not available in statically
3411 linked applications. Implemented by Ulrich Drepper.
3413 * Low-overhead boundary checking variants of string and some stdio functions
3414 were added. These are to be used in conjunction with a gcc patch by
3415 Jakub Jelinek which adds calls to these functions if possible.
3416 Implemented by Jakub Jelinek and Ulrich Drepper.
3418 * Old code for several operating systems and machine architectures that
3419 have not been in working condition in a long time have been removed from
3420 the main source tree maintained by the GNU C Library's maintainers.
3421 These files are now reside in the separate `ports' source module
3422 that is usable as an add-on when building the library.
3426 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3427 interrogate the dynamic linker, compatible with the Solaris interface.
3429 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3430 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3432 * getifaddrs now uses the netlink interface on Linux to get its information.
3433 Implemented by Thorsten Kukuk.
3435 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3436 Implemented by Ulrich Drepper.
3438 * support for non-executable stacks on x86 has been added. Changes mostly
3441 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3444 * getaddrinfo now performs destination address selection according to
3449 * Thread-safe interfaces for many functions that access locale data
3450 were added in version 2.3, but these features were omitted from NEWS.
3451 Many functions have variants with an `_l' suffix that take a `locale_t'
3452 object as a parameter rather than consulting the current locale.
3453 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3454 create and maintain `locale_t' objects. Additionally, the new function
3455 `uselocale' sets "the current locale" (as used by functions not so
3456 parameterized) set for an individual thread. These features were added
3457 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3459 * The functions getresuid, getresgid, setresuid, and setresgid, which
3460 have long been available on Linux, are now declared in <unistd.h>
3461 and are now also available on the Hurd.
3463 * ELF thread-local storage support (TLS) now works on x86-64.
3465 * The new dynamic string token $LIB is expanded in shared library names.
3466 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3468 * Aldy Hernandez contributed complete software floating point support for
3469 PowerPC machines with no FPU.
3471 * fexecve is implemented on Linux.
3473 * The `btowc' function should work at least twice as fast due to
3474 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3476 * With approriate thread add-ons cancelable functions are now implemented
3477 in libc.so as well. No need to call the function in libpthread. This
3478 change allowed to finally disable the incorrect and expensive handling
3479 of weak definition in ld.so.
3481 * Yet more PLT entries in libc.so have been removed. We finally arrived
3482 at the bare minimum. Startup times improved appropriately.
3484 * Support for the new Linux/x86 system call interface was added. The
3485 AT_SYSINFO auxiliary vector entry is recognized and handled.
3489 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3492 * iconv (the program and the interface) now accepts empty names (excluding
3493 options like //TRANSLIT) to mean "use charset of current locale".
3495 * localedef can now transliterate characters in strings which are not in
3496 the provided charmap. The information from the input locale is used.
3498 * Prelinking support was added for ELF targets. This requires additional
3499 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3501 * Read-only stdio streams now use mmap to speed up operation by eliminating
3502 copying and buffer underflows. To use add 'm' to the mode string of
3503 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3505 * The malloc functions were completely rewritten by Wolfram Gloger based
3506 on Doug Lea's malloc-2.7.0.c.
3508 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3509 implementation of regex.
3511 * Bruno Haible upgraded the iconv and locale implementation to support
3514 * Contents of the LC_* and LANG environment variables in the CEN style are
3515 not recognized anymore. It never was used. Change by Ulrich Drepper.
3517 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3518 thread-local storage (TLS) ABI on some platforms.
3519 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3521 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3522 EUC-JISX0213, and TSCII.
3524 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3525 BSD-compatible interface for getting all network interface addresses.
3526 Implementation for IPv4 by Roland McGrath.
3528 * Loading of locale data is faster due to the introduction of a locale
3529 archive. Implemented by Roland McGrath and Ulrich Drepper.
3531 * Startup times are significantly reduced by not using exported functions
3532 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3535 * Steven Munroe contributed a port to PowerPC64/Linux.
3539 * The Hurd now uses the GNU libio implementation of stdio rather than the
3540 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3542 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3543 with the same behavior as the Linux system call of the same name.
3547 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3548 128-bit long double format.
3550 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3551 IBM1161, and IBM1162 charsets.
3553 * Andreas Jaeger contributed a port to x86-64/Linux.
3555 * Peter Bruin contributed a port to PowerPC/Hurd.
3557 * libc functions using I/O streams now can handle wide-oriented streams
3560 * optimizations in the dynamic linker. Binaries created by recent binutils
3561 versions start up quicker due to reduced time spent on relocations.
3563 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3567 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3568 asinh, atanh, j0 for the 128-bit long double format.
3570 * Bruno Haible updated all the code handling Unicode in some form to
3571 support Unicode 3.1.
3573 * Speed of regex for single-byte locales is back to previous levels.
3574 Patch by Isamu Hasegawa.
3576 * Alpha, SPARC, and IA-64 now also using floating stacks.
3578 * Startup time of internationalized applications greatly improved through
3579 iconv cache. Use iconvconfig to generate the cache file.
3580 Contributed by Ulrich Drepper.
3582 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3583 severe performance problems. Patch by David Mosberger.
3585 * The Hurd port got a lot more functionality like AIO, various stdio
3586 extensions, etc. Mainly done by Roland McGrath.
3588 * mtrace can now lookup symbols in shared libraries.
3592 * Intel's IA-64 math library is largely integrated. It provides fast and
3593 accurate implementatations for most basic and standard math functions
3594 in float, double, and long double format.
3596 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3597 and asin for the 96-bit long double format and asin, log, tan for the
3598 128-bit long double format.
3600 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3601 The basic double functions exist today. Contributed by Abraham Ziv
3602 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3603 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3605 * An asynchronous name lookup library was added. The interface is designed
3606 after POSIX AIO. The proposal was circulated beforehand to get comments.
3607 No negative ones came in. Implemented by Ulrich Drepper.
3609 * Port to S390/64bit contributed by Martin Schwidefsky
3610 <schwidefsky@de.ibm.com>.
3612 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3613 of functions for Linux/IA-64.
3615 * The RPC code is now thread safe. Threads can now use the same service
3616 of different services at the same time. Patch by Eric Norum
3617 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3619 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3620 family of functions for Linux/S390.
3622 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3623 of functions for Linux/x86.
3625 * Port to Linux/CRIS contributed by Axis Communications.
3629 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3630 we can now check for namespace violations and missing declarations. The
3631 result is that almost all headers are now Unix-compliant (as defined in
3632 the upcoming XPG6). The negative side is that some programs might need
3633 corrections, too, if they depend on the incorrect form of the headers in
3634 previous versions which defined too many symbols and included too many
3637 * regex now handles multibyte character sets correctly.
3638 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3640 * iconv (the program) does now conform to the upcoming XPG6 and handles
3641 charmaps. Instead of the charset names the path of charmaps can be
3642 provided and the conversion happens based on this data.
3643 Contributed by Ulrich Drepper.
3645 * The locale program now provides detailed information about the installed
3650 only lists the names of the supported locales
3654 provides details such as country, language, and codeset name.
3655 Contributed by Ulrich Drepper.
3659 * The gencat program now parses the input file according to the charset
3660 selected by the LC_CTYPE category. This is important for stateful
3661 character sets. To make generating catalogs easier there is a way
3662 to overwrite the charset selected by the locale: before the first
3663 message or $ quote line the catalog can contain a line like
3665 $ codeset=ISO-8859-2
3667 to select the charset (ISO-8859-2 in this case).
3669 Implemented by Shinya Hanataka and Ulrich Drepper.
3671 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3672 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3673 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3675 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3676 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3678 * The LANGUAGE environment variable is now ignored unless the locale is
3679 changed from the default "C" locale.
3681 * The usual bug fixes.
3685 * Greg McGary added runtime support for bounds checking using gcc's
3686 new -fbounded-pointers option. ix86 target is complete. PowerPC
3689 * Thorsten Kukuk added secure mode support to nscd.
3691 * The Berkeley db libraries have been removed.
3693 Related, the nss_db module is now in a separate package since it
3694 obviously requires a database library being available.
3696 * Wide character I/O streams implemented by Ulrich Drepper.
3698 * Functions from the extended socket API added by Ulrich Drepper.
3700 * Functions feenableexcept and fedisableexcept to control the
3701 behaviour of individual exceptions have been added by Andreas Jaeger.
3703 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3705 * The resolver code has been updated from bind 8.2.3-T5B which supports
3706 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3709 This change could in some situations effect backward compatibility. Since
3710 now `_res' is a thread-local instead of a global variable, modifying it
3711 in one thread does not have any effect in other threads.
3713 The resolver library was also extended to allow IPv6 as the transport
3714 protocol for the requests. This work was done by Stig Venaas.
3716 * Compatibility code for K&R C compilers has been removed from the
3717 header files. A ISO C compiler is needed to use the library
3718 (conforming to either C89 or C99 standard).
3720 * Complete rewrite of the localedef program to support multibyte character
3721 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3722 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3723 Implemented by Ulrich Drepper.
3725 Bruno Haible significantly improved the generation and use of the data
3726 structures for the wide character tables.
3728 * Plural handling in gettext implemented by Ulrich Drepper.
3730 * The utmp daemon has been removed.
3732 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3734 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3737 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3739 * POSIX barriers implemented by Kaz Kylheku.
3741 * POSIX spawn function family implemented by Ulrich Drepper.
3743 * POSIX spinlocks are now available.
3745 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3747 * the configure option --enable-kernel=X.Y.Z allows to strip out
3748 compatibility for kernel versions before X.Y.Z. This is currently only
3749 implemented for Linux.
3751 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3752 field sin6_scope_id. This means that all programs using IPv6 should be
3753 recompiled. Don't expect binary compatibility with previous glibc
3756 * various conversion modules for IBM character sets contributed by
3759 * IA-64 port by Jes Sorensen and HJ Lu.
3773 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3775 * New xdr functions are added; some rpc functions are now 64bit clean.
3777 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3779 * Fixed known incompatibilities with glibc 2.0.
3781 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3783 * Optimized a number of functions (especially the ELF dynamic loader).
3785 * Update timezone data files.
3787 * lots of charmaps corrections
3789 * some new locale definitions and charmaps
3794 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3795 conform to POSIX member type requirements. Symbol versions have been
3796 adjusted as needed within the library, and for direct use by applications,
3797 but there is potential for problems if third-party libraries use
3798 struct timeval as part of their interface. This does not present
3799 a problem for X and other "essential" system libraries.
3801 * An additional locale model to support C++ Standard Library locale
3802 model and probably more was implemented by Ulrich Drepper.
3804 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3807 * Miles Bader provided the `argp' function family to support hierachical
3808 command line argument parsing, layered on top of getopt.
3810 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3812 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3815 * scanf recognizes the %a and %A format for scanning floating point numbers.
3817 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3818 information and interfaces for the available integer types.
3820 * about 130 new math functions were added to implement the ISO C9x math
3823 * the new header <complex.h> contains definitions of the complex math
3824 functions from ISO C 9X.
3826 * the new header <tgmath.h> defines generic macros to use complex or
3827 real valued functions.
3829 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3831 * Andreas Jaeger provided a test suite for the math library.
3833 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3835 * Ulrich Drepper added character set conversion functions (iconv).
3837 * Optimized string functions have been added.
3839 * The localedata addon is now part of glibc.
3841 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3843 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3844 daemon for NSS (nscd).
3846 Missing a better place here are some numbers on improvements. Under
3847 Linux 2.1.125 un-tar-ing the kernel sources takes
3851 using local files 12.19s 6.88s 22.91s
3853 using NIS 13.92s 8.91s 26.34s
3855 using NIS & nscd 10.37s 7.34s 25.30s
3857 using NIS+ 27.57s 30.37s 640.46s
3859 using NIS+ & nscd 10.25s 7.83s 26.51s
3861 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3863 Keep in mind that non-namelookup related operations dominate above times.
3864 It was just a common complain that using NIS+ unpacking the kernel is
3867 [1] The old nscd implementation is not available anymore since it was
3868 distributed with glibc up to version 2.0.98 and thus is now replaced.
3870 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3872 * Mark Kettenis provided a Hesiod NSS module.
3874 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3875 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3877 * rcmd can now handle netgroups (Dick Streefland).
3879 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3882 * Support for the IPv6 protocol has been added to the socket API, as per the
3883 latest draft standards.
3885 * Support for Linux 2.2 has been added.
3887 * Interface changes relative to the latest 2.0.x release:
3888 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3889 addseverity NEW: Unix98
3890 alphasort64 NEW: LFS
3891 argp_err_exit_status NEW: argp, GNU ext
3892 argp_error NEW: argp, GNU ext
3893 argp_failure NEW: argp, GNU ext
3894 argp_help NEW: argp, GNU ext
3895 argp_parse NEW: argp, GNU ext
3896 argp_program_bug_address NEW: argp, GNU ext
3897 argp_program_version NEW: argp, GNU ext
3898 argp_program_version_hook NEW: argp, GNU ext
3899 argp_state_help NEW: argp, GNU ext
3900 argp_usage NEW: argp, GNU ext
3901 authdes_create NEW: Secure RPC
3902 authdes_getucred NEW: Secure RPC
3903 authdes_pk_create NEW: Secure RPC
3904 backtrace NEW: GNU ext.
3905 backtrace_symbols NEW: GNU ext.
3906 backtrace_symbols_fd NEW: GNU ext.
3908 cacosf NEW: ISO C 9x
3909 cacosh NEW: ISO C 9x
3910 cacoshf NEW: ISO C 9x
3911 cacoshl NEW: ISO C 9x
3912 cacosl NEW: ISO C 9x
3919 casinf NEW: ISO C 9x
3920 casinh NEW: ISO C 9x
3921 casinhf NEW: ISO C 9x
3922 casinhl NEW: ISO C 9x
3923 casinl NEW: ISO C 9x
3925 catanf NEW: ISO C 9x
3926 catanh NEW: ISO C 9x
3927 catanhf NEW: ISO C 9x
3928 catanhl NEW: ISO C 9x
3929 catanl NEW: ISO C 9x
3930 cbc_crypt NEW: Secure RPC
3934 ccoshf NEW: ISO C 9x
3935 ccoshl NEW: ISO C 9x
3941 cimagf NEW: ISO C 9x
3942 cimagl NEW: ISO C 9x
3943 clearerr_locked REMOVED
3944 clntunix_create NEW: sunrpc ext
3946 clog10 NEW: ISO C 9x
3947 clog10f NEW: ISO C 9x
3948 clog10l NEW: ISO C 9x
3958 cprojf NEW: ISO C 9x
3959 cprojl NEW: ISO C 9x
3961 crealf NEW: ISO C 9x
3962 creall NEW: ISO C 9x
3967 csinhf NEW: ISO C 9x
3968 csinhl NEW: ISO C 9x
3971 csqrtf NEW: ISO C 9x
3972 csqrtl NEW: ISO C 9x
3976 ctanhf NEW: ISO C 9x
3977 ctanhl NEW: ISO C 9x
3979 des_setparity NEW: Secure RPC
3980 ecb_crypt NEW: Secure RPC
3981 endutxent NEW: Unix98
3983 exp10f NEW: ISO C 9x
3984 exp10l NEW: ISO C 9x
3988 fattach NEW: STREAMS
3989 fdetach NEW: STREAMS
3993 feclearexcept NEW: ISO C 9x
3994 fegetenv NEW: ISO C 9x
3995 fegetexceptflag NEW: ISO C 9x
3996 fegetround NEW: ISO C 9x
3997 feholdexcept NEW: ISO C 9x
3999 feraiseexcept NEW: ISO C 9x
4000 ferror_locked REMOVED
4001 fesetenv NEW: ISO C 9x
4002 fesetexceptflag NEW: ISO C 9x
4003 fesetround NEW: ISO C 9x
4004 fetestexcept NEW: ISO C 9x
4005 feupdateenv NEW: ISO C 9x
4006 fflush_locked REMOVED
4010 fgets_unlocked NEW: GNU ext.
4011 fileno_locked REMOVED
4023 fputc_locked REMOVED
4024 fputs_unlocked NEW: GNU ext.
4025 fread_unlocked NEW: GNU ext.
4030 fstatvfs NEW: Unix98
4034 ftruncate64 NEW: LFS
4036 fwrite_unlocked NEW: GNU ext.
4037 gai_strerror NEW: IPv6
4041 getchar_locked REMOVED
4043 getdate_err NEW: Unix98
4044 getdate_r NEW: GNU ext.
4046 getnameinfo NEW: IPv6
4047 getnetname NEW: Secure RPC
4048 getpmsg NEW: STREAMS
4049 getpt NEW: Unix98 PTY
4050 getrlimit64 NEW: LFS
4051 getutxent NEW: Unix98
4052 getutxid NEW: Unix98
4053 getutxline NEW: Unix98
4054 glob64 NEW: GNU ext.
4055 globfree64 NEW: GNU ext.
4056 gnu_get_libc_release NEW: GNU ext.
4057 gnu_get_libc_version NEW: GNU ext.
4058 grantpt NEW: Unix98 PTY
4059 host2netname NEW: Secure RPC
4061 iconv_close NEW: iconv
4062 iconv_open NEW: iconv
4063 if_freenameindex NEW: IPv6
4064 if_indextoname NEW: IPv6
4065 if_nameindex NEW: IPv6
4066 if_nametoindex NEW: IPv6
4067 in6addr_any NEW: IPv6
4068 in6addr_loopback NEW: IPv6
4069 inet6_isipv4mapped NEW: IPv6
4070 isastream NEW: STREAMS
4071 iswblank NEW: GNU ext.
4072 key_decryptsession NEW: Secure RPC
4073 key_decryptsession_pk NEW: Secure RPC
4074 key_encryptsession NEW: Secure RPC
4075 key_encryptsession_pk NEW: Secure RPC
4076 key_gendes NEW: Secure RPC
4077 key_get_conv NEW: Secure RPC
4078 key_secretkey_is_set NEW: Secure RPC
4079 key_setnet NEW: Secure RPC
4080 key_setsecret NEW: Secure RPC
4081 llrint NEW: ISO C 9x
4082 llrintf NEW: ISO C 9x
4083 llrintl NEW: ISO C 9x
4084 llround NEW: ISO C 9x
4085 llroundf NEW: ISO C 9x
4086 llroundl NEW: ISO C 9x
4091 lrintf NEW: ISO C 9x
4092 lrintl NEW: ISO C 9x
4093 lround NEW: ISO C 9x
4094 lroundf NEW: ISO C 9x
4095 lroundl NEW: ISO C 9x
4097 makecontext NEW: Unix98
4098 mempcpy NEW: GNU ext.
4101 modify_ldt NEW: kernel
4105 nearbyint NEW: ISO C 9x
4106 nearbyintf NEW: ISO C 9x
4107 nearbyintl NEW: ISO C 9x
4108 netname2host NEW: Secure RPC
4109 netname2user NEW: Secure RPC
4110 nexttoward NEW: ISO C 9x
4111 nexttowardf NEW: ISO C 9x
4112 nexttowardl NEW: ISO C 9x
4116 passwd2des NEW: Secure RPC
4118 pow10f NEW: GNU ext.
4119 pow10l NEW: GNU ext.
4122 printf_size NEW: GNU ext.
4123 printf_size_info NEW: GNU ext.
4124 profil_counter REMOVED
4125 pthread_mutexattr_getkind_np REPLACED
4126 pthread_mutexattr_setkind_np REPLACED
4127 ptsname NEW: Unix98 PTY
4128 ptsname_r NEW: Unix98 PTY
4130 putchar_locked REMOVED
4131 putgrent NEW: GNU ext.
4133 putpmsg NEW: STREAMS
4134 pututxline NEW: Unix98
4138 readdir64_r NEW: LFS
4139 remquo NEW: ISO C 9x
4140 remquof NEW: ISO C 9x
4141 remquol NEW: ISO C 9x
4143 roundf NEW: ISO C 9x
4144 roundl NEW: ISO C 9x
4146 scalbln NEW: ISO C 9x
4147 scalblnf NEW: ISO C 9x
4148 scalblnl NEW: ISO C 9x
4150 sendfile NEW: kernel
4151 setcontext NEW: Unix98
4152 setrlimit64 NEW: LFS
4153 setutxent NEW: Unix98
4155 sigignore NEW: Unix98
4156 sigqueue NEW: POSIX.1b
4157 sigrelse NEW: Unix98
4158 sigset NEW: POSIX.1b
4159 sigtimedwait NEW: POSIX.1b
4160 sigwaitinfo NEW: POSIX.1b
4161 sincos NEW: GNU ext.
4162 sincosf NEW: GNU ext.
4163 sincosl NEW: GNU ext.
4167 strcasestr NEW: GNU ext.
4168 strtoimax NEW: ISO C 9x
4169 strtoumax NEW: ISO C 9x
4170 strverscmp NEW: GNU ext.
4171 svcauthdes_stats NEW: Secure RPC
4172 svcunix_create NEW: sunrpc etx
4173 svcunixfd_create NEW: sunrpc ext
4174 swapcontext NEW: Unix98
4175 tcgetsid NEW: Unix98 PTY
4176 tdestroy NEW: GNU ext.
4177 tgamma NEW: ISO C 9x
4178 tgammaf NEW: ISO C 9x
4179 tgammal NEW: ISO C 9x
4183 truncf NEW: ISO C 9x
4184 truncl NEW: ISO C 9x
4186 unlockpt NEW: Unix98 PTY
4187 updwtmpx NEW: Unix98
4188 user2netname NEW: Secure RPC
4189 utmpxname NEW: Unix98
4190 versionsort NEW: GNU ext.
4191 versionsort64 NEW: GNU ext.
4193 wcscasecmp NEW: GNU ext.
4194 wcsncasecmp NEW: GNU ext.
4195 wcsnlen NEW: GNU ext.
4196 wcstoimax NEW: ISO C 9x
4197 wcstoll NEW: ISO C 9x
4198 wcstoull NEW: ISO C 9x
4199 wcstoumax NEW: ISO C 9x
4201 wordexp NEW: POSIX.2
4202 wordfree NEW: POSIX.2
4203 write_profiling REMOVED
4204 xdecrypt NEW: Secure RPC
4205 xdr_authdes_cred NEW: Secure RPC
4206 xdr_authdes_verf NEW: Secure RPC
4207 xdr_cryptkeyarg NEW: Secure RPC
4208 xdr_cryptkeyarg2 NEW: Secure RPC
4209 xdr_cryptkeyres NEW: Secure RPC
4210 xdr_getcredres NEW: Secure RPC
4211 xdr_key_netstarg NEW: Secure RPC
4212 xdr_key_netstres NEW: Secure RPC
4213 xdr_keybuf NEW: Secure RPC
4214 xdr_keystatus NEW: Secure RPC
4215 xdr_netnamestr NEW: Secure RPC
4216 xdr_sizeof NEW: Secure RPC
4217 xdr_unixcred NEW: sunrpc ext
4218 xencrypt NEW: Secure RPC
4219 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4230 * inet_ntoa is thread-safe
4232 * updwtmp is moved from libutil to libc
4234 * rewrite of cbrt function
4236 * update of timezone data
4250 * add atoll function
4252 * fix complex problems in Berkeley DB code
4254 * fix math functions
4258 * fixed lots of header problems (especially Linux/GNU specific)
4260 * dynamic loader preserves all registers
4262 * Roland McGrath provided support for handling of auxiliary objects in
4263 the ELF dynamic loader.
4265 * support for parallel builds is improved
4269 * GNU extensions are no longer declared by default. To enable them you
4270 must define the macro `_GNU_SOURCE' in your program or compile with
4273 * The library has changed from using GNU ld symbol aliases to using weak
4274 symbols where available. The ELF object file format supports weak
4275 symbols; GNU ld also supports weak symbols in the a.out format. (There
4276 is also now support for other GNU ld extensions in ELF. Use the
4277 `--with-elf' option to configure to indicate you have ELF, and
4278 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4279 of many files which contained only symbol aliases, reducing the size of
4280 the source and the compiled library; many other files were renamed to
4281 less cryptic names previously occupied by the symbol alias files.
4282 There is a new header file <elf.h> for programs which operate on
4283 files in the ELF format.
4285 * Converted to Autoconf version 2, so `configure' has more options.
4286 Run `configure --help' to see the details.
4288 * The library can now be configured to build profiling, highly-optimized
4289 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4290 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4291 `configure' enable building these extra libraries. The shared library is
4292 built by default when using both ELF and GNU ld. When shared libraries
4293 are enabled, the new library `-ldl' is available for arbitrary run-time
4294 loading of shared objects; its interface is defined in <dlfcn.h>. The
4295 new header file <link.h> gives access to the internals of the run-time
4296 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4297 application of same name on other systems and it provides information
4298 about dynamically linked binaries.
4300 * The C library now provides the run-time support code for profiling
4301 executables compiled with `-pg'. Programs can control the profiling code
4302 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4303 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4304 the support for this file format was contributed by David Mosberger-Tang.
4306 * The math code has been replaced with a math library based on fdlibm from
4307 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4308 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4309 functions. The math functions now reside in a separate library, so
4310 programs using them will need to use `-lm' their linking commands.
4312 * John C. Bowman contributed optimized ix87 assembler inline functions.
4314 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4315 mechanism similar to that found in Solaris 2. This is now used for the
4316 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4317 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4318 file controls what services are used for each individual database. This
4319 works by loading shared libraries with names specified in `nsswitch.conf',
4320 so service modules can be changed or added at any time without even
4321 relinking any program. Currently there are the file, db, and NIS based
4322 NSS services available.
4324 * The new functions `strtoq' and `strtouq' parse integer values from
4325 strings, like `strtol' and `strtoul', but they return `long long int' and
4326 `unsigned long long int' values, respectively (64-bit quantities).
4328 * The new functions `strtof' and `strtold' parse floating-point values from
4329 strings, like `strtod', but they return `float' and `long double' values,
4330 respectively (on some machines `double' and `long double' are the same).
4332 * Ulrich Drepper has contributed new implementations of the floating-point
4333 printing and reading code used in the `printf' family of functions and
4334 `strtod', `strtof', and `strtold'. These new functions are perfectly
4335 accurate, and much faster than the old ones.
4337 * The implementation of the POSIX locale model was completely rewritten by
4338 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4339 to compile the POSIX locale definition.
4341 * The former dummy implementations of the strcoll and strxfrm function are
4342 now replaced by fully functional code contributed by Ulrich Drepper. The
4343 collation information comes from the POSIX locale definitions.
4345 * The new header <langinfo.h> defines an interface for accessing
4346 various locale-dependent data (using the locale chosen with `setlocale').
4348 * Ulrich Drepper has contributed a new suite of functions for operation on
4349 wide-character and multibyte-character strings, in <wchar.h>;
4350 and classification and case conversion of wide characters, in <wctype.h>.
4351 These new functions are conforming to the ISO C, Amendement 1 specification.
4353 * There is now a second implementation of the standard I/O library available.
4354 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4355 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4357 * You can now use positional parameter specifications in format strings
4358 for the `printf' and `scanf' families of functions. For example,
4359 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4360 ``Number 6, Mr Jones''. This is mainly useful when providing different
4361 format strings for different languages, whose grammars may dictate
4362 different orderings of the values being printed. To support this
4363 feature, the interface for `register_printf_handler' has changed; see
4364 the header file <printf.h> for details.
4366 * The `printf' and `scanf' families of functions now understand a new
4367 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4368 to group numbers as indicated by the locale; for `scanf' and friends, this
4369 says to accept as valid only a number with all the proper grouping
4370 separators in the right places. In the default "C" locale, numbers are
4371 not grouped; but locales for specific countries will define the usual
4372 conventions (i.e. separate thousands with `,' in the US locale).
4374 * The pgrp functions have been regularized, slightly incompatibly but much
4375 less confusingly. The core functions are now `getpgid' and `setpgid',
4376 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4377 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4378 provided for compatibility. There is no longer an incompatible `getpgrp'
4379 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4381 * The new header file <fts.h> and suite of functions simplify programs that
4382 operate on directory trees. This code comes from 4.4 BSD.
4384 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4385 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4386 NSS scheme used in glibc.
4388 * There is a new malloc debugging hook `__memalign_hook'.
4390 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4391 `unsigned int' in <sys/types.h>. These are for compatibility only and
4392 their use is discouraged.
4394 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4395 done differently, so that it works even without GNU ld.
4397 * New function `euidaccess' checks allowed access to a file like `access',
4398 but using the effective IDs instead of the real IDs.
4400 * The time zone data files have been updated for the latest and greatest
4401 local time conventions of the countries of the world.
4403 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4406 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4407 for formatting floating-point numbers. They are provided only for
4408 compatibility; new programs should use `sprintf' instead. There are
4409 also equivalent function for the `long double' floating-point type and
4410 all functions also exist in a reentrant form.
4412 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4413 functions for maintaining the login-record files (primarily of use to
4414 system programs such as `login'), and convenient functions for
4415 allocating and initializing a pseudo-terminal (pty) device.
4417 * Ulrich Drepper has contributed new support for System V style
4418 shared memory and IPC on systems that support it.
4420 * Ulrich Drepper has contributed several miscellaneous new functions found
4421 in System V: The `hsearch' family of functions provide an effective
4422 implementation of hash tables; `a64l' and `l64a' provide a very simple
4423 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4426 * Ulrich Drepper has contributed new reentrant counterparts for the
4427 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4429 * Ulrich Drepper has contributed new, highly-optimized versions of several
4430 string functions for the i486/Pentium family of processors.
4432 * Ulrich Drepper has updated the Linux-specific code, based largely
4433 on work done in Hongjiu Lu's version of GNU libc for Linux.
4434 The GNU library now supports Linux versions 2.0.10 and later,
4435 using the ELF object file format (i[3456]86-*-linux).
4437 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4439 * David Mosberger-Tang and Richard Henderson have ported the C library
4440 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4441 dynamic linking support for ELF/Alpha.
4443 * Richard Henderson contributed several Alpha optimized assembler function
4444 for arithmetic and string handling.
4446 * Ulrich Drepper has contributed a new set of message catalog functions to
4447 support multiple languages using the <libintl.h> interface, for use with
4448 his new package GNU gettext. Translation volunteers have contributed
4449 catalogs of the library's messages in Spanish, German, and Korean.
4451 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4452 program and the `catgets' function for reading the catalog files it
4453 creates. (The <libintl.h> interface is preferred; we include the
4454 <nl_types.h> interface using `catgets' only for source compatibility with
4455 programs already written to use it.)
4457 * New header file <values.h> gives SVID-compatible names for <limits.h>
4460 * Various new macros, declarations, and small header files for compatibility
4463 * New function `group_member' is a convenient way to check if a process has
4464 a given effective group ID.
4466 * When using GCC 2.7 and later, the socket functions are now declared in a
4467 special way so that passing an argument of type `struct sockaddr_in *',
4468 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4469 `struct sockaddr *' type, does not generate a type-clash warning.
4471 * New function `error' declared in header file <error.h> is a convenient
4472 function for printing error messages and optionally exiting; this is the
4473 canonical function used in GNU programs. The new functions `err', `warn',
4474 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4475 doing the same thing.
4477 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4478 POSIX.2 `glob' function to do ~ and {...} expansion.
4480 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4481 `clearenv' which is used in POSIX.9 is also available.
4483 * New function `getsid' returns session ID number on systems that support it.
4485 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4486 files <db.h> and <mpool.h> provide a rich set of functions for several
4487 types of simple databases stored in memory and in files, and <ndbm.h> is
4488 an old `ndbm'-compatible interface using the `db' functions. Link with
4489 `-ldb' to get these functions.
4491 * New macro `strdupa' copies a string like `strdup', but uses local stack
4492 space from `alloca' instead of dynamic heap space from `malloc'.
4494 * New function `strnlen' is like `strlen' but searches only a given maximum
4495 number of characters for the null terminator. `stpncpy', `strndup' and
4496 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4499 * New function `statfs' in header <sys/statfs.h>.
4501 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4502 provide convenient functions for operating on blocks of null-terminated
4505 * A new suite of functions in <utmp.h> handle all the details of reading
4506 and writing the utmp file.
4508 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4511 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4512 completely POSIX compliant and also implemented the extended functionality
4513 to handle alternate digit representation and alternate era date formats.
4515 * Ulrich Drepper provided an implementation of the `strptime' function
4516 defined in XPG4.2 which transforms a string into a `struct tm' value.
4518 * Paul Eggert provided the tzselect shell script as part of the timezone
4519 code. The shell script makes it easy to select the correct timezone
4522 * The implementation of the malloc family of functions is completely replaced
4523 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4524 The implementation uses the mmap function (if available) and it is
4525 optimized for the use in multi threaded programs.
4527 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4528 functions. This new functionality is usable by specifying a special
4529 salt string and it is compatible with implementation on *BSD systems.
4531 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4532 `getsubopt' to handle second level command line options, `bsd_signal'
4533 to access BSD style `signal' functionality, the obsolete `regexp' style
4536 * the `lchown' function is available on system which support this
4539 * The implementation of the shadow password handling function was contributed
4542 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4544 * POSIX.1g support was added. The <sys/select.h> header is available,
4545 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4546 implementation of `getaddrinfo'.
4550 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4552 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4553 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4555 * The new function `daemon' from 4.4 BSD is useful for server programs that
4556 want to put themselves in the background.
4558 * Joel Sherrill has contributed support for several standalone boards that
4559 run without an operating system.
4561 * `printf', `scanf' and friends now accept a `q' type modifier for long
4562 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4564 * All of the code taken from BSD (notably most of the math and networking
4565 routines) has been updated from the BSD 4.4-Lite release.
4567 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4569 * The new functions `getdomainname' and `setdomainname' fetch or change the
4570 YP/NIS domain name. These are system calls which exist on systems which
4573 * The time zone data files have been updated for the latest international
4576 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4577 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4581 * The C library now includes support for Sun RPC, from Sun's free
4582 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4583 are included. (There is still no support for YP.)
4585 * Tom Quinn has contributed a port of the C library to SGI machines running
4586 Irix 4 (mips-sgi-irix4).
4588 * The new `lockf' function is a simplified interface to the locking
4589 facilities of `fcntl', included for compatibility.
4591 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4593 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4595 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4598 * The error message printed by `assert' for a failed assertion now includes
4599 the name of the program (if using GNU ld) and the name of the calling
4600 function (with versions of GCC that support this).
4602 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4604 * The library now includes the <sys/mman.h> header file and memory
4605 management functions `mmap', `munmap', `mprotect', `msync', and
4606 `madvise', on systems that support those facilities.
4608 * The interface for `mcheck' has changed slightly: the function called to
4609 abort the program when an allocation inconsistency is detected now takes
4610 an argument that indicates the type of failure. The new function
4611 `mprobe' lets you request a consistency check for a particular block at
4612 any time (checks are normally done only when you call `free' or `realloc'
4615 * It is now possible to easily cross-compile the C library, building on one
4616 system a library to run on another machine and/or operating system. All
4617 you need to do is set the variable `HOST_CC' in `configparms' to the
4618 native compiler for programs to run on the machine you are building on (a
4619 few generator programs are used on Unix systems); set `CC' to the
4622 * The new function `fexecve' (only implemented on the GNU system) executes
4623 a program file given a file descriptor already open on the file.
4627 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4628 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4630 * You can set the variable `libprefix' in `configparms' to specify a prefix
4631 to be prepended to installed library files; this makes it easy to install
4632 the GNU C library to be linked as `-lgnuc' or whatever.
4634 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4635 copies a limited number of characters from a string, and returns the
4636 address of the last character written.
4638 * You no longer need to check for whether the installed `stddef.h' is
4639 compatible with the GNU C library. configure now checks for you.
4641 * You can now define a per-stream `fileno' function to convert the
4642 stream's cookie into an integral file descriptor.
4644 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4645 allocates zero bytes of storage, and returns a unique pointer which
4646 you can pass to `realloc' or `free'. The behavior is undefined if
4647 you dereference this pointer.
4649 * The C library now runs on Sony NEWS m68k machines running either
4650 NewsOS 3 or NewsOS 4.
4652 * The new `syscall' function is a system-dependent primitive function
4653 for invoking system calls. It has the canonical behavior on Unix
4654 systems, including unreliable return values for some calls (such as
4655 `pipe', `fork' and `getppid').
4657 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4658 to `EAGAIN', which is the preferred name. On systems whose kernels
4659 use two distinct codes, the C library now translates EWOULDBLOCK to
4660 EAGAIN in every system call function.
4664 * The GNU C Library Reference Manual is now distributed with the library.
4665 `make dvi' will produce a DVI file of the printed manual.
4666 `make info' will produce Info files that you can read on line using C-h i
4667 in Emacs or the `info' program.
4668 Please send comments on the manual to bug-glibc-manual@gnu.org.
4670 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4672 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4674 * Jason Merrill has contributed a port to the Sequent Symmetry running
4675 Dynix version 3 (i386-sequent-dynix).
4677 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4678 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4680 * New function `memory_warnings' lets you arrange to get warnings when
4681 malloc is running out of memory to allocate, like Emacs gives you.
4683 * The C library now contains the relocating allocator used in Emacs 19 for
4684 its editing buffers. This allocator (ralloc) minimizes allocation
4685 overhead and fragmentation by moving allocated regions around whenever it
4686 needs to. You always refer to a ralloc'd region with a "handle" (a
4687 pointer to a pointer--an object of type `void **').
4689 * There is a new `printf' format: `%m' gives you the string corresponding
4690 to the error code in `errno'.
4692 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4693 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4694 pass, the `a' modifier says to fill in a `char **' you pass with a
4697 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4698 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4699 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4701 * `mkstemp' is a traditional Unix function to atomically create and open a
4702 uniquely-named temporary file.
4706 * The standard location for the file that says what the local timezone is
4707 has changed again. It is now `/usr/local/etc/localtime' (or more
4708 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4710 * The distribution no longer contains any files with names longer than 14
4713 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4714 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4716 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4720 * The configuration process has changed quite a bit. The `configure'
4721 script is now used just like the configuration scripts for other GNU
4722 packages. The `sysdeps' directory hierarchy is much rearranged.
4723 The file `INSTALL' explains the new scheme in detail.
4725 * The header files no longer need to be processed into ANSI C and
4726 traditional C versions. There is just one set of files to install, and
4727 it will work with ANSI or old C compilers (including `gcc -traditional').
4729 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4730 MIPS DECStation running Ultrix 4.
4732 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4733 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4734 made itself into a shared library.
4736 * Yet further improved support for the i386, running 4.3 BSD-like systems
4737 (such as Mach 3 with the Unix single-server), or System V.
4739 * New function `strncasecmp' to do case-insensitive string comparison
4740 with limited length.
4742 * New function `strsep' is a reentrant alternative to `strtok'.
4744 * New functions `scandir' and `alphasort' for searching directories.
4746 * New function `setenv' is a better interface to `putenv'.
4748 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4749 function for traversing a directory tree.
4751 * The GNU obstack package is now also part of the C library.
4752 The new function `open_obstack_stream' creates a stdio stream that
4753 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4754 formatted output directly to an obstack.
4756 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4757 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4759 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4761 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4762 things to your strings.
4764 * There are some new test programs: test-fseek, testmb, and testrand.
4766 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4767 These ports are not finished, but are a good starting place for really
4768 supporting those systems.
4770 * `/etc/localtime' is now the standard location for the file that says what
4771 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4772 This follows the general principle that `/etc' is the place for all local
4773 configuration files.
4775 * The C library header files now use `extern "C"' when used by the C++
4776 compiler, so the C library should now work with C++ code.
4778 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4779 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4782 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4783 function which uses the `merge sort' algorithm, and is said to be
4784 significantly faster than the old GNU `qsort' function. Merge sort is
4785 now the standard `qsort' function. The new algorithm can require a lot
4786 of temporary storage; so, the old sorting function is called when the
4787 required storage is not available.
4789 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4790 provides the Unix `crypt' function, plus some other entry points.
4792 * The code and header files taken from 4.4 BSD have been updated with the
4793 latest files released from Berkeley.
4795 ----------------------------------------------------------------------
4796 Copyright information:
4798 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4800 Permission is granted to anyone to make or distribute verbatim copies
4801 of this document as received, in any medium, provided that the
4802 copyright notice and this permission notice are preserved,
4803 thus giving the recipient permission to redistribute in turn.
4805 Permission is granted to distribute modified versions
4806 of this document, or of portions of it,
4807 under the above conditions, provided also that they
4808 carry prominent notices stating who last changed them.
4811 version-control: never