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 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
24 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
27 * Nominative and genitive month names are now supported for the Catalan and
28 Czech languages. The Catalan and Greek languages now support abbreviated
29 alternative month names.
31 * Building and running on GNU/Hurd systems now works without out-of-tree
34 Deprecated and removed features, and other changes affecting compatibility:
36 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
37 installed. Software that was using either header should be updated to
38 use standard <stdio.h> interfaces instead.
40 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
41 This was never required by the C standard, and the macros just expanded
42 to call alternative names for the same functions. If you hoped getc and
43 putc would provide performance improvements over fgetc and fputc, instead
44 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
45 necessary, flockfile and funlockfile.
47 * All stdio functions now treat end-of-file as a sticky condition. If you
48 read from a file until EOF, and then the file is enlarged by another
49 process, you must call clearerr or another function with the same effect
50 (e.g. fseek, rewind) before you can read the additional data. This
51 corrects a longstanding C99 conformance bug. It is most likely to affect
52 programs that use stdio to read interactive input from a terminal.
55 * The macros 'major', 'minor', and 'makedev' are now only available from
56 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
57 headers that happen to include <sys/types.h>. These macros are rarely
58 used, not part of POSIX nor XSI, and their names frequently collide with
59 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
62 <sys/sysmacros.h> is a GNU extension. Portable programs that require
63 these macros should first include <sys/types.h>, and then include
64 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
66 * The tilegx*-*-linux-gnu configurations are no longer supported.
68 * The obsolete function ustat is no longer available to newly linked
69 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
70 function has been deprecated in favor of fstatfs and statfs.
72 Changes to build and runtime requirements:
74 [Add changes to build and runtime requirements here]
76 Security related changes:
78 [Add security related changes here]
80 The following bugs are resolved with this release:
82 [The release manager will add the list generated by
83 scripts/list-fixed-bugs.py just before the release.]
90 * The GNU C Library can now be compiled with support for building static
91 PIE executables (See --enable-static-pie in INSTALL). These static PIE
92 executables are like static executables but can be loaded at any address
93 and provide additional security hardening benefits at the cost of some
94 memory and performance. When the library is built with --enable-static-pie
95 the resulting libc.a is usable with GCC 8 and above to create static PIE
96 executables using the GCC option '-static-pie'. This feature is currently
97 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
98 aarch64 with binutils 2.30 or later.
100 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
101 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
104 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
106 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
108 * In order to support faster and safer process termination the malloc API
109 family of functions will no longer print a failure address and stack
110 backtrace after detecting heap corruption. The goal is to minimize the
111 amount of work done after corruption is detected and to avoid potential
112 security issues in continued process execution. Reducing shutdown time
113 leads to lower overall process restart latency, so there is benefit both
114 from a security and performance perspective.
116 * The abort function terminates the process immediately, without flushing
117 stdio streams. Previous glibc versions used to flush streams, resulting
118 in deadlocks and further data corruption. This change also affects
119 process aborts as the result of assertion failures.
121 * On platforms where long double has the IEEE binary128 format (aarch64,
122 alpha, mips64, riscv, s390 and sparc), the math library now implements
123 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
124 These are the same interfaces added in version 2.26 for some platforms where
125 this format is supported but is not the format of long double.
127 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
128 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
129 implements interfaces for that type, as defined by ISO/IEC TS
130 18661-3:2015. These are corresponding interfaces to those supported for
133 * The math library now implements interfaces for the _Float32, _Float64 and
134 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
135 corresponding interfaces to those supported for _Float128.
137 * glibc now implements the memfd_create and mlock2 functions on Linux.
139 * Support for memory protection keys was added. The <sys/mman.h> header now
140 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
143 * The copy_file_range function was added.
145 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
147 * The ldconfig utility now processes `include' directives using the C/POSIX
148 collation ordering. Previous glibc versions used locale-specific
149 ordering, the change might break systems that relied on that.
151 * Support for two grammatical forms of month names has been added.
152 In a call to strftime, the "%B" and "%b" format specifiers will now
153 produce the grammatical form required when the month is used as part
154 of a complete date. New "%OB" and "%Ob" specifiers produce the form
155 required when the month is named by itself. For instance, in Greek
156 and in many Slavic and Baltic languages, "%B" will produce the month
157 in genitive case, and "%OB" will produce the month in nominative case.
159 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
160 are all valid and will all accept any known form of month
161 name---standalone or complete, abbreviated or full. In a call to
162 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
163 the strings used by "%B" and "%b", respectively. New query
164 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
165 used by "%OB" and "%Ob", respectively.
167 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
168 define the strings for %OB and %Ob, respectively; these have the
169 same syntax as "mon" and "abmon". These arrays are optional; if they
170 are not provided then they have the same content as "mon" and "abmon",
173 These features are provided for locales which define "alt_mon" and/or
174 "ab_alt_mon" in their locale source data. This release includes such
175 alternative month name data for the following languages: Belarusian,
176 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
178 This feature is currently a GNU extension, but it is expected to
179 be added to the next revision of POSIX, and it is also already
180 available on some BSD-derived operating systems.
182 This feature will cause existing statically compiled applications
183 to fail to load locales and fall back to the builtin C/POSIX locales.
184 See notes below for other changes affecting compatibility.
186 * Support for the RISC-V ISA running on Linux has been added. This port
187 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
188 for the following ISA and ABI pairs:
194 Deprecated and removed features, and other changes affecting compatibility:
196 * Statically compiled applications attempting to load locales compiled for the
197 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
198 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
199 support for two grammatical forms of the month names, also extends the locale
200 data binary format. Static applications needing locale support must be
201 recompiled to match the runtime and data they are deployed with. In some
202 distributions there is an upgrade window where dynamically linked applications
203 may use a new library but the old locale data and also fall back to the
204 builtin C/POSIX locales; restarting the application process is sufficient to
207 * Support for statically linked applications which call dlopen is deprecated
208 and will be removed in a future version of glibc. Applications which call
209 dlopen need to be linked dynamically instead.
211 * Support for old programs which use internal stdio data structures and
212 functions is deprecated. This includes programs which use the C++ streams
213 provided by libstdc++ in GCC 2.95. Programs which use the internal
214 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
215 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
216 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
217 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
218 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
219 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
220 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
221 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
222 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
223 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
224 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
225 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
226 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
227 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
228 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
229 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
230 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
231 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
232 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
233 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
234 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
235 future version of glibc. Unlike other symbol removals, these old
236 applications will not be supported using compatibility symbols.
238 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
239 defined by <sys/ptrace.h>.
241 * libm no longer supports SVID error handling (calling a user-provided
242 matherr function on error) or the _LIB_VERSION variable to control error
243 handling. (SVID error handling and the _LIB_VERSION variable still work
244 for binaries linked against older versions of the GNU C Library.) The
245 libieee.a library is no longer provided. math.h no longer defines struct
246 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
247 TLOSS, PLOSS and HUGE.
249 * The libm functions pow10, pow10f and pow10l are no longer supported for
250 new programs. Programs should use the standard names exp10, exp10f and
251 exp10l for these functions instead.
253 * The mcontext_t type is no longer the same as struct sigcontext. On
254 platforms where it was previously the same, this changes the C++ name
255 mangling for interfaces involving this type.
257 * The add-ons mechanism for building additional packages at the same time as
258 glibc has been removed. The --enable-add-ons configure option is now
261 * The --without-fp configure option is now ignored. Whether hardware
262 floating-point instructions are used is now configured based on whether
263 the compiler used at configure time (without any options implied by a
264 --with-cpu= configure option) uses such instructions.
266 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
267 the specified string can be parsed as a domain name.
269 * In the malloc_info output, the <heap> element may contain another <aspace>
270 element, "subheaps", which contains the number of sub-heaps.
272 * The libresolv function p_secstodate is no longer supported for new
275 * The tilepro-*-linux-gnu configuration is no longer supported.
277 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
278 and will be removed in a future release. Software that is still using
279 either header should be updated to use standard <stdio.h> interfaces
282 libio.h was originally the header for a set of supported GNU extensions,
283 but they have not been maintained as such in many years, they are now
284 standing in the way of improvements to stdio, and we don't think there are
285 any remaining external users. _G_config.h was never intended for public
286 use, but predates the bits convention.
288 Changes to build and runtime requirements:
290 * bison version 2.7 or later is required to generate code in the 'intl'
293 Security related changes:
295 CVE-2009-5064: The ldd script would sometimes run the program under
296 examination directly, without preventing code execution through the
297 dynamic linker. (The glibc project disputes that this is a security
298 vulnerability; only trusted binaries must be examined using the ldd
301 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
302 suffered from a one-byte overflow during ~ operator processing (either
303 on the stack or the heap, depending on the length of the user name).
304 Reported by Tim Rühsen.
306 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
307 would sometimes fail to free memory allocated during ~ operator
308 processing, leading to a memory leak and, potentially, to a denial
311 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
312 without GLOB_NOESCAPE, could write past the end of a buffer while
313 unescaping user names. Reported by Tim Rühsen.
315 CVE-2017-17426: The malloc function, when called with an object size near
316 the value SIZE_MAX, would return a pointer to a buffer which is too small,
317 instead of NULL. This was a regression introduced with the new malloc
318 thread cache in glibc 2.26. Reported by Iain Buclaw.
320 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
321 to the allocation of too much memory. (This is not a security bug per se,
322 it is mentioned here only because of the CVE assignment.) Reported by
325 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
326 of the number of search path components. (This is not a security
327 vulnerability per se because no trust boundary is crossed if the fix for
328 CVE-2017-1000366 has been applied, but it is mentioned here only because
329 of the CVE assignment.) Reported by Qualys.
331 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
332 for AT_SECURE or SUID binaries could be used to load libraries from the
335 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
336 succeeds without returning an absolute path due to unexpected behaviour
337 of the Linux kernel getcwd syscall. Reported by halfdog.
339 CVE-2018-6485: The posix_memalign and memalign functions, when called with
340 an object size near the value of SIZE_MAX, would return a pointer to a
341 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
343 CVE-2018-6551: The malloc function, when called with an object size near
344 the value of SIZE_MAX, would return a pointer to a buffer which is too
345 small, instead of NULL.
347 The following bugs are resolved with this release:
349 [866] glob: glob should match dangling symlinks
350 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
351 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
353 [5997] math: Very slow execution of sinf function
354 [10580] localedata: hr_HR: updated locale
355 [10871] locale: 'mon' array should contain both nominative and genitive
357 [12349] localedata: eu_ES: incorrect thousands separator
358 [13605] localedata: shn_MM: new Shan locale
359 [13805] localedata: ru_RU: currency should use ',' as radix point
360 [13953] localedata: km_KH: locale update
361 [13994] localedata: mjw_IN: new locale
362 [14121] build: make writes .mo files in po directory
363 [14333] libc: Fix the race between atexit() and exit()
364 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
365 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
367 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
368 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
369 width Latin characters
370 [15332] localedata: es_CU: locale update
371 [15436] stdio: Don't close or flush stdio streams on abort
372 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
374 [16148] localedata: ca_ES: incorrect thousands separator
375 [16750] dynamic-link: ldd should not try to execute the binaries
377 [16777] localedata: pl_PL: incorrect thousands separator in locale
378 [16905] localedata: hanzi: new collation
379 [17563] localedata: cmn_TW: add hanzi collation
380 [17750] localedata: wrong collation order of diacritics in most locales
381 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
383 [17956] build: Build fails on missing definitions from header file
384 nss/nss.h when Mozilla NSS is used for cryptography
385 [18203] libc: realpath() does not handle unreachable paths correctly
386 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
387 [18812] localedata: kab_DZ: new Kabyle Algeria locale
388 [18822] libc: Internal functions are called via PLT
389 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
390 [19170] libc: __gmon_start__ defined in hppa in crtn.S
391 [19574] libc: glibc should support building static PIE binaries
392 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
394 [19971] glob: glob: Do not skip entries with zero d_ino values
395 [19982] localedata: fr.po: spelling mistake for error code EXDEV
396 [20008] localedata: km_KH: convert to translit_neutral
397 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
398 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
399 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
400 [20482] localedata: de_CH: abbreviated weekdays should be two letters
401 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
403 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
404 they're set, wrong errors returned by gaih_inet when lookup functions
406 [20756] localedata: [PATCH] Use Unicode wise thousands separator
407 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
409 [20952] localedata: yuw_PG: new locale
410 [21084] localedata: charmaps/IBM858: new codepage
411 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
412 [21242] libc: assert gives pedantic warning in old gcc versions
413 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
414 __regcall calling convention
415 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
416 [21326] libc: C99 functions are not declared for C++11 and later if
417 _GNU_SOURCE is not predefined by g++
418 [21457] libc: sys/ucontext.h namespace
419 [21530] libc: tmpfile() should be implemented using O_TMPFILE
420 [21660] math: GCC fails to compile a formula with tgmath.h
421 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
422 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
423 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
424 [21684] math: tgmath.h handling of complex integers
425 [21685] math: tgmath.h handling of bit-fields
426 [21686] math: tgmath.h handling of __int128
427 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
429 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
431 [21750] localedata: column width of characters incompatible with classical
433 [21754] malloc: malloc: Perform as little work as possible after heap
434 consistency check failures
435 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
436 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
437 [21791] string: Unused XXX_chk_XXX functions in libc.a
438 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
440 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
442 [21845] localedata: Added new Locale bho_NP
443 [21853] localedata: Fix abday Which looks same as day in zh_SG
444 [21854] localedata: Added New Locale en_SC
445 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
446 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
447 _dl_runtime_resolve_avx_slow
448 [21885] network: getaddrinfo: gethosts does not release resolver context
449 on memory allocation failure
450 [21899] libc: XPG4.2 sigaction namespace
451 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
453 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
455 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
456 errno value, causing getaddrinfo to fail
457 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
458 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
459 instead of EAI_NODATA
460 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
461 constant PTRACE_SEIZE_DEVEL
462 [21930] math: C-only gcc builtins used in <math.h> isinf
463 [21932] network: Unpaired __resolv_context_get in generic get*_r
465 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
466 [21944] libc: sigval namespace
467 [21951] localedata: Update hanzi collation by stroke
468 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
469 sysdeps/x86_64/fpu/e_expf.S
470 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
471 allocation in syscall loops)
472 [21959] localedata: Fix Country name for xh_ZA
473 [21960] localedata: Fix abmon for bem_ZM
474 [21966] math: AVX2 mathvec functions use FMA without checking
475 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
476 [21971] localedata: Added New Locale for mfe_MU
477 [21972] libc: assert macro requires operator== (int) for its argument type
478 [21973] math: [sparc] libm missing sqrtl compat symbol
479 [21974] libc: Remove __bb_init_func and __bb_exit_func
480 [21982] string: stratcliff.c: error: assuming signed overflow does not
482 [21986] stdio: __guess_grouping is called incorrectly
483 [21987] math: [sparc32] wrong bits/long-double.h installed
484 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
486 [22022] localedata: Missing country_name for mni_IN
487 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
489 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
491 [22026] locale: iconv_open: heap overflow on gconv_init failure
492 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
493 [22035] math: [m68k] bits/math-inline.h macro namespace
494 [22038] localedata: Fix abbreviated weeks and months for Somali
495 [22044] localedata: Remove redundant data for Limburgish Language
496 [22050] malloc: Linking with -lmcheck does not hook
497 __malloc_initialize_hook correctly
498 [22051] libc: zero terminator in the middle of glibc's .eh_frame
499 [22052] malloc: malloc failed to compile with GCC 7 and -O3
500 [22070] localedata: charmaps/UTF-8: wcwidth for
501 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
502 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
503 Jungseong and Jongseong) should be 0
504 [22078] nss: nss_files performance issue in multi mode
505 [22082] math: bits/math-finite.h exp10 condition
506 [22086] libc: pcprofiledump incorrect cross-endian condition
507 [22093] dynamic-link: ld.so no longer searches in .../x86_64
508 [22095] network: Name server address allocation memory leak in resolv.conf
510 [22096] network: __resolv_conf_attach can incorrectly free passed conf
512 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
513 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
515 [22111] malloc: malloc: per thread cache is not returned when thread exits
516 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
517 [22134] libc: [linux] implement fexecve with execveat
518 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
520 [22145] libc: ttyname() gives up too early in the face of namespaces
521 [22146] math: C++ build issue with float128 on x86_64
522 [22153] nptl: nptl: save error code before process termination
523 [22156] libc: [hppa,ia64,microblaze] Executable stack default
524 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
525 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
526 [22165] libc: [hppa] Text relocations in libc.so
527 [22180] libc: destructor registered via __cxa_atexit is called twice
528 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
529 symlinks") cause "make" segfaults
530 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
532 [22207] libc: FAIL: stdlib/test-atexit-race
533 [22225] math: nearbyint arithmetic moved before feholdexcept
534 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
535 [22235] math: iscanonical in C++ and float128
536 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
537 the svid compat wrapper
538 [22244] math: ynf and yn are wrong without the svid compat wrapper
539 [22273] libc: Improper assert in Linux posix_spawn implementation
540 [22284] libc: -pg -pie doesn't work
541 [22292] locale: localedef exits with error 4 when it should be error 1
542 [22294] locale: Allow "" for int_currency_symbol definition in locales.
543 [22295] locale: Don't warn on non-symbolic characters in locale sources in
545 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
546 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
548 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
549 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
550 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
551 [22322] libc: [mips64] wrong bits/long-double.h installed
552 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
553 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
555 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
556 [22343] malloc: Integer overflow in posix_memalign
557 [22347] libc: getrandom() returns the number of bytes that were copied to
558 the buffer even though the comments say "Return 0 on success and -1 on
560 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
561 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
563 [22370] dynamic-link: Incorrect note padding check
564 [22375] libc: malloc returns pointer from tcache_get when should return
565 NULL (CVE-2017-17426)
566 [22377] math: iseqsig, float128 and C++
567 [22382] localedata: Error in tpi_PG locale
568 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
569 inside the ASCII printable range
570 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
572 [22403] localedata: Slash needs escaping in some locales
573 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
574 [22409] network: res_hnok does not accept some host names used on the
576 [22412] network: res_dnok, res_hnok should perform syntax checks
577 [22413] network: ns_name_pton ignores syntactically invalid trailing
579 [22415] stdio: setvbuf can lead to invalid free/segfault
580 [22432] build: Non-deterministic build
581 [22439] malloc: malloc_info should compute summary statistics for all sub-
583 [22442] network: if_nametoindex could report index for the wrong
585 [22446] build: aliasing violation calling readlink in handle_request
586 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
587 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
588 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
589 __nscd_hash/__nss_hash
590 [22463] network: p_secstodate overflow handling
591 [22469] localedata: pl_PL LC_COLLATE does not use i18n
592 [22478] libc: sigwait can fail with EINTR
593 [22505] libc: ldconfig processes include directive in locale-specific
595 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
596 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
597 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
598 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
599 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
600 [22534] localedata: Collation rules for Serbian and Bosnian should be the
602 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
603 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
604 [22577] libc: missing newline after "cannot allocate TLS data structures
606 [22588] manual: manual/conf.texi: missing underscore in front of
608 [22593] math: nextafter and nexttoward are declared with const attribute
609 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
610 [22603] string: ia64 memchr overflows internal pointer check
611 [22605] libc: SH clone does not set the exit code correctly
612 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
614 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
615 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
616 modified in case of success
617 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
618 [22615] manual: manual: ambiguous wording about errno value in case of
620 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
622 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
623 binaries or if /proc is not mounted (CVE-2017-16997)
624 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
625 [22630] build: $(no-pie-ldflag) is no longer effective
626 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
627 [22635] nptl: pthread_self returns NULL before libpthread is loaded
628 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
629 [22637] nptl: guard size is subtracted from thread stack size instead of
631 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
632 [22657] localedata: hu_HU: Avoid double space in date
633 [22660] math: fmax, fmin sNaN handling on alpha
634 [22664] libc: New warning of GCC8
635 [22665] math: alpha: ceil and floor raise inexact exceptions
636 [22666] math: alpha: trunc raise inexact exceptions
637 [22667] libc: makecontext lacks stack alignment on i386
638 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
639 [22679] libc: getcwd(3) can succeed without returning an absolute path
641 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
643 [22687] math: [powerpc-nofpu] complex long double functions spurious
645 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
646 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
647 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
649 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
650 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
652 [22701] nis: Incomplete removal of libnsl
653 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
654 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
655 [22715] dynamic-link: FAIL: elf/tst-audit10
656 [22719] libc: Backtrace tests fail on hppa
657 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
658 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
659 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
660 before the first call to crypt_r () in crypt/badsalttest.c
667 * A per-thread cache has been added to malloc. Access to the cache requires
668 no locks and therefore significantly accelerates the fast path to allocate
669 and free small amounts of memory. Refilling an empty cache requires locking
670 the underlying arena. Performance measurements show significant gains in a
671 wide variety of user workloads. Workloads were captured using a special
672 instrumented malloc and analyzed with a malloc simulator. Contributed by
673 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
675 * Unicode 10.0.0 Support: Character encoding, character type info, and
676 transliteration tables are all updated to Unicode 10.0.0, using
677 generator scripts contributed by Mike FABIAN (Red Hat).
678 These updates cause user visible changes, especially the changes in
679 wcwidth for many emoji characters cause problems when emoji sequences
680 are rendered with pango, see for example:
681 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
683 * Collation of Hungarian has been overhauled and is now consistent with "The
684 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
687 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
689 - The GNU C Library will now detect when /etc/resolv.conf has been
690 modified and reload the changed configuration. The new resolver option
691 “no-reload” (RES_NORELOAD) disables this behavior.
693 - The GNU C Library now supports an arbitrary number of search domains
694 (configured using the “search” directive in /etc/resolv.conf);
695 previously, there was a hard limit of six domains. For backward
696 compatibility, applications that directly modify the ‘_res’ global
697 object are still limited to six search domains.
699 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
700 Library will now randomly pick a name server from the configuration as a
701 starting point. (Previously, the second name server was always used.)
703 * The tunables feature is now enabled by default. This allows users to tweak
704 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
706 * New function reallocarray, which resizes an allocated block (like realloc)
707 to the product of two sizes, with a guaranteed clean failure upon integer
708 overflow in the multiplication. Originally from OpenBSD, contributed by
709 Dennis Wölfing and Rüdiger Sonderfeld.
711 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
712 These are extended versions of preadv and pwritev, respectively, taking an
713 additional flags argument. The set of supported flags depends on the
714 running kernel; full support currently requires kernel 4.7 or later.
716 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
717 create a new session ID for the spawned process. This feature is
718 scheduled to be added to the next major revision of POSIX; for the time
719 being, it is available under _GNU_SOURCE.
721 * errno.h is now safe to use from C-preprocessed assembly language on all
722 supported operating systems. In this context, it will only define the
723 Exxxx constants, as preprocessor macros expanding to integer literals.
725 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
726 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
727 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
728 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
730 To compile programs that use this feature, the compiler must support
731 128-bit floating point with the type name _Float128 (as defined by TS
732 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
733 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
734 must be defined to make the new interfaces visible.
736 The new functions and macros correspond to those present for other
737 floating-point types (except for a few obsolescent interfaces not
738 supported for the new type), with F128 or f128 suffixes; for example,
739 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
740 printf or scanf formats for the new type; the strfromf128 and strtof128
741 interfaces should be used instead.
743 Deprecated and removed features, and other changes affecting compatibility:
745 * The synchronization that pthread_spin_unlock performs has been changed to
746 now be equivalent to a C11 atomic store with release memory order to the
747 spin lock's memory location. Previously, several (but not all)
748 architectures used stronger synchronization (e.g., containing what is
749 often called a full barrier). This change can improve performance, but
750 may affect odd fringe uses of spin locks that depend on the previous
751 behavior (e.g., using spin locks as atomic variables to try to implement
752 Dekker's mutual exclusion algorithm).
754 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
757 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
758 will only be built and installed when the GNU C Library is configured with
759 --enable-obsolete-rpc. This allows alternative RPC implementations, such
760 as TIRPC or rpcsvc-proto, to be used.
762 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
763 libnss_compat, are deprecated, and will not be built or installed by
766 The NIS(+) support library, libnsl, is also deprecated. By default, a
767 compatibility shared library will be built and installed, but not headers
768 or development libraries. Only a few NIS-related programs require this
769 library. (In particular, the GNU C Library has never required programs
770 that use 'gethostbyname' to be linked with libnsl.)
772 Replacement implementations based on TIRPC, which additionally support
773 IPv6, are available from <https://github.com/thkukuk/>. The configure
774 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
775 name service modules, to be built and installed.
777 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
778 support is enabled, the configured recursive resolver must support EDNS.
779 (Responding to EDNS-enabled queries with responses which are not
780 EDNS-enabled is fine, but FORMERR responses are not.)
782 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
783 servers have not supported this opcode for a long time.
785 * The _res_opcodes variable has been removed from libresolv. It had been
786 exported by accident.
788 * <string.h> no longer includes inline versions of any string functions,
789 as this kind of optimization is better done by the compiler. The macros
790 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
792 * The nonstandard header <xlocale.h> has been removed. Most programs should
793 use <locale.h> instead. If you have a specific need for the definition of
794 locale_t with no other declarations, please contact
795 libc-alpha@sourceware.org and explain.
797 * The obsolete header <sys/ultrasound.h> has been removed.
799 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
801 * The obsolete function cfree has been removed. Applications should use
804 * The stack_t type no longer has the name struct sigaltstack. This changes
805 the C++ name mangling for interfaces involving this type.
807 * The ucontext_t type no longer has the name struct ucontext. This changes
808 the C++ name mangling for interfaces involving this type.
810 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
811 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
812 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
813 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
814 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
815 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
816 fpregset_t type no longer has the name struct fpu. This changes the C++
817 name mangling for interfaces involving those types.
819 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
820 synced with the kernel:
822 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
823 are not supported on this architecture and have been removed.
825 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
826 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
827 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
829 Programs that assume the GET/SETREGS ptrace requests are universally
830 available will now fail to build, instead of malfunctioning at runtime.
832 Changes to build and runtime requirements:
834 * Linux kernel 3.2 or later is required at runtime, on all architectures
835 supported by that kernel. (This is a change from version 2.25 only for
838 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
840 * On most architectures, GCC 4.9 or later is required to build the GNU C
841 Library. On powerpc64le, GCC 6.2 or later is required.
843 Older GCC versions and non-GNU compilers are still supported when
844 compiling programs that use the GNU C Library. (We do not know exactly
845 how old, and some GNU extensions to C may be _de facto_ required. If you
846 are interested in helping us make this statement less vague, please
847 contact libc-alpha@sourceware.org.)
849 Security related changes:
851 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
852 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
854 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
855 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
857 * Avoid printing a backtrace from the __stack_chk_fail function since it is
858 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
861 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
862 fixed (CVE-2017-12133).
864 The following bugs are resolved with this release:
866 [984] network: Respond to changed resolv.conf in gethostbyname
867 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
868 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
870 [12189] libc: __stack_chk_fail should not attempt a backtrace
872 [14096] time: Race condition on timezone/tst-timezone.out
873 [14172] localedata: az_IR: new locale
874 [14995] build: glibc fails to build if gold is the default linker, even if
876 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
877 [16637] network: inet_pton function is accepting IPv6 with bad format
878 [16640] string: string/strtok.c: undefined behaviour inconsistent between
879 x86 and other generic code
880 [16875] localedata: ko_KR: fix lang_name
881 [17225] localedata: ar_SY: localized month names for May and June are
883 [17297] localedata: da_DK: wrong date_fmt string
884 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
885 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
886 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
887 [19066] localedata: ar_SA abbreviated day and month names are in English
888 [19569] network: resolv: Support an arbitrary number of search domains
889 [19570] network: Implement random DNS server selection in the stub
891 [19838] locale: localedef fails on PA-RISC
892 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
893 order of 0D36 and 0D37
894 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
896 [20098] libc: FAIL: debug/backtrace-tst on hppa
897 [20257] network: sunrpc: clntudp_call does not enforce timeout when
899 [20275] localedata: locale day/abday/mon/abmon should not have trailing
901 [20313] localedata: Update locale data to Unicode 9.0
902 [20424] manual: Document how to provide a malloc replacement
903 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
905 [20686] locale: Add el_GR@euro to SUPPORTED.
906 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
907 failures consistently
908 [21015] dynamic-link: Document and fix --enable-bind-now
909 [21016] nptl: pthread_cond support is broken on hppa
910 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
911 frame-pointer on i386
912 [21049] libc: segfault in longjmp_chk() due to clobbered processor
914 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
915 [21088] libc: Build fails with --enable-static-nss
916 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
917 generic c code is used
918 [21109] libc: Tunables broken on big-endian
919 [21112] math: powf has large ulp errors with base close to 1 and exponent
921 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
923 [21120] malloc: glibc malloc is incompatible with GCC 7
924 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
926 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
927 (-0.0) when linking with -lieee
928 [21171] math: log10, log2 and lgamma return incorrect results
929 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
930 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
931 [21207] localedata: ce_RU: update weekdays from CLDR
932 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
933 [21217] localedata: Update months from CLDR-31
934 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
935 [21243] libc: support_delete_temp_file should issue warning for failed
937 [21244] libc: support resolv_test_start() socket fd close should be
939 [21253] libc: localedef randomly segfaults when using -fstack-check due to
940 new posix_spawn implementation
941 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
942 leads to lower CPU frequency
943 [21259] libc: [alpha] termios.h missing IXANY for POSIX
944 [21261] libc: [sparc64] bits/setjmp.h namespace
945 [21267] network: [mips] bits/socket.h IOC* namespace
946 [21268] libc: [alpha] termios.h NL2, NL3 namespace
947 [21270] libc: mmap64 silently truncates large offset values
948 [21275] libc: posix_spawn always crashes on ia64 now
949 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
950 [21280] math: [powerpc] logbl for POWER7 return incorrect results
951 [21289] libc: Incorrect declaration for 32-bit platforms with
952 _FILE_OFFSET_BITS=64 causes build error
953 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
954 not support gethostbyname4_r
955 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
957 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
959 [21340] libc: Support POSIX_SPAWN_SETSID
960 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
961 [21359] network: ns_name_pack needs additional byte in destination buffer
962 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
963 against fragmentation attacks (CVE-2017-12132)
964 [21369] network: resolv: Remove EDNS fallback
965 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
967 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
968 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
969 [21393] stdio: Missing dup3 error check in freopen, freopen64
970 [21396] libc: Use AVX2 memcpy/memset on Skylake server
971 [21399] localedata: Bad description for U00EC in
972 localedata/charmaps/CP1254
973 [21411] malloc: realloc documentation error
974 [21426] network: sys/socket.h uio.h namespace
975 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
976 [21445] libc: signal.h bsd_signal namespace
977 [21455] network: Network headers stdint.h namespace
978 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
980 [21475] network: resolv: Overlong search path is truncated mid-label
981 [21511] libc: sigstack namespace
982 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
983 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
984 bits/syscall.h: No such file or directory
985 [21517] libc: struct sigaltstack namespace
986 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
987 [21533] localedata: Update locale data to Unicode 10.0
989 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
990 junk at end of line, first unrecognized character is `@'
991 [21538] libc: SIG_HOLD missing for XPG4
992 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
993 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
994 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
995 [21543] libc: sigevent namespace
996 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
997 size for general purpose registers in mcontext_t structure
998 [21550] libc: sigwait namespace
999 [21552] libc: XPG4 bsd_signal namespace
1000 [21554] libc: sigpause namespace
1001 [21560] libc: sys/wait.h signal.h namespace
1002 [21561] libc: waitid namespace
1003 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1004 '/usr/include/stdlib.h': Permission denied
1005 [21575] libc: sys/wait.h missing struct rusage definition
1006 [21584] libc: sigaltstack etc namespace
1007 [21597] libc: siginterrupt namespace
1008 [21607] math: hppa: FAIL: math/test-tgmath
1009 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1011 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1012 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1013 stack and heap (CVE-2017-1000366)
1014 [21625] libc: wait3 namespace
1015 [21654] nss: Incorrect pointer alignment in NSS group merge result
1017 [21657] network: Parse interface zone id for node-local multicast
1018 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1020 [21666] libc: .symver is used on common symbol
1021 [21668] network: resolv: res_init cross-thread broadcast introduces race
1023 [21687] math: tgmath.h totalorder, totalordermag return type
1024 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1026 [21696] libc: Incorrect assumption of of __cpu_mask in
1027 posix/sched_cpucount.c
1028 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1029 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1031 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1032 leading to relocation crash
1033 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1034 `resolv_conf_matches (resp, conf)' failed.
1035 [21710] localedata: Added Samoan language locale for Samoa
1036 [21711] localedata: Pashto yesstr/nostr locale are missing
1037 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1038 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1039 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1040 [21723] localedata: yesstr/nostr missing for Chinese language locale
1041 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1043 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1045 [21728] localedata: New Locale for Tongan language
1046 [21729] localedata: incorrect LC_NAME fields for hi_IN
1047 [21733] localedata: yesstr and nostr are missing for zh_HK
1048 [21734] localedata: Missing yesstr and nostr are for kw_GB
1049 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1050 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1051 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1052 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1054 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1056 [21749] localedata: Wrong abbreviated day name (“abday”) for
1058 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1059 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1060 [21759] localedata: missing yesstr and nostr for Tigrinya
1061 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1062 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1063 [21767] localedata: Missing Bislama locales
1064 [21768] localedata: Missing yesstr and nostr for aa_ET
1065 [21770] localedata: Missing Field in li_NL
1066 [21778] nptl: Robust mutex may deadlock
1067 [21779] libc: MicroBlaze segfaults when loading libpthread
1068 [21783] localedata: Fix int_select international_call_prefixes
1069 [21784] localedata: Inconsistency in country_isbn
1070 [21788] localedata: Missing Country Postal Abbreviations
1071 [21794] localedata: Added-country_isbn-for-Italy
1072 [21795] localedata: Add/Fix country_isbn for France
1073 [21796] localedata: Added country_isbn for Republic of Korea
1074 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1075 [21799] localedata: Added int_select international_call_prefixes
1076 [21801] localedata: Added int_select international_call_prefixes
1077 [21804] nptl: Double semicolon in thread-shared-types.h
1078 [21807] localedata: LC_ADDRESS fix for pap_CW
1079 [21808] localedata: Fix LC_ADDRESS for pap_AW
1080 [21821] localedata: Added country_name in mai_IN
1081 [21822] localedata: Fix LC_TIME for mai_IN
1082 [21823] localedata: missing yesstr, nostr for sa_IN
1083 [21825] localedata: Fix name_mrs for mag_IN
1084 [21828] localedata: 2.26 changelog should mention user visible changes
1086 [21835] localedata: Added Maithili language locale for Nepal
1087 [21838] localedata: Removed redundant data for the_NP
1088 [21839] localedata: Fix LC_MONETARY for ta_LK
1089 [21844] localedata: Fix Latin characters and Months Sequence.
1090 [21848] localedata: Fix mai_NP Title Name
1095 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1096 24731-2:2010, is supported to enable declarations of functions from that
1097 TR. Note that not all functions from that TR are supported by the GNU C
1100 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1101 18661-1:2014, is supported to enable declarations of functions and macros
1102 from that TS. Note that not all features from that TS are supported by
1105 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1106 18661-4:2015, is supported to enable declarations of functions and macros
1107 from that TS. Note that most features from that TS are not supported by
1110 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1111 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1112 Since the GNU C Library defaults to a much newer revision of POSIX, this
1113 will only affect programs that specifically request an old conformance
1114 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1115 see a change in the visible declarations, but a program compiled with
1116 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1119 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1120 defined by all multithreaded code, but glibc has not required this for
1123 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1124 means that in a future release, the macros “major”, “minor”, and “makedev”
1125 will only be available from <sys/sysmacros.h>.
1127 These macros are not part of POSIX nor XSI, and their names frequently
1128 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1129 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1130 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1133 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1134 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1135 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1137 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1138 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1139 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1140 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1141 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1142 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1143 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1144 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1145 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1146 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1147 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1150 * New <math.h> features are added from TS 18661-1:2014:
1152 - Signaling NaN macros: SNANF, SNAN, SNANL.
1154 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1155 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1156 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1158 - llogb functions: the llogb, llogbf and llogbl functions, and the
1159 FP_LLOGB0 and FP_LLOGBNAN macros.
1161 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1164 - Comparison macros: iseqsig.
1166 - Classification macros: iscanonical, issubnormal, iszero.
1168 - Total order functions: totalorder, totalorderf, totalorderl,
1169 totalordermag, totalordermagf, totalordermagl.
1171 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1173 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1174 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1176 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1177 are added to libc. They convert a floating-point number into string.
1179 * Most of glibc can now be built with the stack smashing protector enabled.
1180 It is recommended to build glibc with --enable-stack-protector=strong.
1181 Implemented by Nick Alcock (Oracle).
1183 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1184 intended to be used instead of memset() to erase sensitive data after use;
1185 the compiler will not optimize out calls to explicit_bzero even if they
1186 are "unnecessary" (in the sense that no _correct_ program can observe the
1187 effects of the memory clear).
1189 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1190 to float instead of double. This does not affect the ABI of any libraries
1191 that are part of the GNU C Library, but may affect the ABI of other
1192 libraries that use this type in their interfaces.
1194 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1195 float_t and double_t types are now defined to long double instead of float
1196 and double. These options are not the default, and this does not affect
1197 the ABI of any libraries that are part of the GNU C Library, but it may
1198 affect the ABI of other libraries that use this type in their interfaces,
1199 if they are compiled or used with those options.
1201 * The getentropy and getrandom functions, and the <sys/random.h> header file
1204 * The buffer size for byte-oriented stdio streams is now limited to 8192
1205 bytes by default. Previously, on Linux, the default buffer size on most
1206 file systems was 4096 bytes (and thus remains unchanged), except on
1207 network file systems, where the buffer size was unpredictable and could be
1208 as large as several megabytes.
1210 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1211 for the Linux quota interface which predates kernel version 2.4.22 has
1214 * The malloc_get_state and malloc_set_state functions have been removed.
1215 Already-existing binaries that dynamically link to these functions will
1216 get a hidden implementation in which malloc_get_state is a stub. As far
1217 as we know, these functions are used only by GNU Emacs and this change
1218 will not adversely affect already-built Emacs executables. Any undumped
1219 Emacs executables, which normally exist only during an Emacs build, should
1220 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1222 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1223 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1224 “no-ip6-dotint” had already been the default, and support for the
1225 “ip6-dotint” option was removed from the Internet in 2006.
1227 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1228 flag from <resolv.h> have been removed. The option relied on a
1229 backwards-incompatible DNS extension which was never deployed on the
1232 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1233 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1234 They were already unimplemented.
1236 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1237 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1238 removed again from the IETF name lookup interface specification in RFC
1239 2553. Applications should use getaddrinfo instead.
1241 * DNSSEC-related declarations and definitions have been removed from the
1242 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1243 decode the data part of DNSSEC record types. Previous versions of glibc
1244 only implemented minimal support for the previous version of DNSSEC, which
1245 is incompatible with the currently deployed version.
1247 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1248 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1249 <arpa/nameser.h> header file because the distinction between RR types and
1250 meta-RR types is not officially standardized, subject to revision, and
1251 thus not suitable for encoding in a macro.
1253 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1254 and rhook members of the res_state type in <resolv.h> have been removed.
1255 The glibc stub resolver did not support these hooks, but the header file
1256 did not reflect that.
1258 * For multi-arch support it is recommended to use a GCC which has
1259 been built with support for GNU indirect functions. This ensures
1260 that correct debugging information is generated for functions
1261 selected by IFUNC resolvers. This support can either be enabled by
1262 configuring GCC with '--enable-gnu-indirect-function', or by
1263 enabling it by default by setting 'default_gnu_indirect_function'
1264 variable for a particular architecture in the GCC source file
1267 * GDB pretty printers have been added for mutex and condition variable
1268 structures in POSIX Threads. When installed and loaded in gdb these pretty
1269 printers show various pthread variables in human-readable form when read
1270 using the 'print' or 'display' commands in gdb.
1272 * Tunables feature added to allow tweaking of the runtime for an application
1273 program. This feature can be enabled with the '--enable-tunables' configure
1274 flag. The GNU C Library manual has details on usage and README.tunables has
1275 instructions on adding new tunables to the library.
1277 * A new version of condition variables functions have been implemented in
1278 the NPTL implementation of POSIX Threads to provide stronger ordering
1281 * A new version of pthread_rwlock functions have been implemented to use a more
1282 scalable algorithm primarily through not using a critical section anymore to
1285 Security related changes:
1287 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1288 have been created with makecontext could fail to terminate due to a
1289 missing .cantunwind annotation. This has been observed to lead to a hang
1290 (denial of service) in some Go applications compiled with gccgo. Reported
1291 by Andreas Schwab. (CVE-2016-6323)
1293 * The DNS stub resolver functions would crash due to a NULL pointer
1294 dereference when processing a query with a valid DNS question type which
1295 was used internally in the implementation. The stub resolver now uses a
1296 question type which is outside the range of valid question type values.
1299 The following bugs are resolved with this release:
1301 [4099] stdio: Overly agressive caching by stream i/o functions.
1302 [7065] build: Support building glibc with -fstack-protector or -fstack-
1304 [9842] localedata: en_CA: incorrect date format
1305 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1306 before it started waiting
1307 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1308 [15765] nptl: sem_open is wrongly a cancellation point
1309 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1310 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1312 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1313 library linked with pthread
1314 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1315 "FIXME: Ingo" issue)
1316 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1317 [17252] libc: getrandom and getentropy syscall
1318 [17730] malloc: thread-local storage is sometimes improperly free()'d
1319 after being __libc_memalign()'d
1320 [18241] stdio: failed fseek on memstream does not set errno and can fail
1322 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1324 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1325 PTHREAD_PRIO_INHERIT mutex on ARM
1326 [18784] network: res_query and related function crash for special record
1327 type queries (CVE-2015-5180)
1328 [19380] math: strtod does not raise "inexact"
1329 [19387] string: Integer overflow in memchr
1330 [19390] string: Integer overflow in strncat
1331 [19398] build: linknamespace tests fail with massively parallel build
1332 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1334 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1335 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1336 [19514] libc: [PATCH] Fix spelling errors in spelling
1337 "implement"/"implementation" in several places
1338 [19582] network: Deprecate RES_USE_INET6
1339 [19673] manual: clog10 docs appear to be erroneous
1340 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1341 causes a segmentation fault
1342 [19826] libc: invalid pointer returned from __tls_get_addr with static
1344 [20016] network: resolv: Remove hooks support from the API
1345 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1347 [20033] math: [x86_64] vectorized math function don't call the __finite
1349 [20116] nptl: use after free in pthread_create
1350 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1351 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1353 [20311] nptl: please install proc_service.h
1354 [20366] libc: Compilation errors in installed headers in strict-compliance
1356 [20370] malloc: malloc: Arena free list management is still racy
1357 (incorrect fix in bug 19243)
1358 [20386] libc: assert (X = 0) does not result in GCC warning
1359 [20432] malloc: malloc: Minimize interface required for interposition
1360 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1361 in _Unwind_Backtrace (CVE-2016-6323)
1362 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1363 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1365 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1366 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1367 [20477] network: resolv: incorrect double-checked locking related to
1369 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1371 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1373 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1374 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1375 cause transition penalty
1376 [20517] math: sparcv9 missing fdiml compat symbol
1377 [20524] manual: strverscmp is inconsistent
1378 [20525] libc: <sys/quota.h> should be based on kernel headers
1379 [20539] math: GCC 7 -static -lm fails to link at -O3
1380 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1382 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1384 [20591] network: Remove obsolete DNSSEC support
1385 [20592] network: DNS resource record type classification macros in
1386 <arpa/nameser.h> are incorrect
1387 [20593] network: Update DNS RR type definitions
1388 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1389 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1391 [20629] network: libresolv: Remove support for bitstring labels
1393 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1394 assertion: "maxidx >= 2"
1395 [20660] math: [arm] Use VSQRT
1396 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1397 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1398 [20689] libc: Test for FMA should also check for AVX.
1399 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1400 [20715] math: iszero macro breaks existing code
1401 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1402 [20728] libc: powerpc: Missing TOC stub in clone
1403 [20729] build: glibc-2.24 fails to build for i486 with -Os
1404 [20750] build: Build fails with default PIE enabled gcc-6.x
1405 [20768] math: [alpha] sqrt fegetenv namespace
1406 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1408 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1409 __FLT_EVAL_METHOD__ is 2
1410 [20790] network: rpcgen buffer overrun in get_prog_declaration
1411 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1412 [20829] libc: crypt snprintf namespace
1413 [20847] libc: tst-vfork3 failure
1414 [20855] math: Default bits/mathdef.h has inappropriate float_t
1415 [20859] math: [sh4] FP_ILOGB0 invalid
1416 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1417 U+20AC), not same as GBK
1418 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1419 [20916] math: pow handling of sNaN arguments
1420 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1421 [20919] math: Bad pow (qNaN, 0) result with -lieee
1422 [20940] math: hypot sNaN handling
1423 [20947] math: fmax, fmin sNaN handling
1424 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1425 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1427 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1428 [20973] nptl: robust mutexes: Lost wake-ups
1429 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1430 [20978] nis: strlen on null pointer in nss_nisplus
1431 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1433 [21014] string: i686 memchr overflows internal pointer check
1434 [21019] libc: [mips] n32 lseek incorrect on overflow
1435 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1436 [21026] libc: [MIPS] readahead syscall is broken on n64
1437 [21028] math: Fallback fesetexceptflag should always succeed
1438 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1439 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1441 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1442 [21061] librt: [microblaze] librt lost clock_* exports
1443 [21073] libc: tunables: insecure environment variables passed to
1444 subprocesses with AT_SECURE
1445 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1449 * The minimum Linux kernel version that this version of the GNU C Library
1450 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1451 kernel version 2.6.32 or later suffices (on architectures that already
1452 required kernel versions more recent than 3.2, those requirements remain
1453 unchanged). Linux 3.2 or later kernel headers are required on all
1456 * The pap_AN locale has been deleted. This has been deprecated for a long
1457 time. It has been replaced by pap_AW & pap_CW, both of which have long
1458 been included in previous releases.
1460 * The readdir_r and readdir64_r functions have been deprecated. It is
1461 recommended to use readdir and readdir64 instead.
1463 * The type “union wait” has been removed. It was deprecated in the early
1464 1990s and never part of POSIX. Application code should use the int type
1465 instead of “union wait”.
1467 * A new NSS action is added to facilitate large distributed system
1468 administration. The action, MERGE, allows remote user stores like LDAP
1469 to be merged into local user stores like /etc/groups in order to provide
1470 easy to use, updated, and managed sets of merged credentials. The new
1471 action can be used by configuring it in /etc/nsswitch.conf:
1472 group: files [SUCCESS=merge] nis
1473 Implemented by Stephen Gallagher (Red Hat).
1475 * The deprecated __malloc_initialize_hook variable has been removed from the
1478 * The long unused localedef --old-style option has been removed. It hasn't
1479 done anything in over 16 years. Scripts using this option can safely
1482 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1483 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1484 return the next representable value in the direction of positive infinity
1485 and the nextdown functions return the next representable value in the
1486 direction of negative infinity. These are currently enabled as GNU
1489 Security related changes:
1491 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1492 could result in a stack overflow when getnetbyname was called with an
1493 overly long name. (CVE-2016-3075)
1495 * Previously, getaddrinfo copied large amounts of address data to the stack,
1496 even after the fix for CVE-2013-4458 has been applied, potentially
1497 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1498 instead. Reported by Michael Petlan. (CVE-2016-3706)
1500 * The glob function suffered from a stack-based buffer overflow when it was
1501 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1502 Reported by Alexander Cherepanov. (CVE-2016-1234)
1504 * The Sun RPC UDP client could exhaust all available stack space when
1505 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1506 alloca plugin for GCC. (CVE-2016-4429)
1508 * The IPv6 name server management code in libresolv could result in a memory
1509 leak for each thread which is created, performs a failing naming lookup,
1510 and exits. Over time, this could result in a denial of service due to
1511 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1513 The following bugs are resolved with this release:
1515 [1170] localedata: ne_NP: update Nepali locale definition file
1516 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1518 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1519 [6796] math: fdim() does not set errno on overflow
1520 [10354] libc: posix_spawn should use vfork() in more cases than presently
1521 [11213] localedata: localedata: add copyright disclaimer to locale files
1522 [12143] localedata: chr_US: new Cherokee locale
1523 [12450] localedata: sgs_LT: new locale
1524 [12676] localedata: ln_CD: new locale
1525 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1527 [13304] math: fma, fmaf, fmal produce wrong results
1528 [14259] build: --localedir arg to configure is ignored
1529 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1530 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1531 [14934] localedata: es_CL: wrong first weekday chilean locale
1532 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1534 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1536 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1537 [15368] nptl: raise() is not async-signal-safe
1538 [15479] math: ceil, floor, round and trunc raise inexact exception
1539 [15578] localedata: kk_KZ: various updates
1540 [16003] localedata: pap_AN: punt old locale
1541 [16137] localedata: iw_IL: punt old locale
1542 [16190] localedata: eo: new esperanto locale
1543 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1544 [16742] malloc: race condition: pthread_atfork() called before first
1545 malloc() results in unexpected locking behaviour/deadlocks
1546 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1548 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1549 [17565] localedata: pt_PT: wrong (work-)week start
1550 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1551 [17950] build: Build fails with -msse
1552 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1553 [18433] libc: posix_spawn does not return correctly upon failure to
1555 [18453] localedata: charmaps/IBM875: incorrect codes
1556 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1558 [18896] localedata: he_IL: improvements for currency
1559 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1560 Romanian locale data
1561 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1563 [19056] libc: Deprecate readdir_r
1564 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1566 [19198] localedata: nl_NL: small improvements for Dutch locales
1567 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1568 nameservers (CVE-2016-5417)
1569 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1571 [19400] locale: Language missing in "iso-639.def", trivial fix in
1573 [19431] malloc: Deadlock between fflush, getdelim, and fork
1574 [19505] libc: Incorrect file descriptor validity checks in
1575 posix_spawn_file_actions_add{open,close,dup2}
1576 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1577 when using RTLD_NEXT
1578 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1579 `intl/{gettextP,loadinfo}.h'
1580 [19534] libc: execle, execlp may use malloc
1581 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1582 [19573] network: res_nclose and __res_maybe_init disagree about name
1583 server initialization, breaking Hesiod
1584 [19575] localedata: Status of GB18030 tables
1585 [19581] localedata: sr_* date_fmt string contains additional newline
1586 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1588 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1589 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1590 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1591 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1593 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1595 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1597 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1598 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1599 [19633] locale: strfmon_l applies global locale to number formatting
1600 [19642] network: Memory leak in getnameinfo
1601 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1602 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1604 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1605 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1607 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1608 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1610 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1612 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1614 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1615 pointers and lengths in error-case.
1616 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1617 report errors on UTF-16 surrogates values.
1618 [19755] nscd: nscd assertion failure in gc
1619 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1620 [19759] libc: mempcpy shouldn't be inlined
1621 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1622 [19765] libc: s390 needs an optimized mempcpy
1623 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1624 NAME_MAX limit assumption (CVE-2016-1234)
1625 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1626 [19787] network: Missing and incorrect truncation checks in getnameinfo
1627 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1629 [19791] network: Assertion failure in res_query.c with un-connectable name
1631 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1632 [19822] math: libm.so install clobbers old version
1633 [19825] network: resolv: send_vc can return uninitialized data in second
1634 response to getaddrinfo
1635 [19830] network: nss_dns: should check RDATA length against buffer length
1636 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1637 confronted with A/AAAA records of invalid size
1638 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1639 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1640 [19853] stdio: Printing IBM long double in decimal with high precision is
1642 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1643 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1644 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1646 [19865] network: Assertion failure or memory leak in
1647 _nss_dns_getcanonname_r
1648 [19868] network: nss_dns: netent code does not skip over non-PTR records
1649 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1651 [19881] string: Improve x86-64 memset
1652 [19907] string: Incorrect memcpy tests
1653 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1655 [19925] libc: termios.h XCASE namespace
1656 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1657 [19929] libc: limits.h NL_NMAX namespace
1658 [19931] stdio: Memory leak in vfprintf
1659 [19957] libc: clone(CLONE_VM) access invalid parent memory
1660 [19963] localedata: en_IL: New locale
1661 [19989] stdio: stdio.h cuserid namespace
1662 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1664 [19996] locale: langinfo.h nl_langinfo_l namespace
1665 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1666 [20010] network: getaddrinfo: Stack overflow in hostent translation
1668 [20012] stdio: libio: fmemopen append mode failure
1669 [20014] stdio: stdio.h namespace for pre-threads POSIX
1670 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1671 [20023] libc: fcntl.h timespec namespace
1672 [20024] math: [x86_64] vectorized sincos trashes the stack
1673 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1674 [20041] time: sys/time.h timespec namespace
1675 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1676 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1677 [20051] libc: ttyslot in wrong header under wrong conditions
1678 [20054] libc: gethostname not declared for XPG4
1679 [20055] libc: termios.h missing tcgetsid for XPG4
1680 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1682 [20073] libc: sys/stat.h fchmod namespace
1683 [20074] libc: stdlib.h rand_r namespace
1684 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1685 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1687 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1688 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1690 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1691 [20119] libc: Wrong mask for processors level type from CPUID
1692 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1693 AS not supporting AVX512
1694 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1696 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1697 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1699 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1700 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1701 [20178] libc: posix_spawn{p} should not call exit
1702 [20191] stdio: libio: vtables hardening
1703 [20195] string: FMA4 detection requires CPUID execution with register
1705 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1706 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1708 [20212] math: acos (sNaN) returns sNaN
1709 [20213] math: asin (sNaN) returns sNaN
1710 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1711 [20218] math: [i386] asinhl (sNaN) returns sNaN
1712 [20219] math: [i386] atanhl (sNaN) returns sNaN
1713 [20222] stdio: fopencookie: Mangle function pointers
1714 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1715 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1716 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1718 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1719 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1720 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1721 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1722 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1723 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1724 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1725 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1726 [20240] math: modf (sNaN) returns sNaN
1727 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1728 [20250] math: frexp (sNaN) returns sNaN
1729 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1730 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1732 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1733 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1734 subscript is below array bounds [-Werror=array-bounds]
1735 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1736 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1738 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1739 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1741 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1742 "invalid" exceptions
1743 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1744 [20316] localedata: id_ID: Februari instead of Pebruari
1745 [20327] string: POWER8 strcasecmp returns incorrect result
1746 [20347] math: Failure: Test: j0_downward (0xap+0)
1747 [20348] libc: FAIL: misc/tst-preadvwritev64
1748 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1749 [20350] libc: There is no test for p{read,write}64
1750 [20357] math: Incorrect cos result for 1.5174239687223976
1751 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1755 * Unicode 8.0.0 Support: Character encoding, character type info, and
1756 transliteration tables are all updated to Unicode 8.0.0, using new
1757 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1758 These updates cause user visible changes, such as the fixes for bugs
1759 89, 16061, and 18568.
1761 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1762 kernel-internal CPU set size. This means that requests that change the
1763 CPU affinity which failed before (for example, an all-ones CPU mask) will
1764 now succeed. Applications that need to determine the effective CPU
1765 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1766 setting it because the kernel can adjust it (and the previous size check
1767 would not detect this in the majority of cases).
1769 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1770 the following new symbols are used: fts64_children, fts64_close,
1771 fts64_open, fts64_read and fts64_set.
1773 * getaddrinfo now detects certain invalid responses on an internal netlink
1774 socket. If such responses are received, an affected process will
1775 terminate with an error message of "Unexpected error <number> on netlink
1776 descriptor <number>" or "Unexpected netlink response of size <number> on
1777 descriptor <number>". The most likely cause for these errors is a
1778 multi-threaded application which erroneously closes and reuses the netlink
1779 file descriptor while it is used by getaddrinfo.
1781 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1782 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1783 unnecessary serialization of memory allocation requests across threads.
1784 The defect is now corrected. Users should see a substantial increase in
1785 the concurent throughput of allocation requests for applications which
1786 trigger this bug. Affected applications typically create and
1787 destroy threads frequently. (Bug 19048 was reported and analyzed by
1790 * There is now a --disable-timezone-tools configure option for disabling the
1791 building and installing of the timezone related utilities (zic, zdump, and
1792 tzselect). This is useful for people who build the timezone data and code
1793 independent of the GNU C Library.
1795 * The obsolete header <regexp.h> has been removed. Programs that require
1796 this header must be updated to use <regex.h> instead.
1798 * The obsolete functions bdflush, create_module, get_kernel_syms,
1799 query_module and uselib are no longer available to newly linked binaries;
1800 the header <sys/kdaemon.h> has been removed. These functions and header
1801 were specific to systems using the Linux kernel and could not usefully be
1802 used with the GNU C Library on systems with version 2.6 or later of the
1805 * Optimized string, wcsmbs and memory functions for IBM z13.
1806 Implemented by Stefan Liebler.
1808 * Newly linked programs that define a variable called signgam will no longer
1809 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1810 require signgam to be set by those functions must ensure that they use the
1811 variable provided by the GNU C Library and declared in <math.h>, without
1812 defining their own copy.
1814 * The minimum GCC version that can be used to build this version of the GNU
1815 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1816 still be used to compile programs using the GNU C Library.
1818 Security related changes:
1820 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1821 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1823 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1824 the pointer guard feature. It is always enabled. Previously,
1825 LD_POINTER_GUARD could be used to disable security hardening in binaries
1826 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1829 * An integer overflow in hcreate and hcreate_r could lead to an
1830 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1832 * The catopen function no longer has unbounded stack usage. Reported by
1833 Max. (CVE-2015-8779)
1835 * The nan, nanf and nanl functions no longer have unbounded stack usage
1836 depending on the length of the string passed as an argument to the
1837 functions. Reported by Joseph Myers. (CVE-2014-9761)
1839 * A stack-based buffer overflow was found in libresolv when invoked from
1840 libnss_dns, allowing specially crafted DNS responses to seize control
1841 of execution flow in the DNS client. The buffer overflow occurs in
1842 the functions send_dg (send datagram) and send_vc (send TCP) for the
1843 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1844 family. The use of AF_UNSPEC triggers the low-level resolver code to
1845 send out two parallel queries for A and AAAA. A mismanagement of the
1846 buffers used for those queries could result in the response of a query
1847 writing beyond the alloca allocated buffer created by
1848 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1849 the overflow. Thanks to the Google Security Team and Red Hat for
1850 reporting the security impact of this issue, and Robert Holiday of
1851 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1853 The following bugs are resolved with this release:
1855 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1856 [887] math: Math library function "logb" and "nextafter" inconsistent
1857 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1858 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1859 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1860 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1862 [4404] localedata: German translation of "Alarm clock" is misleading
1863 [6799] math: nextafter() and nexttoward() doen't set errno on
1864 overflow/underflow errors
1865 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1867 [10432] nis: _nss_nis_setnetgrent assertion failure
1868 [11460] libc: fts has no LFS support
1869 [12926] network: getaddrinfo()/make_request() may spin forever
1870 [13065] nptl: Race condition in pthread barriers
1871 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1872 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1874 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1875 [14912] libc: Rename non-installed bits/*.h headers
1876 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1877 [15367] math: Let gcc use __builtin_isinf
1878 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1879 [15421] math: lgamma wrongly sets signgam for ISO C
1880 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1881 FE_INVALID with argument out of range
1882 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1884 [15786] dynamic-link: ifunc resolver functions can smash function
1886 [15918] math: Unnecessary check for equality in hypotf()
1887 [16061] localedata: Review / update transliteration data
1888 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1890 [16141] time: strptime %z offset restriction
1891 [16171] math: drem should be alias of remainder
1892 [16296] math: fegetround is pure?
1893 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1894 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1896 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1897 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1899 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1900 "Exception "Inexact" set" on ppc32
1901 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1902 [16517] math: Missing underflow exception from tanf/tan/tanl
1903 [16519] math: Missing underflow exception from sinhf
1904 [16520] math: Missing underflow exception from tanhf
1905 [16521] math: Missing underflow exception from exp2
1906 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1908 [16734] stdio: fopen calls mmap to allocate its buffer
1909 [16961] math: nan function incorrect handling of bad sequences
1910 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1911 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1912 [16985] locale: localedef: confusing error message when opening output
1914 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1915 [17197] locale: Redundant shift character in iconv conversion output at
1917 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1918 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1919 [17250] dynamic-link: static linking breaks nss loading
1920 (getaddrinfo/getpwnam/etc...)
1921 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1923 [17441] math: isnan() should use __builtin_isnan() in GCC
1924 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1925 (related to lock elision)
1926 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1927 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1928 [17887] time: strptime should be able to parse "+01:00" style timezones
1929 [17905] libc: catopen() Multiple unbounded stack allocations
1931 [18084] libc: backtrace (..., 0) dumps core on x86
1932 [18086] libc: nice() sets errno to 0 on success
1933 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1934 is too large (CVE-2015-8778)
1935 [18251] dynamic-link: SONAME missing when audit modules provides path
1936 [18265] libc: add attributes for wchar string and memory functions
1937 [18370] math: csqrt missing underflows
1938 [18421] libc: [hppa] read-only segment has dynamic relocations
1939 [18472] libc: Obsolete syscall wrappers should be compat symbols
1940 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1941 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1942 [18525] localedata: Remove locale timezone information
1943 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1944 [18568] localedata: Update locale data to Unicode 8.0
1945 [18589] locale: sort-test.sh fails at random
1946 [18595] math: ctan, ctanh missing underflows
1947 [18604] libc: assert macro-expands its argument
1948 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1949 contains a vector instruction exception.
1950 [18611] math: j1, jn missing errno setting on underflow
1951 [18618] localedata: sync Chechen locale definitions with other *_RU
1953 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1954 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1955 [18665] network: In send_dg, the recvfrom function is NOT always using the
1956 buffer size of a newly created buffer (CVE-2015-7547)
1957 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1959 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1961 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1962 [18699] math: tilegx cproj() for various complex infinities does not yield
1964 [18724] libc: Harden put*ent functions against data injection
1965 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1966 [18755] build: build errors with -DNDEBUG
1967 [18757] stdio: fmemopen fails to set errno on failure
1968 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1970 [18781] libc: openat64 lacks O_LARGEFILE
1971 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1972 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1973 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1974 [18790] math: [ldbl-128ibm] tanhl inaccurate
1975 [18795] libc: stpncpy fortification misses buffer lengths that are
1976 statically too large
1977 [18796] build: build fails for --disable-mathvec
1978 [18803] math: hypot missing underflows
1979 [18820] stdio: fmemopen may leak memory on failure
1980 [18823] math: csqrt spurious underflows
1981 [18824] math: fma spurious underflows
1982 [18825] math: pow missing underflows
1983 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1984 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1985 [18870] build: sem_open.c fails to compile with missing symbol
1987 [18872] stdio: Fix memory leak in printf_positional
1988 [18873] libc: posix_fallocate overflow check ineffective
1989 [18875] math: Excess precision leads incorrect libm
1990 [18877] libc: arm: mmap offset regression
1991 [18887] libc: memory corruption when using getmntent on blank lines
1992 [18918] localedata: hu_HU: change time to HH:MM:SS format
1993 [18921] libc: Regression: extraneous stat() and fstat() performed by
1995 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1996 binaries (CVE-2015-8777)
1997 [18951] math: tgamma missing underflows
1998 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2000 [18953] localedata: lt_LT: change currency symbol to the euro
2001 [18956] math: powf inaccuracy
2002 [18961] math: [i386] exp missing underflows
2003 [18966] math: [i386] exp10 missing underflows
2004 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2005 [18969] build: multiple string test failures due to missing locale
2007 [18970] libc: Reference of pthread_setcancelstate in libc.a
2008 [18977] math: float / long double Bessel functions not in XSI POSIX
2009 [18980] math: i386 libm functions return with excess range and precision
2010 [18981] math: i386 scalb*, ldexp return with excess range and precision
2011 [18982] stdio: va_list and vprintf
2012 [18985] time: Passing out of range data to strftime() causes a segfault
2014 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2015 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2016 [19012] locale: iconv_open leaks memory on error path
2017 [19016] math: clog, clog10 inaccuracy
2018 [19018] nptl: Mangle function pointers in tls_dtor_list
2019 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2020 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2021 [19048] malloc: malloc: arena free list can become cyclic, increasing
2023 [19049] math: [powerpc] erfc incorrect zero sign
2024 [19050] math: [powerpc] log* incorrect zero sign
2025 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2026 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2027 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2028 [19074] network: Data race in _res_hconf_reorder_addrs
2029 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2030 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2031 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2033 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2035 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2036 [19086] manual: posix_fallocate64 documented argument order is wrong.
2037 [19088] math: lround, llround missing exceptions close to overflow
2039 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2040 [19095] math: dbl-64 lrint incorrect for 64-bit long
2041 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2042 [19124] dynamic-link: ld.so failed to build with older assmebler
2043 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2044 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2045 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2046 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2047 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2048 pthread_setaffinity_np
2049 [19156] math: [ldbl-128] j0l spurious underflows
2050 [19164] nptl: tst-getcpu fails with many possible CPUs
2051 [19168] math: math/test-ildoubl and math/test-ldouble failure
2052 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2053 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2055 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2056 (FE_NOMASK_ENV) do not clear SSE exceptions
2057 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2058 _int_new_arena/reused_arena
2059 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2060 [19201] math: dbl-64 remainder incorrect sign of zero result
2061 [19205] math: bits/math-finite.h conditions do not match math.h and
2063 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2064 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2065 for C99-based standards
2066 [19212] libc: features.h not -Wundef clean
2067 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2069 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2070 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2071 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2073 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2074 "inexact" exceptions
2075 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2077 [19242] libc: strtol incorrect in Turkish locales
2078 [19243] malloc: reused_arena can pick an arena on the free list, leading
2079 to an assertion failure and reference count corruption
2080 [19253] time: tzset() ineffective when temporary TZ did not include DST
2082 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2083 [19270] math: [hppa] Shared libm missing __isnanl
2084 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2085 MAP_HUGETLB and MAP_STACK defines
2086 [19313] nptl: Wrong __cpu_mask for x32
2087 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2088 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2089 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2090 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2091 [19363] time: x32: times() return value wrongly truncates/sign extends
2093 [19367] dynamic-link: Improve branch prediction on Silvermont
2094 [19369] network: Default domain name not reset by res_ninit when "search"
2095 / "domain" entry is removed from resolv.conf
2096 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2098 [19385] localedata: bg_BG: time separator should be colon, not comma
2099 [19408] libc: linux personality syscall wrapper may erroneously return an
2100 error on 32-bit architectures
2101 [19415] libc: dladdr returns wrong names on hppa
2102 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2103 IBM905, IBM907, and IBM909
2104 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2105 [19443] build: build failures with -DDEBUG
2106 [19451] build: Make check fails on test-double-vlen2
2107 [19462] libc: Glibc failed to build with -Os
2108 [19465] math: Wrong code with -Os
2109 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2111 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2113 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2114 [19486] math: S390: Math tests fail with "Exception Inexact set".
2115 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2116 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2118 [19590] math: Fail to build shared objects that use libmvec.so functions.
2122 * The following bugs are resolved with this release:
2124 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2125 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2126 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2127 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2128 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2129 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2130 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2131 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2132 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2133 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2134 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2135 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2136 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2137 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2138 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2139 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2140 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2141 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2142 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2143 18657, 18676, 18694, 18696, 18887.
2145 * Cache information can be queried via sysconf() function on s390 e.g. with
2146 _SC_LEVEL1_ICACHE_SIZE as argument.
2148 * A buffer overflow in gethostbyname_r and related functions performing DNS
2149 requests has been fixed. If the NSS functions were called with a
2150 misaligned buffer, the buffer length change due to pointer alignment was
2151 not taken into account. This could result in application crashes or,
2152 potentially arbitrary code execution, using crafted, but syntactically
2153 valid DNS responses. (CVE-2015-1781)
2155 * The time zone file parser has been made more robust against crafted time
2156 zone files, avoiding heap buffer overflows related to the processing of
2157 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2158 large time zone data files. Overly long time zone specifiers in the TZ
2159 variable no longer result in stack overflows and crashes.
2161 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2162 for LD and GD on x86 and x86-64, has been implemented. You will need
2163 binutils-2.24 or later to enable this optimization.
2165 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2166 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2167 Hat). These updates cause user visible changes, such as the fix for bug
2170 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2171 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2172 condition in some applications.
2174 * Added vector math library named libmvec with the following vectorized x86_64
2175 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2177 The library can be disabled with --disable-mathvec. Use of the functions is
2178 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2179 Shared library libmvec.so is linked in as needed when using -lm (no need to
2180 specify -lmvec explicitly for not static builds).
2181 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2183 * A new fmemopen implementation has been added with the goal of POSIX
2184 compliance. The new implementation fixes the following long-standing
2185 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2186 old implementation is still present for use be by existing binaries.
2188 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2189 releases. It has been fixed to match 2.19 and older, but binaries built
2190 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2192 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2193 Contributed by Roland McGrath (Google).
2195 * The header <regexp.h> is deprecated, and will be removed in a future
2196 release. Use of this header will trigger a deprecation warning.
2197 Application developers should update their code to use <regex.h> instead.
2199 This header was formerly part of SUS, but was deprecated in 1994 and
2200 removed from the standard in 2001. Also, the glibc implementation
2201 leaks memory. See BZ#18681 for more details.
2205 * The following bugs are resolved with this release:
2207 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2208 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2209 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2210 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2211 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2212 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2213 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2214 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2215 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2216 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2219 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2220 too little memory for the to-be-scanned arguments and overflow the
2221 allocated buffer. The implementation now correctly computes the required
2222 buffer size when using malloc, and switches to malloc from alloca as
2225 * A new semaphore algorithm has been implemented in generic C code for all
2226 machines. Previous custom assembly implementations of semaphore were
2227 difficult to reason about or ensure that they were safe. The new version
2228 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2229 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2230 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2232 * Port to Altera Nios II has been contributed by Mentor Graphics.
2234 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2235 implementations for powerpc64/powerpc64le.
2236 Implemented by Adhemerval Zanella (IBM).
2238 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2239 and powerpc64le. This may improve lock scaling of existing programs on
2240 HTM capable systems. The lock elision code is only enabled with
2241 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2242 powerpc will issue a transaction abort on every syscall to avoid side
2243 effects being visible outside transactions.
2245 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2246 AArch64. Contributed by ARM Ltd.
2248 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2250 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2251 under certain input conditions resulting in the execution of a shell for
2252 command substitution when the application did not request it. The
2253 implementation now checks WRDE_NOCMD immediately before executing the
2254 shell and returns the error WRDE_CMDSUB as expected.
2256 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2257 processing format strings with a large number of format specifiers.
2259 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2260 infinite loop if the DNS response contained a PTR record of an unexpected
2263 * The minimum GCC version that can be used to build this version of the GNU
2264 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2265 still be used to compile programs using the GNU C Library.
2267 * The GNU C Library is now built with -Werror by default. This can be
2268 disabled by configuring with --disable-werror.
2270 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2272 * The obsolete sigvec function has been removed. This was the original
2273 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2274 programs have been using instead for about 25 years. Of course, ABI
2275 compatibility for old binaries using sigvec remains intact.
2277 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2278 with newer versions of bison.
2280 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2281 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2282 registers overlay two consecutive single-precision registers. MIPS32R2
2283 introduced a new FPU mode (FR=1) where double-precision registers extend the
2284 corresponding single-precision registers which is incompatible with the
2285 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2286 require the use of FR=1 making a transition necessary. New o32 ABI
2287 extensions enable users to migrate over time from the original o32 ABI
2288 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2289 tracks the ABI of any loaded object and verifies that new objects are
2290 compatible. Mode transitions will also be requested as required and
2291 unsupportable objects will be rejected. The ABI checks include both soft and
2292 hard float ABIs for o32, n32 and n64.
2294 GCC 5 with GNU binutils 2.25 onwards:
2295 It is strongly recommended that all o32 system libraries are built using the
2296 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2297 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2298 compiler using --with-fp-32=xx to set this by default.
2302 * The following bugs are resolved with this release:
2304 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2305 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2306 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2307 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2308 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2309 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2310 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2311 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2312 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2313 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2314 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2315 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2316 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2317 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2318 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2320 * Reverted change of ABI data structures for s390 and s390x:
2321 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2322 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2323 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2324 are preserved pointing straight to the same implementation as the old ones.
2325 Given that, new callers will simply provide a too-big buffer to these
2326 functions. Any applications/libraries out there that embed jmp_buf or
2327 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2328 against 2.19 headers will have to rebuilt again. This is necessary in any
2329 case to revert the breakage in their ABI caused by the glibc change.
2331 * Support for file description locks is added to systems running the
2332 Linux kernel. The standard file locking interfaces are extended to
2333 operate on file descriptions, not file descriptors, via the use of
2334 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2335 are associated with an open file instead of a process.
2337 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2339 * The minimum Linux kernel version that this version of the GNU C Library
2340 can be used with is 2.6.32.
2342 * Running the testsuite no longer terminates as soon as a test fails.
2343 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2344 with PASS or FAIL lines for individual tests. A summary of the results is
2345 printed, including a list of failing lists, and "make check" exits with
2346 error status if there were any unexpected failures. "make check
2347 stop-on-test-failure=y" may be used to keep the old behavior.
2349 * The am33 port, which had not worked for several years, has been removed
2352 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2353 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2354 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2355 conflicted with POSIX (support for which was removed in 2.19), the
2356 interfaces those macros enabled remain available when compiling with
2357 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2358 test macros defined.
2360 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2362 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2363 This may improve lock scaling of existing programs on TX capable systems.
2364 The lock elision code is only built with --enable-lock-elision=yes and
2365 then requires a GCC version supporting the TX builtins. With lock elision
2366 default mutexes are elided via __builtin_tbegin, if the cpu supports
2367 transactions. By default lock elision is not enabled and the elision code
2370 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2371 copy the path argument. This allowed programs to cause posix_spawn to
2372 deference a dangling pointer, or use an unexpected pathname argument if
2373 the string was modified after the posix_spawn_file_actions_addopen
2376 * All supported architectures now use the main glibc sysdeps directory
2377 instead of some being in a separate "ports" directory (which was
2378 distributed separately before glibc 2.17).
2380 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2381 On configurations that support it (all Linux configurations), it's now
2382 used regardless of the --enable-add-ons switch to configure. It is no
2383 longer possible to build such configurations without pthreads support.
2385 * Locale names, including those obtained from environment variables (LANG
2386 and the LC_* variables), are more tightly checked for proper syntax.
2387 setlocale will now fail (with EINVAL) for locale names that are overly
2388 long, contain slashes without starting with a slash, or contain ".." path
2389 components. (CVE-2014-0475) Previously, some valid locale names were
2390 silently replaced with the "C" locale when running in AT_SECURE mode
2391 (e.g., in a SUID program). This is no longer necessary because of the
2394 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2395 with application code using Intel MPX instructions. (With all previous
2396 versions, the MPX register state could be clobbered when making calls
2397 into or out of a shared library.) Note that while the new dynamic
2398 linker is compatible with all known x86 hardware whether or not it
2399 supports Intel MPX, some x86 instruction-set emulators might fail to
2400 handle the new instruction encodings. This is known to affect Valgrind
2401 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2402 release), and might affect other tools that do instruction emulation.
2404 * Support for loadable gconv transliteration modules has been removed.
2405 The support for transliteration modules has been non-functional for
2406 over a decade, and the removal is prompted by security defects. The
2407 normal gconv conversion modules are still supported. Transliteration
2408 with //TRANSLIT is still possible, and the //IGNORE specifier
2409 continues to be supported. (CVE-2014-5119)
2411 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2412 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2413 resulting a denial-of-service security vulnerability in applications which
2414 use functions related to iconv. (CVE-2014-6040)
2418 * The following bugs are resolved with this release:
2420 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2421 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2422 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2423 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2424 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2425 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2426 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2427 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2428 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2429 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2430 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2431 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2432 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2433 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2434 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2435 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2436 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2437 16431, 16453, 16474, 16506, 16510, 16529
2439 * Slovenian translations for glibc messages have been contributed by the
2440 Translation Project's Slovenian team of translators.
2442 * The public headers no longer use __unused nor __block. This change is to
2443 support compiling programs that are derived from BSD sources and use
2444 __unused internally, and to support compiling with Clang's -fblock
2445 extension which uses __block.
2447 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2448 large collation sequences to optimize multiple passes. This cache
2449 computation may overflow for large collation sequences and may cause a
2450 stack or buffer overflow. This is now fixed to use a slower algorithm
2451 which does not use a cache if there is an integer overflow.
2453 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2454 rules for large collation sequences to optimize multiple passes and falls
2455 back to alloca if malloc fails, resulting in a possible stack overflow.
2456 The implementation now falls back to an uncached collation sequence lookup
2459 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2460 initialized for static applications resulting in the security feature
2461 being disabled. The pointer guard is now correctly initialized to a
2462 random value for static applications. Existing static applications need
2463 to be recompiled to take advantage of the fix (bug 15754).
2465 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2466 to the d_name member of struct dirent, or omit the terminating NUL
2467 character. (Bugzilla #14699).
2469 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2470 aligned_alloc functions could allocate too few bytes or corrupt the
2471 heap when passed very large allocation size values (Bugzilla #15855,
2474 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2475 for AF_INET6 has been fixed (Bugzilla #16072).
2477 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2478 pap_CW, quz_PE, the_NP.
2480 * Substantially revised locales: gd_GB, ht_HT
2482 * The LC_ADDRESS field was updated to support country_car for almost all
2485 * ISO 1427 definitions were updated.
2487 * ISO 3166 definitions were updated.
2489 * The localedef utility now supports --big-endian and --little-endian
2490 command-line options to generate locales for a different system from that
2491 for which the C library was built.
2493 * Binary locale files now only depend on the endianness of the system for
2494 which they are generated and not on other properties of that system. As a
2495 consequence, binary files generated with new localedef may be incompatible
2496 with old versions of the GNU C Library, and binary files generated with
2497 old localedef may be incompatible with this version of the GNU C Library,
2498 in the following circumstances:
2500 + Locale files may be incompatible on m68k systems.
2502 + Locale archive files (but not separate files for individual locales) may
2503 be incompatible on systems where plain "char" is signed.
2505 * The configure option --disable-versioning has been removed. Builds with
2506 --disable-versioning had not worked for several years.
2508 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2509 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2511 * SystemTap probes for malloc have been introduced.
2513 * SystemTap probes for slow multiple precision fallback paths of
2514 transcendental functions have been introduced.
2516 * Support for powerpc64le has been added.
2518 * The soft-float powerpc port now supports e500 processors.
2520 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2522 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2523 set of header declarations that are enabled by default, even when other
2524 feature test macros or compiler options such as -std=c99 would otherwise
2525 disable some of those declarations.
2527 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2528 conflict with POSIX. The libbsd-compat library (which was a dummy library
2529 that did nothing) has also been removed.
2531 * Preliminary documentation about Multi-Thread, Async-Signal and
2532 Async-Cancel Safety has been added.
2534 * Change of ABI data structures for s390 and s390x:
2535 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2536 allow for future hardware extensions. All library functions that accept or
2537 return these structures were versioned in order to provide backwards
2538 compatibility with existing code. However, not all cases can be handled
2539 with interface versioning. If an application embeds either structure into
2540 another structure and that structure is passed to another compilation unit
2541 which expects the newer larger-sized object then any access to the new
2542 fields will result in undefined behaviour. Similarly any access to fields
2543 that were moved by the enlarging of the structures will cause undefined
2544 behaviour. To fix the undefined behaviour all objects that are part of the
2545 final in-memory image must agree on the size of structures, and this may
2546 require recompilation.
2550 * The following bugs are resolved with this release:
2552 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2553 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2554 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2555 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2556 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2557 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2558 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2559 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2560 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2561 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2562 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2563 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2564 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2567 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2568 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2569 Distributions can re-enable building and using pt_chown via the new configure
2570 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2571 considerable security risks and should only be used if the distribution
2572 understands and accepts the risks.
2574 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2577 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2578 fixed (Bugzilla #15330).
2580 * Add support for calling C++11 thread_local object destructors on thread
2581 and program exit. This needs compiler support for offloading C++11
2582 destructor calls to glibc.
2584 * Improved worst case performance of libm functions with double inputs and
2587 * Support for priority inherited mutexes in pthread condition variables on
2588 non-x86 architectures.
2590 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2592 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2594 * Optimized string functions for ARM. Implemented by Will Newton and
2597 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2599 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2602 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2603 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2605 * Added a benchmark framework to track performance of functions in glibc.
2607 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2608 It is based on draft TS 18661 and currently enabled as a GNU extension.
2610 * On Linux, the clock function now uses the clock_gettime system call
2611 for improved precision, rather than old times system call.
2613 * Added support for version-3 tz format files. This is needed when using
2614 the tz database release 2013e or later, and affects a few unusual cases --
2615 currently only TZ='America/Godthab' for time stamps after 2037.
2617 * Added new API functions pthread_getattr_default_np and
2618 pthread_setattr_default_np to get and set the default pthread
2619 attributes of a process.
2621 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2622 This may improve lock scaling of existing programs on TSX capable systems.
2623 When the --enable-lock-elision=yes parameter is specified at configure
2624 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2627 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2628 availability). Implemented by Ryan S. Arnold.
2630 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2634 * The following bugs are resolved with this release:
2636 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2637 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2638 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2639 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2640 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2641 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2642 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2643 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2644 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2645 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2646 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2647 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2648 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2649 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2650 14879, 14889, 14893, 14898, 14914.
2652 * Optimization of memcpy for MIPS.
2654 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2655 EMFILE has been fixed (Bugzilla #14889).
2657 * The add-on ports collection is now distributed in the "ports" subdirectory
2658 of the main GNU C Library distribution, rather than separately.
2660 * Port to ARM AArch64 contributed by Linaro.
2662 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2663 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2665 Implemented by Andreas Krebbel.
2667 * The new function secure_getenv allows secure access to the environment,
2668 returning NULL if running in a SUID/SGID process. This function replaces
2669 the internal function __secure_getenv.
2671 * SystemTap static probes have been added into the dynamic linker.
2672 Implemented by Gary Benson.
2674 * Optimizations of string functions strstr, strcasestr and memmem.
2675 Implemented by Maxim Kuvyrkov.
2677 * The minimum Linux kernel version that this version of the GNU C Library
2678 can be used with is 2.6.16.
2680 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2681 powerpc POWER7. Implemented by Will Schmidt.
2683 * New configure option --disable-nscd builds the C library such that it
2684 never attempts to contact the Name Service Caching Daemon (nscd).
2685 New configure option --disable-build-nscd avoids building nscd itself;
2686 this is the default if --disable-nscd is used.
2688 * Improved support for cross-compilation, including cross-testing and
2689 bootstrap builds without a previously built glibc.
2691 * Several testsuite tests are now able to test multiple IFUNC variants of an
2692 interface, rather than just testing the one that would be chooen by
2695 * New configure options --with-bugurl and --with-pkgversion, for
2696 distributors to use to embed their bug-reporting and package version
2697 information in --help and --version output.
2699 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2700 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2701 allows creation of chroots without the procfs mounted on /proc.
2703 * The `crypt' function now fails if passed salt bytes that violate the
2704 specification for those values. On Linux, the `crypt' function will
2705 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2706 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2707 when the mode is enabled.
2709 * The `clock_*' suite of functions (declared in <time.h>) is now available
2710 directly in the main C library. Previously it was necessary to link with
2711 -lrt to use these functions. This change has the effect that a
2712 single-threaded program that uses a function such as `clock_gettime' (and
2713 is not linked with -lrt) will no longer implicitly load the pthreads
2714 library at runtime and so will not suffer the overheads associated with
2715 multi-thread support in other code such as the C++ runtime library.
2717 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2722 * The following bugs are resolved with this release:
2724 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2725 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2726 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2727 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2728 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2729 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2730 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2731 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2732 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2733 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2734 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2735 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2736 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2737 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2738 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2739 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2740 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2741 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2742 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2743 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2744 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2745 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2748 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2749 configuring glibc with:
2750 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2751 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2752 Implemented by H.J. Lu.
2756 + define static_assert
2758 + do not declare gets
2760 + declare at_quick_exit and quick_exit also for ISO C11
2762 + aligned_alloc. NB: The code is deliberately allows the size parameter
2763 to not be a multiple of the alignment. This is a moronic requirement
2764 in the standard but it is only a requirement on the caller, not the
2767 + timespec_get added
2769 + uchar.h support added
2771 + CMPLX, CMPLXF, CMPLXL added
2773 Implemented by Ulrich Drepper.
2775 * Support for the IA-64 has been moved to ports.
2777 * Remove support for anything but ELF binary format
2779 * Checking versions of poll, ppoll added.
2780 Implemented by Ulrich Drepper.
2782 * More generic and 64-bit performance optimizations to math functions.
2783 Implemented by Ulrich Drepper.
2785 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2786 headers and functions available at compile time as they were before
2787 version 2.14. This option will be removed at some time in the future
2788 after the TI-RPC library becomes fully sufficient for the needs of
2789 existing applications.
2791 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2792 Note that glibc is not expected to work with any Linux kernel version
2795 * New header <sys/auxv.h> and function getauxval allowing easy access to
2796 the AT_* key-value pairs passed from the Linux kernel. The header also
2797 defines the HWCAP_* bits associated with the AT_HWCAP key.
2799 * New locales: mag_IN
2801 * New configure option --enable-systemtap builds SystemTap static probes
2802 into libc for setjmp and longjmp and into libpthread for various operations.
2803 So far the setjmp/longjmp probes and some of the libpthread probes are
2804 provided only for i*86 and x86_64.
2805 Implemented by Roland McGrath and Rayson Ho.
2807 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2809 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2812 * More optimized functions for SPARC. Implemented by David S. Miller.
2814 * Improved support for cross-compilation, especially bootstrap builds
2815 without a previously built glibc.
2817 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2818 Chris Metcalf from Tilera.
2820 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2821 now supported for ARM processors.
2823 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2824 as the name of the dynamic linker, to distinguish it from the
2825 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2827 * Support for CRIS has been removed from ports.
2829 * A new class of installed header has been documented for low-level
2830 platform-specific functionality. PowerPC added the first instance with a
2831 function to provide time base register access. Contributed by Tulio
2832 Magno Quites Machado Filho.
2834 * ix86 configurations now install header files that are consistent with
2835 what x86-64 configurations install. These same header files can be used
2836 for -m32, -m64, or -mx32 builds.
2837 Contributed by H.J. Lu.
2839 * Math library bug fixes. A thorough audit of all open math library bugs was
2840 conducted by Joseph Myers. Significant progress was made on many math
2841 library bugs resulting in more accurate exceptions and function results.
2842 Many thanks to all those that contributed including Andreas Jaeger for his
2843 patch review and work on the x87 trigonometric instruction issues.
2845 * Timezone data is no longer installed. Timezone-related binaries and scripts
2846 will continue to be installed. Users should obtain their timezone data from
2847 their distribution provider or from the tzdata package at
2848 <ftp://munnari.oz.au/pub/>.
2852 * The following bugs are resolved with this release:
2854 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2855 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2856 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2857 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2858 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2859 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2860 13484, 13506, 13515, 13523, 13524, 13538, 13540
2862 * New program pldd to list loaded object of a process
2863 Implemented by Ulrich Drepper.
2865 * Add nss_db support back to glibc. No more dependency on Berkeley db
2866 and support for initgroups lookups.
2867 Implemented by Ulrich Drepper.
2869 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2870 Contributed by HJ Lu.
2872 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2873 Contributed by HJ Lu.
2875 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2876 on x86-32 and x86-64.
2877 Contributed by Liubov Dmitrieva.
2879 * Optimized strchr and strrchr for SSE on x86-32.
2880 Contributed by Liubov Dmitrieva.
2882 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2883 for x86-64 and x86-32.
2884 Contributed by Liubov Dmitrieva.
2886 * New interfaces: scandirat, scandirat64
2887 Implemented by Ulrich Drepper.
2889 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2890 Implemented by Ulrich Drepper.
2892 * nscd now also caches the netgroup database.
2893 Implemented by Ulrich Drepper.
2895 * Integrate libm with gcc's -ffinite-math-only option.
2896 Implemented by Ulrich Drepper.
2898 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2899 to math functions. Implemented by Ulrich Drepper.
2901 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2902 Implemented by Ulrich Drepper.
2904 * New Linux interfaces: process_vm_readv, process_vm_writev
2906 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2907 Implemented by Ulrich Drepper.
2909 * Optimized nearbyint and strcasecmp for PPC.
2910 Implemented by Adhemerval Zanella.
2912 * New locales: bho_IN, unm_US, es_CU, ta_LK
2916 * The following bugs are resolved with this release:
2918 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2919 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2920 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2921 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2922 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2923 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2924 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2925 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2926 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2927 12795, 12811, 12813, 12814, 12841
2929 * The RPC implementation in libc is obsoleted. Old programs keep working
2930 but new programs cannot be linked with the routines in libc anymore.
2931 Programs in need of RPC functionality must be linked against TI-RPC.
2932 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2934 Visible changes of this change include (obviously) the inability to link
2935 programs using RPC functions without referencing the TI-RPC library and the
2936 removal of the RPC headers from the glibc headers.
2937 Implemented by Ulrich Drepper.
2939 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2940 syncfs, setns, sendmmsg
2942 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2943 yue_HK, lij_IT, mhr_RU
2945 * New encodings: CP770, CP771, CP772, CP773, CP774
2947 * New program sotruss to trace calls through PLTs
2948 Implemented by Ulrich Drepper.
2950 * The malloc hook implementation is marked deprecated and will be removed
2951 from the default implementation in the next version. The design never
2952 worked ever since the introduction of threads. Even programs which do
2953 not create threads themselves can use multiple threads created internally.
2957 * The following bugs are resolved with this release:
2959 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2960 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2961 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2962 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2965 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2967 * POWER7 optimizations: memset, memcmp, strncmp
2969 * New optimized string functions for x86-64: strnlen (SSE2),
2970 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2971 Implemented by Ulrich Drepper.
2975 * The following bugs are resolved with this release:
2977 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2978 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2979 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2980 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2981 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2982 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2983 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2984 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2986 * New interfaces: pthread_getname_np, pthread_setname_np
2988 * New Linux interface: recvmmsg
2990 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2992 * The dynamic linker now recognizes supported ABI versions from the
2993 EI_ABIVERSION field in the ELF header.
2994 Implemented by Ulrich Drepper.
2996 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2997 will not be used to fill in password fields in the passwd.byname replies.
2998 Instead it is used to synthesize the shadow.byname table, should it be
2999 missing. This is a useful mode in some installations involving Solaris.
3000 Implemented by Ulrich Drepper.
3002 * New locales: kok_IN, sq_MK, cv_RU
3006 * The following bugs are resolved with this release:
3008 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3009 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3010 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3011 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3012 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3013 10789, 10817, 10823, 10840, 10847
3015 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3017 Implemented by Ulrich Drepper.
3019 * Checking version of longjmp added that fails if an uninitialized stack
3020 frame would be created. Implemented by Ulrich Drepper.
3022 * STT_GNU_IFUNC is now supported in static executables.
3023 Implemented by H.J. Lu.
3025 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3027 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3028 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3029 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3030 Contributed by H.J. Lu.
3032 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3033 Implemented by Ulrich Drepper.
3035 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3037 Contributed by H.J. Lu.
3039 * Support for fma instruction in AVX on x86-64.
3040 Implemented by H.J. Lu and Ulrich Drepper.
3042 * AVX support in x86-64 auditing support in ld.so.
3043 Implemented by H.J. Lu.
3045 * STB_GNU_UNIQUE support added.
3046 Implemented by Ulrich Drepper.
3048 * Implement second fallback mode for optimized DNS lookup for even more
3049 broken environments. If two requests from the same port are not
3050 handled correctly close the socket and open a new one before sending
3051 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3052 can be used to select this mode right away, instead of rediscovering the
3053 necessity is every process again.
3054 Implemented by Ulrich Drepper.
3056 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3057 Implemented by Adam Tkac.
3059 * Optimized iconv conversions for S390x.
3060 Implemented by Andreas Krebbel.
3062 * Using condvars with PI mutexes is now more efficient due to kernel
3063 support for requeueing to PI futexes. NPTL support added for x86-64.
3064 Implemented by Ulrich Drepper.
3066 * New locales: ps_AF, my_MM
3070 * The following bugs are resolved with this release:
3072 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3073 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3074 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3075 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3076 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3078 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3079 Implemented by Ulrich Drepper.
3081 * New Linux interfaces: accept4, fallocate, fallocate64.
3082 Implemented by Ulrich Drepper.
3084 * Correct declarations of string function when used in C++ code. This
3085 could lead to compile errors for invalid C++ code.
3087 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3090 * New POSIX 2008 interface: psiginfo
3091 Implemented by Ulrich Drepper.
3093 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3094 Implemented by Ulrich Drepper.
3096 * Support for selecting between multiple function definitions at runtime
3097 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3099 * The libcrypt library can now use the hash function implementations in
3100 NSS. Implemented by Ulrich Drepper.
3102 * The malloc implementation can be compiled to be less memory efficient
3103 but higher performing in multi-threaded programs.
3104 Implemented by Ulrich Drepper.
3106 * New locales: nan_TW@latin, ks_IN
3108 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3109 Implemented by Ulrich Drepper.
3111 * Extended printf hook support. It is possible to use user-defined types
3112 and extend existing format specifiers.
3113 Implemented by Ulrich Drepper.
3115 * Handling for group shadow files has been added.
3116 Implemented by Ulrich Drepper.
3118 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3119 servers (the case, e.g., for some people using the built-in DNS
3120 server in ADSL modems/routers). There is a once-per-process timeout
3121 in case of a broken server. To avoid it, users can run nscd or put
3122 'options single-request' in /etc/resolv.conf.
3123 Implemented by Ulrich Drepper.
3127 * The following bugs are resolved with this release:
3129 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3130 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3131 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3132 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3133 6968, 6974, 6980, 6995, 7008, 7009, 7029
3135 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3136 up at the same time. Implemented by Ulrich Drepper.
3138 * TLS descriptors for LD and GD on x86 and x86-64.
3139 Implemented by Alexandre Oliva.
3141 * getaddrinfo now handles DCCP and UDPlite.
3142 Implemented by Ulrich Drepper.
3144 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3145 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3146 Implemented by Ulrich Drepper.
3148 * New implementation of memmem, strstr, and strcasestr which is O(n).
3149 Implemented by Eric Blake.
3151 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3153 * Implement "e" option for popen to open file descriptor with the
3154 close-on-exec flag set. Implemented by Ulrich Drepper.
3156 * Implement "b" mode for fmemopen. In this mode writes writes don't
3157 implicitly add a NUL byte and seeks from the end of the buffer really
3158 use the buffer end, not the string length as the basis.
3159 Implemented by Ulrich Drepper.
3161 * Many functions, exported and internal, now atomically set the close-on-exec
3162 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3164 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3166 Implemented by Pravin Satpute.
3168 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3172 * The following bugs are resolved with this release:
3174 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3175 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3176 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3177 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3178 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3179 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3180 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3182 * New locales: bo_CN, bo_IN, shs_CA.
3184 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3186 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3187 Implemented by Pravin Satpute.
3189 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3191 * nscd caches now all timeouts for DNS entries
3192 Implemented by Ulrich Drepper.
3194 * nscd is more efficient and wakes up less often.
3195 Implemented by Ulrich Drepper.
3197 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3198 vdprintf, and obstack_vprintf.
3199 Implemented by Jakub Jelinek.
3201 * Faster memset for x86-64.
3202 Implemented by Harsha Jagasia and H.J. Lu.
3204 * Faster memcpy on x86.
3205 Implemented by Ulrich Drepper.
3207 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3208 Implemented by Ulrich Drepper.
3210 * Faster sqrt and sqrtf implemention for some PPC variants.
3211 Implemented by Steven Munroe.
3215 * The following bugs are resolved with this release:
3217 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3218 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3219 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3220 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3221 5071, 5103, 5104, 5112, 5113, 5184, 5186
3223 * More checking functions: fread, fread_unlocked, open*, mq_open.
3224 Implemented by Jakub Jelinek and Ulrich Drepper.
3226 * Extend fortification to C++. Implemented by Jakub Jelinek.
3228 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3229 by not recognizing 'a' as a modifier when those specs are requested.
3230 Implemented by Jakub Jelinek.
3232 * PPC optimizations to math and string functions.
3233 Implemented by Steven Munroe.
3235 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3236 options to be passed. Implemented by Ulrich Drepper.
3238 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3240 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3241 Implemented by Ulrich Drepper.
3243 * Handle private futexes in the NPTL implementation.
3244 Implemented by Jakub Jelinek and Ulrich Drepper.
3246 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3247 Implemented by Roland McGrath and Ulrich Drepper.
3249 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3251 * SHA-256 and SHA-512 based password encryption.
3252 Implemented by Ulrich Drepper.
3254 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3255 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3258 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3259 Implemented by Ulrich Drepper.
3263 * The following bugs are resolved with this release:
3265 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3266 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3267 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3268 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3269 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3270 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3271 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3274 * New Linux interfaces: epoll_pwait, sched_getcpu.
3276 * New generic interfaces: strerror_l.
3278 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3282 * The following bugs are resolved with this release:
3284 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3285 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3286 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3287 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3288 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3289 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3290 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3291 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3292 3225, 3251, 3252, 3253, 3273
3294 * For Linux, the sorting of addresses returned by getaddrinfo now also
3295 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3296 Implemented by Ulrich Drepper.
3298 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3299 Implemented by Ulrich Drepper.
3301 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3303 * New iconv module for MIK. Contributed by Alexander Shopov.
3305 * For sites with broken group and/or passwd database, the auto-propagate
3306 option of nscd can prevent creating ID lookup entries from the results
3307 of a name lookup and vice versa. This usually is no problem but some
3308 site might have problems with the default behavior.
3309 Implemented by Ulrich Drepper.
3311 * Iterating over entire database in NIS can be slow. With the
3312 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3313 to trade time for memory. The entire database will be read at once.
3314 Implemented by Ulrich Drepper.
3316 * The interfaces introduced in RFC 3542 have been implemented by
3319 * Support for the new ELF hash table format was added by Ulrich Drepper.
3321 * Support for priority inheritance mutexes added by Jakub Jelinek and
3324 * Support for priority protected mutexes added by Jakub Jelinek.
3326 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3327 or_IN, csb_PL, fy_NL, sr_ME.
3331 * More overflow detection functions.
3333 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3334 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3336 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3337 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3338 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3339 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3340 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3341 by Masahide Washizawa.
3343 * It is now possible to install audit libraries for the dynamic linker using
3344 LD_AUDIT. Implemented by Ulrich Drepper.
3346 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3347 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3348 For a libc and libpthread that works well on Linux 2.4 kernels, we
3349 recommend using the stable 2.3 branch.
3351 * The new function open_wmemstream acts like open_memstream,
3352 but uses a wchar_t wide character buffer.
3354 * The new function ppoll is an improved version of poll, similar to pselect.
3356 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3357 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3358 renameat, unlinkat, linkat, symlinkat, readlinkat.
3360 * New Linux kernel system calls: unshare,
3361 inotify_init, inotify_add_watch, inotify_rm_watch.
3363 * The euidaccess function is now also known by the alias eaccess,
3364 for compatibility with some other systems.
3366 * Timezone data updated to 2006b version.
3370 * The following bugs are resolved with this release:
3372 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3373 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3374 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3375 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3376 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3377 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3379 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3381 * As of this release, GCC 4 can be used to compile the C Library.
3383 * Timezone data updated to 2005m version.
3387 * The following bugs are resolved with this release:
3389 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3390 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3391 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3392 777, 787, 821, 822, 823, 825
3394 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3398 * Support for RFC 3678. Real implementations exist only for Linux so far.
3399 Implemented by Ulrich Drepper.
3401 * nscd can now cache entries persistently. Expiring entries are reloaded.
3402 For speedups the cache can be shared in memory with client processes.
3403 Implemented by Ulrich Drepper.
3405 * nscd can now perform SELinux checks.
3406 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3408 * getaddrinfo queries are now cached. Canonical name lookup is performed
3410 Implemented by Ulrich Drepper.
3412 * The nothrow function attribute is used when headers are used by gcc when
3413 compiling C code. This can avoid significant amounts of exception
3416 * The malloc functions perform more error checking and are stricter when
3417 it comes to reacting on errors. The default action is to terminate
3418 the process after showing an error message. Implemented by Ulrich Drepper.
3420 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3421 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3423 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3424 namespaces using the new function dlmopen(). This feature is of course,
3425 like most other dynamic loading functionality, not available in statically
3426 linked applications. Implemented by Ulrich Drepper.
3428 * Low-overhead boundary checking variants of string and some stdio functions
3429 were added. These are to be used in conjunction with a gcc patch by
3430 Jakub Jelinek which adds calls to these functions if possible.
3431 Implemented by Jakub Jelinek and Ulrich Drepper.
3433 * Old code for several operating systems and machine architectures that
3434 have not been in working condition in a long time have been removed from
3435 the main source tree maintained by the GNU C Library's maintainers.
3436 These files are now reside in the separate `ports' source module
3437 that is usable as an add-on when building the library.
3441 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3442 interrogate the dynamic linker, compatible with the Solaris interface.
3444 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3445 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3447 * getifaddrs now uses the netlink interface on Linux to get its information.
3448 Implemented by Thorsten Kukuk.
3450 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3451 Implemented by Ulrich Drepper.
3453 * support for non-executable stacks on x86 has been added. Changes mostly
3456 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3459 * getaddrinfo now performs destination address selection according to
3464 * Thread-safe interfaces for many functions that access locale data
3465 were added in version 2.3, but these features were omitted from NEWS.
3466 Many functions have variants with an `_l' suffix that take a `locale_t'
3467 object as a parameter rather than consulting the current locale.
3468 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3469 create and maintain `locale_t' objects. Additionally, the new function
3470 `uselocale' sets "the current locale" (as used by functions not so
3471 parameterized) set for an individual thread. These features were added
3472 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3474 * The functions getresuid, getresgid, setresuid, and setresgid, which
3475 have long been available on Linux, are now declared in <unistd.h>
3476 and are now also available on the Hurd.
3478 * ELF thread-local storage support (TLS) now works on x86-64.
3480 * The new dynamic string token $LIB is expanded in shared library names.
3481 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3483 * Aldy Hernandez contributed complete software floating point support for
3484 PowerPC machines with no FPU.
3486 * fexecve is implemented on Linux.
3488 * The `btowc' function should work at least twice as fast due to
3489 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3491 * With approriate thread add-ons cancelable functions are now implemented
3492 in libc.so as well. No need to call the function in libpthread. This
3493 change allowed to finally disable the incorrect and expensive handling
3494 of weak definition in ld.so.
3496 * Yet more PLT entries in libc.so have been removed. We finally arrived
3497 at the bare minimum. Startup times improved appropriately.
3499 * Support for the new Linux/x86 system call interface was added. The
3500 AT_SYSINFO auxiliary vector entry is recognized and handled.
3504 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3507 * iconv (the program and the interface) now accepts empty names (excluding
3508 options like //TRANSLIT) to mean "use charset of current locale".
3510 * localedef can now transliterate characters in strings which are not in
3511 the provided charmap. The information from the input locale is used.
3513 * Prelinking support was added for ELF targets. This requires additional
3514 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3516 * Read-only stdio streams now use mmap to speed up operation by eliminating
3517 copying and buffer underflows. To use add 'm' to the mode string of
3518 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3520 * The malloc functions were completely rewritten by Wolfram Gloger based
3521 on Doug Lea's malloc-2.7.0.c.
3523 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3524 implementation of regex.
3526 * Bruno Haible upgraded the iconv and locale implementation to support
3529 * Contents of the LC_* and LANG environment variables in the CEN style are
3530 not recognized anymore. It never was used. Change by Ulrich Drepper.
3532 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3533 thread-local storage (TLS) ABI on some platforms.
3534 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3536 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3537 EUC-JISX0213, and TSCII.
3539 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3540 BSD-compatible interface for getting all network interface addresses.
3541 Implementation for IPv4 by Roland McGrath.
3543 * Loading of locale data is faster due to the introduction of a locale
3544 archive. Implemented by Roland McGrath and Ulrich Drepper.
3546 * Startup times are significantly reduced by not using exported functions
3547 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3550 * Steven Munroe contributed a port to PowerPC64/Linux.
3554 * The Hurd now uses the GNU libio implementation of stdio rather than the
3555 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3557 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3558 with the same behavior as the Linux system call of the same name.
3562 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3563 128-bit long double format.
3565 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3566 IBM1161, and IBM1162 charsets.
3568 * Andreas Jaeger contributed a port to x86-64/Linux.
3570 * Peter Bruin contributed a port to PowerPC/Hurd.
3572 * libc functions using I/O streams now can handle wide-oriented streams
3575 * optimizations in the dynamic linker. Binaries created by recent binutils
3576 versions start up quicker due to reduced time spent on relocations.
3578 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3582 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3583 asinh, atanh, j0 for the 128-bit long double format.
3585 * Bruno Haible updated all the code handling Unicode in some form to
3586 support Unicode 3.1.
3588 * Speed of regex for single-byte locales is back to previous levels.
3589 Patch by Isamu Hasegawa.
3591 * Alpha, SPARC, and IA-64 now also using floating stacks.
3593 * Startup time of internationalized applications greatly improved through
3594 iconv cache. Use iconvconfig to generate the cache file.
3595 Contributed by Ulrich Drepper.
3597 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3598 severe performance problems. Patch by David Mosberger.
3600 * The Hurd port got a lot more functionality like AIO, various stdio
3601 extensions, etc. Mainly done by Roland McGrath.
3603 * mtrace can now lookup symbols in shared libraries.
3607 * Intel's IA-64 math library is largely integrated. It provides fast and
3608 accurate implementatations for most basic and standard math functions
3609 in float, double, and long double format.
3611 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3612 and asin for the 96-bit long double format and asin, log, tan for the
3613 128-bit long double format.
3615 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3616 The basic double functions exist today. Contributed by Abraham Ziv
3617 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3618 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3620 * An asynchronous name lookup library was added. The interface is designed
3621 after POSIX AIO. The proposal was circulated beforehand to get comments.
3622 No negative ones came in. Implemented by Ulrich Drepper.
3624 * Port to S390/64bit contributed by Martin Schwidefsky
3625 <schwidefsky@de.ibm.com>.
3627 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3628 of functions for Linux/IA-64.
3630 * The RPC code is now thread safe. Threads can now use the same service
3631 of different services at the same time. Patch by Eric Norum
3632 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3634 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3635 family of functions for Linux/S390.
3637 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3638 of functions for Linux/x86.
3640 * Port to Linux/CRIS contributed by Axis Communications.
3644 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3645 we can now check for namespace violations and missing declarations. The
3646 result is that almost all headers are now Unix-compliant (as defined in
3647 the upcoming XPG6). The negative side is that some programs might need
3648 corrections, too, if they depend on the incorrect form of the headers in
3649 previous versions which defined too many symbols and included too many
3652 * regex now handles multibyte character sets correctly.
3653 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3655 * iconv (the program) does now conform to the upcoming XPG6 and handles
3656 charmaps. Instead of the charset names the path of charmaps can be
3657 provided and the conversion happens based on this data.
3658 Contributed by Ulrich Drepper.
3660 * The locale program now provides detailed information about the installed
3665 only lists the names of the supported locales
3669 provides details such as country, language, and codeset name.
3670 Contributed by Ulrich Drepper.
3674 * The gencat program now parses the input file according to the charset
3675 selected by the LC_CTYPE category. This is important for stateful
3676 character sets. To make generating catalogs easier there is a way
3677 to overwrite the charset selected by the locale: before the first
3678 message or $ quote line the catalog can contain a line like
3680 $ codeset=ISO-8859-2
3682 to select the charset (ISO-8859-2 in this case).
3684 Implemented by Shinya Hanataka and Ulrich Drepper.
3686 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3687 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3688 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3690 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3691 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3693 * The LANGUAGE environment variable is now ignored unless the locale is
3694 changed from the default "C" locale.
3696 * The usual bug fixes.
3700 * Greg McGary added runtime support for bounds checking using gcc's
3701 new -fbounded-pointers option. ix86 target is complete. PowerPC
3704 * Thorsten Kukuk added secure mode support to nscd.
3706 * The Berkeley db libraries have been removed.
3708 Related, the nss_db module is now in a separate package since it
3709 obviously requires a database library being available.
3711 * Wide character I/O streams implemented by Ulrich Drepper.
3713 * Functions from the extended socket API added by Ulrich Drepper.
3715 * Functions feenableexcept and fedisableexcept to control the
3716 behaviour of individual exceptions have been added by Andreas Jaeger.
3718 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3720 * The resolver code has been updated from bind 8.2.3-T5B which supports
3721 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3724 This change could in some situations effect backward compatibility. Since
3725 now `_res' is a thread-local instead of a global variable, modifying it
3726 in one thread does not have any effect in other threads.
3728 The resolver library was also extended to allow IPv6 as the transport
3729 protocol for the requests. This work was done by Stig Venaas.
3731 * Compatibility code for K&R C compilers has been removed from the
3732 header files. A ISO C compiler is needed to use the library
3733 (conforming to either C89 or C99 standard).
3735 * Complete rewrite of the localedef program to support multibyte character
3736 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3737 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3738 Implemented by Ulrich Drepper.
3740 Bruno Haible significantly improved the generation and use of the data
3741 structures for the wide character tables.
3743 * Plural handling in gettext implemented by Ulrich Drepper.
3745 * The utmp daemon has been removed.
3747 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3749 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3752 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3754 * POSIX barriers implemented by Kaz Kylheku.
3756 * POSIX spawn function family implemented by Ulrich Drepper.
3758 * POSIX spinlocks are now available.
3760 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3762 * the configure option --enable-kernel=X.Y.Z allows to strip out
3763 compatibility for kernel versions before X.Y.Z. This is currently only
3764 implemented for Linux.
3766 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3767 field sin6_scope_id. This means that all programs using IPv6 should be
3768 recompiled. Don't expect binary compatibility with previous glibc
3771 * various conversion modules for IBM character sets contributed by
3774 * IA-64 port by Jes Sorensen and HJ Lu.
3788 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3790 * New xdr functions are added; some rpc functions are now 64bit clean.
3792 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3794 * Fixed known incompatibilities with glibc 2.0.
3796 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3798 * Optimized a number of functions (especially the ELF dynamic loader).
3800 * Update timezone data files.
3802 * lots of charmaps corrections
3804 * some new locale definitions and charmaps
3809 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3810 conform to POSIX member type requirements. Symbol versions have been
3811 adjusted as needed within the library, and for direct use by applications,
3812 but there is potential for problems if third-party libraries use
3813 struct timeval as part of their interface. This does not present
3814 a problem for X and other "essential" system libraries.
3816 * An additional locale model to support C++ Standard Library locale
3817 model and probably more was implemented by Ulrich Drepper.
3819 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3822 * Miles Bader provided the `argp' function family to support hierachical
3823 command line argument parsing, layered on top of getopt.
3825 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3827 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3830 * scanf recognizes the %a and %A format for scanning floating point numbers.
3832 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3833 information and interfaces for the available integer types.
3835 * about 130 new math functions were added to implement the ISO C9x math
3838 * the new header <complex.h> contains definitions of the complex math
3839 functions from ISO C 9X.
3841 * the new header <tgmath.h> defines generic macros to use complex or
3842 real valued functions.
3844 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3846 * Andreas Jaeger provided a test suite for the math library.
3848 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3850 * Ulrich Drepper added character set conversion functions (iconv).
3852 * Optimized string functions have been added.
3854 * The localedata addon is now part of glibc.
3856 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3858 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3859 daemon for NSS (nscd).
3861 Missing a better place here are some numbers on improvements. Under
3862 Linux 2.1.125 un-tar-ing the kernel sources takes
3866 using local files 12.19s 6.88s 22.91s
3868 using NIS 13.92s 8.91s 26.34s
3870 using NIS & nscd 10.37s 7.34s 25.30s
3872 using NIS+ 27.57s 30.37s 640.46s
3874 using NIS+ & nscd 10.25s 7.83s 26.51s
3876 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3878 Keep in mind that non-namelookup related operations dominate above times.
3879 It was just a common complain that using NIS+ unpacking the kernel is
3882 [1] The old nscd implementation is not available anymore since it was
3883 distributed with glibc up to version 2.0.98 and thus is now replaced.
3885 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3887 * Mark Kettenis provided a Hesiod NSS module.
3889 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3890 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3892 * rcmd can now handle netgroups (Dick Streefland).
3894 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3897 * Support for the IPv6 protocol has been added to the socket API, as per the
3898 latest draft standards.
3900 * Support for Linux 2.2 has been added.
3902 * Interface changes relative to the latest 2.0.x release:
3903 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3904 addseverity NEW: Unix98
3905 alphasort64 NEW: LFS
3906 argp_err_exit_status NEW: argp, GNU ext
3907 argp_error NEW: argp, GNU ext
3908 argp_failure NEW: argp, GNU ext
3909 argp_help NEW: argp, GNU ext
3910 argp_parse NEW: argp, GNU ext
3911 argp_program_bug_address NEW: argp, GNU ext
3912 argp_program_version NEW: argp, GNU ext
3913 argp_program_version_hook NEW: argp, GNU ext
3914 argp_state_help NEW: argp, GNU ext
3915 argp_usage NEW: argp, GNU ext
3916 authdes_create NEW: Secure RPC
3917 authdes_getucred NEW: Secure RPC
3918 authdes_pk_create NEW: Secure RPC
3919 backtrace NEW: GNU ext.
3920 backtrace_symbols NEW: GNU ext.
3921 backtrace_symbols_fd NEW: GNU ext.
3923 cacosf NEW: ISO C 9x
3924 cacosh NEW: ISO C 9x
3925 cacoshf NEW: ISO C 9x
3926 cacoshl NEW: ISO C 9x
3927 cacosl NEW: ISO C 9x
3934 casinf NEW: ISO C 9x
3935 casinh NEW: ISO C 9x
3936 casinhf NEW: ISO C 9x
3937 casinhl NEW: ISO C 9x
3938 casinl NEW: ISO C 9x
3940 catanf NEW: ISO C 9x
3941 catanh NEW: ISO C 9x
3942 catanhf NEW: ISO C 9x
3943 catanhl NEW: ISO C 9x
3944 catanl NEW: ISO C 9x
3945 cbc_crypt NEW: Secure RPC
3949 ccoshf NEW: ISO C 9x
3950 ccoshl NEW: ISO C 9x
3956 cimagf NEW: ISO C 9x
3957 cimagl NEW: ISO C 9x
3958 clearerr_locked REMOVED
3959 clntunix_create NEW: sunrpc ext
3961 clog10 NEW: ISO C 9x
3962 clog10f NEW: ISO C 9x
3963 clog10l NEW: ISO C 9x
3973 cprojf NEW: ISO C 9x
3974 cprojl NEW: ISO C 9x
3976 crealf NEW: ISO C 9x
3977 creall NEW: ISO C 9x
3982 csinhf NEW: ISO C 9x
3983 csinhl NEW: ISO C 9x
3986 csqrtf NEW: ISO C 9x
3987 csqrtl NEW: ISO C 9x
3991 ctanhf NEW: ISO C 9x
3992 ctanhl NEW: ISO C 9x
3994 des_setparity NEW: Secure RPC
3995 ecb_crypt NEW: Secure RPC
3996 endutxent NEW: Unix98
3998 exp10f NEW: ISO C 9x
3999 exp10l NEW: ISO C 9x
4003 fattach NEW: STREAMS
4004 fdetach NEW: STREAMS
4008 feclearexcept NEW: ISO C 9x
4009 fegetenv NEW: ISO C 9x
4010 fegetexceptflag NEW: ISO C 9x
4011 fegetround NEW: ISO C 9x
4012 feholdexcept NEW: ISO C 9x
4014 feraiseexcept NEW: ISO C 9x
4015 ferror_locked REMOVED
4016 fesetenv NEW: ISO C 9x
4017 fesetexceptflag NEW: ISO C 9x
4018 fesetround NEW: ISO C 9x
4019 fetestexcept NEW: ISO C 9x
4020 feupdateenv NEW: ISO C 9x
4021 fflush_locked REMOVED
4025 fgets_unlocked NEW: GNU ext.
4026 fileno_locked REMOVED
4038 fputc_locked REMOVED
4039 fputs_unlocked NEW: GNU ext.
4040 fread_unlocked NEW: GNU ext.
4045 fstatvfs NEW: Unix98
4049 ftruncate64 NEW: LFS
4051 fwrite_unlocked NEW: GNU ext.
4052 gai_strerror NEW: IPv6
4056 getchar_locked REMOVED
4058 getdate_err NEW: Unix98
4059 getdate_r NEW: GNU ext.
4061 getnameinfo NEW: IPv6
4062 getnetname NEW: Secure RPC
4063 getpmsg NEW: STREAMS
4064 getpt NEW: Unix98 PTY
4065 getrlimit64 NEW: LFS
4066 getutxent NEW: Unix98
4067 getutxid NEW: Unix98
4068 getutxline NEW: Unix98
4069 glob64 NEW: GNU ext.
4070 globfree64 NEW: GNU ext.
4071 gnu_get_libc_release NEW: GNU ext.
4072 gnu_get_libc_version NEW: GNU ext.
4073 grantpt NEW: Unix98 PTY
4074 host2netname NEW: Secure RPC
4076 iconv_close NEW: iconv
4077 iconv_open NEW: iconv
4078 if_freenameindex NEW: IPv6
4079 if_indextoname NEW: IPv6
4080 if_nameindex NEW: IPv6
4081 if_nametoindex NEW: IPv6
4082 in6addr_any NEW: IPv6
4083 in6addr_loopback NEW: IPv6
4084 inet6_isipv4mapped NEW: IPv6
4085 isastream NEW: STREAMS
4086 iswblank NEW: GNU ext.
4087 key_decryptsession NEW: Secure RPC
4088 key_decryptsession_pk NEW: Secure RPC
4089 key_encryptsession NEW: Secure RPC
4090 key_encryptsession_pk NEW: Secure RPC
4091 key_gendes NEW: Secure RPC
4092 key_get_conv NEW: Secure RPC
4093 key_secretkey_is_set NEW: Secure RPC
4094 key_setnet NEW: Secure RPC
4095 key_setsecret NEW: Secure RPC
4096 llrint NEW: ISO C 9x
4097 llrintf NEW: ISO C 9x
4098 llrintl NEW: ISO C 9x
4099 llround NEW: ISO C 9x
4100 llroundf NEW: ISO C 9x
4101 llroundl NEW: ISO C 9x
4106 lrintf NEW: ISO C 9x
4107 lrintl NEW: ISO C 9x
4108 lround NEW: ISO C 9x
4109 lroundf NEW: ISO C 9x
4110 lroundl NEW: ISO C 9x
4112 makecontext NEW: Unix98
4113 mempcpy NEW: GNU ext.
4116 modify_ldt NEW: kernel
4120 nearbyint NEW: ISO C 9x
4121 nearbyintf NEW: ISO C 9x
4122 nearbyintl NEW: ISO C 9x
4123 netname2host NEW: Secure RPC
4124 netname2user NEW: Secure RPC
4125 nexttoward NEW: ISO C 9x
4126 nexttowardf NEW: ISO C 9x
4127 nexttowardl NEW: ISO C 9x
4131 passwd2des NEW: Secure RPC
4133 pow10f NEW: GNU ext.
4134 pow10l NEW: GNU ext.
4137 printf_size NEW: GNU ext.
4138 printf_size_info NEW: GNU ext.
4139 profil_counter REMOVED
4140 pthread_mutexattr_getkind_np REPLACED
4141 pthread_mutexattr_setkind_np REPLACED
4142 ptsname NEW: Unix98 PTY
4143 ptsname_r NEW: Unix98 PTY
4145 putchar_locked REMOVED
4146 putgrent NEW: GNU ext.
4148 putpmsg NEW: STREAMS
4149 pututxline NEW: Unix98
4153 readdir64_r NEW: LFS
4154 remquo NEW: ISO C 9x
4155 remquof NEW: ISO C 9x
4156 remquol NEW: ISO C 9x
4158 roundf NEW: ISO C 9x
4159 roundl NEW: ISO C 9x
4161 scalbln NEW: ISO C 9x
4162 scalblnf NEW: ISO C 9x
4163 scalblnl NEW: ISO C 9x
4165 sendfile NEW: kernel
4166 setcontext NEW: Unix98
4167 setrlimit64 NEW: LFS
4168 setutxent NEW: Unix98
4170 sigignore NEW: Unix98
4171 sigqueue NEW: POSIX.1b
4172 sigrelse NEW: Unix98
4173 sigset NEW: POSIX.1b
4174 sigtimedwait NEW: POSIX.1b
4175 sigwaitinfo NEW: POSIX.1b
4176 sincos NEW: GNU ext.
4177 sincosf NEW: GNU ext.
4178 sincosl NEW: GNU ext.
4182 strcasestr NEW: GNU ext.
4183 strtoimax NEW: ISO C 9x
4184 strtoumax NEW: ISO C 9x
4185 strverscmp NEW: GNU ext.
4186 svcauthdes_stats NEW: Secure RPC
4187 svcunix_create NEW: sunrpc etx
4188 svcunixfd_create NEW: sunrpc ext
4189 swapcontext NEW: Unix98
4190 tcgetsid NEW: Unix98 PTY
4191 tdestroy NEW: GNU ext.
4192 tgamma NEW: ISO C 9x
4193 tgammaf NEW: ISO C 9x
4194 tgammal NEW: ISO C 9x
4198 truncf NEW: ISO C 9x
4199 truncl NEW: ISO C 9x
4201 unlockpt NEW: Unix98 PTY
4202 updwtmpx NEW: Unix98
4203 user2netname NEW: Secure RPC
4204 utmpxname NEW: Unix98
4205 versionsort NEW: GNU ext.
4206 versionsort64 NEW: GNU ext.
4208 wcscasecmp NEW: GNU ext.
4209 wcsncasecmp NEW: GNU ext.
4210 wcsnlen NEW: GNU ext.
4211 wcstoimax NEW: ISO C 9x
4212 wcstoll NEW: ISO C 9x
4213 wcstoull NEW: ISO C 9x
4214 wcstoumax NEW: ISO C 9x
4216 wordexp NEW: POSIX.2
4217 wordfree NEW: POSIX.2
4218 write_profiling REMOVED
4219 xdecrypt NEW: Secure RPC
4220 xdr_authdes_cred NEW: Secure RPC
4221 xdr_authdes_verf NEW: Secure RPC
4222 xdr_cryptkeyarg NEW: Secure RPC
4223 xdr_cryptkeyarg2 NEW: Secure RPC
4224 xdr_cryptkeyres NEW: Secure RPC
4225 xdr_getcredres NEW: Secure RPC
4226 xdr_key_netstarg NEW: Secure RPC
4227 xdr_key_netstres NEW: Secure RPC
4228 xdr_keybuf NEW: Secure RPC
4229 xdr_keystatus NEW: Secure RPC
4230 xdr_netnamestr NEW: Secure RPC
4231 xdr_sizeof NEW: Secure RPC
4232 xdr_unixcred NEW: sunrpc ext
4233 xencrypt NEW: Secure RPC
4234 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4245 * inet_ntoa is thread-safe
4247 * updwtmp is moved from libutil to libc
4249 * rewrite of cbrt function
4251 * update of timezone data
4265 * add atoll function
4267 * fix complex problems in Berkeley DB code
4269 * fix math functions
4273 * fixed lots of header problems (especially Linux/GNU specific)
4275 * dynamic loader preserves all registers
4277 * Roland McGrath provided support for handling of auxiliary objects in
4278 the ELF dynamic loader.
4280 * support for parallel builds is improved
4284 * GNU extensions are no longer declared by default. To enable them you
4285 must define the macro `_GNU_SOURCE' in your program or compile with
4288 * The library has changed from using GNU ld symbol aliases to using weak
4289 symbols where available. The ELF object file format supports weak
4290 symbols; GNU ld also supports weak symbols in the a.out format. (There
4291 is also now support for other GNU ld extensions in ELF. Use the
4292 `--with-elf' option to configure to indicate you have ELF, and
4293 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4294 of many files which contained only symbol aliases, reducing the size of
4295 the source and the compiled library; many other files were renamed to
4296 less cryptic names previously occupied by the symbol alias files.
4297 There is a new header file <elf.h> for programs which operate on
4298 files in the ELF format.
4300 * Converted to Autoconf version 2, so `configure' has more options.
4301 Run `configure --help' to see the details.
4303 * The library can now be configured to build profiling, highly-optimized
4304 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4305 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4306 `configure' enable building these extra libraries. The shared library is
4307 built by default when using both ELF and GNU ld. When shared libraries
4308 are enabled, the new library `-ldl' is available for arbitrary run-time
4309 loading of shared objects; its interface is defined in <dlfcn.h>. The
4310 new header file <link.h> gives access to the internals of the run-time
4311 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4312 application of same name on other systems and it provides information
4313 about dynamically linked binaries.
4315 * The C library now provides the run-time support code for profiling
4316 executables compiled with `-pg'. Programs can control the profiling code
4317 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4318 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4319 the support for this file format was contributed by David Mosberger-Tang.
4321 * The math code has been replaced with a math library based on fdlibm from
4322 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4323 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4324 functions. The math functions now reside in a separate library, so
4325 programs using them will need to use `-lm' their linking commands.
4327 * John C. Bowman contributed optimized ix87 assembler inline functions.
4329 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4330 mechanism similar to that found in Solaris 2. This is now used for the
4331 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4332 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4333 file controls what services are used for each individual database. This
4334 works by loading shared libraries with names specified in `nsswitch.conf',
4335 so service modules can be changed or added at any time without even
4336 relinking any program. Currently there are the file, db, and NIS based
4337 NSS services available.
4339 * The new functions `strtoq' and `strtouq' parse integer values from
4340 strings, like `strtol' and `strtoul', but they return `long long int' and
4341 `unsigned long long int' values, respectively (64-bit quantities).
4343 * The new functions `strtof' and `strtold' parse floating-point values from
4344 strings, like `strtod', but they return `float' and `long double' values,
4345 respectively (on some machines `double' and `long double' are the same).
4347 * Ulrich Drepper has contributed new implementations of the floating-point
4348 printing and reading code used in the `printf' family of functions and
4349 `strtod', `strtof', and `strtold'. These new functions are perfectly
4350 accurate, and much faster than the old ones.
4352 * The implementation of the POSIX locale model was completely rewritten by
4353 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4354 to compile the POSIX locale definition.
4356 * The former dummy implementations of the strcoll and strxfrm function are
4357 now replaced by fully functional code contributed by Ulrich Drepper. The
4358 collation information comes from the POSIX locale definitions.
4360 * The new header <langinfo.h> defines an interface for accessing
4361 various locale-dependent data (using the locale chosen with `setlocale').
4363 * Ulrich Drepper has contributed a new suite of functions for operation on
4364 wide-character and multibyte-character strings, in <wchar.h>;
4365 and classification and case conversion of wide characters, in <wctype.h>.
4366 These new functions are conforming to the ISO C, Amendement 1 specification.
4368 * There is now a second implementation of the standard I/O library available.
4369 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4370 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4372 * You can now use positional parameter specifications in format strings
4373 for the `printf' and `scanf' families of functions. For example,
4374 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4375 ``Number 6, Mr Jones''. This is mainly useful when providing different
4376 format strings for different languages, whose grammars may dictate
4377 different orderings of the values being printed. To support this
4378 feature, the interface for `register_printf_handler' has changed; see
4379 the header file <printf.h> for details.
4381 * The `printf' and `scanf' families of functions now understand a new
4382 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4383 to group numbers as indicated by the locale; for `scanf' and friends, this
4384 says to accept as valid only a number with all the proper grouping
4385 separators in the right places. In the default "C" locale, numbers are
4386 not grouped; but locales for specific countries will define the usual
4387 conventions (i.e. separate thousands with `,' in the US locale).
4389 * The pgrp functions have been regularized, slightly incompatibly but much
4390 less confusingly. The core functions are now `getpgid' and `setpgid',
4391 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4392 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4393 provided for compatibility. There is no longer an incompatible `getpgrp'
4394 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4396 * The new header file <fts.h> and suite of functions simplify programs that
4397 operate on directory trees. This code comes from 4.4 BSD.
4399 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4400 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4401 NSS scheme used in glibc.
4403 * There is a new malloc debugging hook `__memalign_hook'.
4405 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4406 `unsigned int' in <sys/types.h>. These are for compatibility only and
4407 their use is discouraged.
4409 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4410 done differently, so that it works even without GNU ld.
4412 * New function `euidaccess' checks allowed access to a file like `access',
4413 but using the effective IDs instead of the real IDs.
4415 * The time zone data files have been updated for the latest and greatest
4416 local time conventions of the countries of the world.
4418 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4421 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4422 for formatting floating-point numbers. They are provided only for
4423 compatibility; new programs should use `sprintf' instead. There are
4424 also equivalent function for the `long double' floating-point type and
4425 all functions also exist in a reentrant form.
4427 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4428 functions for maintaining the login-record files (primarily of use to
4429 system programs such as `login'), and convenient functions for
4430 allocating and initializing a pseudo-terminal (pty) device.
4432 * Ulrich Drepper has contributed new support for System V style
4433 shared memory and IPC on systems that support it.
4435 * Ulrich Drepper has contributed several miscellaneous new functions found
4436 in System V: The `hsearch' family of functions provide an effective
4437 implementation of hash tables; `a64l' and `l64a' provide a very simple
4438 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4441 * Ulrich Drepper has contributed new reentrant counterparts for the
4442 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4444 * Ulrich Drepper has contributed new, highly-optimized versions of several
4445 string functions for the i486/Pentium family of processors.
4447 * Ulrich Drepper has updated the Linux-specific code, based largely
4448 on work done in Hongjiu Lu's version of GNU libc for Linux.
4449 The GNU library now supports Linux versions 2.0.10 and later,
4450 using the ELF object file format (i[3456]86-*-linux).
4452 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4454 * David Mosberger-Tang and Richard Henderson have ported the C library
4455 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4456 dynamic linking support for ELF/Alpha.
4458 * Richard Henderson contributed several Alpha optimized assembler function
4459 for arithmetic and string handling.
4461 * Ulrich Drepper has contributed a new set of message catalog functions to
4462 support multiple languages using the <libintl.h> interface, for use with
4463 his new package GNU gettext. Translation volunteers have contributed
4464 catalogs of the library's messages in Spanish, German, and Korean.
4466 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4467 program and the `catgets' function for reading the catalog files it
4468 creates. (The <libintl.h> interface is preferred; we include the
4469 <nl_types.h> interface using `catgets' only for source compatibility with
4470 programs already written to use it.)
4472 * New header file <values.h> gives SVID-compatible names for <limits.h>
4475 * Various new macros, declarations, and small header files for compatibility
4478 * New function `group_member' is a convenient way to check if a process has
4479 a given effective group ID.
4481 * When using GCC 2.7 and later, the socket functions are now declared in a
4482 special way so that passing an argument of type `struct sockaddr_in *',
4483 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4484 `struct sockaddr *' type, does not generate a type-clash warning.
4486 * New function `error' declared in header file <error.h> is a convenient
4487 function for printing error messages and optionally exiting; this is the
4488 canonical function used in GNU programs. The new functions `err', `warn',
4489 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4490 doing the same thing.
4492 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4493 POSIX.2 `glob' function to do ~ and {...} expansion.
4495 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4496 `clearenv' which is used in POSIX.9 is also available.
4498 * New function `getsid' returns session ID number on systems that support it.
4500 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4501 files <db.h> and <mpool.h> provide a rich set of functions for several
4502 types of simple databases stored in memory and in files, and <ndbm.h> is
4503 an old `ndbm'-compatible interface using the `db' functions. Link with
4504 `-ldb' to get these functions.
4506 * New macro `strdupa' copies a string like `strdup', but uses local stack
4507 space from `alloca' instead of dynamic heap space from `malloc'.
4509 * New function `strnlen' is like `strlen' but searches only a given maximum
4510 number of characters for the null terminator. `stpncpy', `strndup' and
4511 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4514 * New function `statfs' in header <sys/statfs.h>.
4516 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4517 provide convenient functions for operating on blocks of null-terminated
4520 * A new suite of functions in <utmp.h> handle all the details of reading
4521 and writing the utmp file.
4523 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4526 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4527 completely POSIX compliant and also implemented the extended functionality
4528 to handle alternate digit representation and alternate era date formats.
4530 * Ulrich Drepper provided an implementation of the `strptime' function
4531 defined in XPG4.2 which transforms a string into a `struct tm' value.
4533 * Paul Eggert provided the tzselect shell script as part of the timezone
4534 code. The shell script makes it easy to select the correct timezone
4537 * The implementation of the malloc family of functions is completely replaced
4538 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4539 The implementation uses the mmap function (if available) and it is
4540 optimized for the use in multi threaded programs.
4542 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4543 functions. This new functionality is usable by specifying a special
4544 salt string and it is compatible with implementation on *BSD systems.
4546 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4547 `getsubopt' to handle second level command line options, `bsd_signal'
4548 to access BSD style `signal' functionality, the obsolete `regexp' style
4551 * the `lchown' function is available on system which support this
4554 * The implementation of the shadow password handling function was contributed
4557 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4559 * POSIX.1g support was added. The <sys/select.h> header is available,
4560 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4561 implementation of `getaddrinfo'.
4565 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4567 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4568 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4570 * The new function `daemon' from 4.4 BSD is useful for server programs that
4571 want to put themselves in the background.
4573 * Joel Sherrill has contributed support for several standalone boards that
4574 run without an operating system.
4576 * `printf', `scanf' and friends now accept a `q' type modifier for long
4577 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4579 * All of the code taken from BSD (notably most of the math and networking
4580 routines) has been updated from the BSD 4.4-Lite release.
4582 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4584 * The new functions `getdomainname' and `setdomainname' fetch or change the
4585 YP/NIS domain name. These are system calls which exist on systems which
4588 * The time zone data files have been updated for the latest international
4591 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4592 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4596 * The C library now includes support for Sun RPC, from Sun's free
4597 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4598 are included. (There is still no support for YP.)
4600 * Tom Quinn has contributed a port of the C library to SGI machines running
4601 Irix 4 (mips-sgi-irix4).
4603 * The new `lockf' function is a simplified interface to the locking
4604 facilities of `fcntl', included for compatibility.
4606 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4608 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4610 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4613 * The error message printed by `assert' for a failed assertion now includes
4614 the name of the program (if using GNU ld) and the name of the calling
4615 function (with versions of GCC that support this).
4617 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4619 * The library now includes the <sys/mman.h> header file and memory
4620 management functions `mmap', `munmap', `mprotect', `msync', and
4621 `madvise', on systems that support those facilities.
4623 * The interface for `mcheck' has changed slightly: the function called to
4624 abort the program when an allocation inconsistency is detected now takes
4625 an argument that indicates the type of failure. The new function
4626 `mprobe' lets you request a consistency check for a particular block at
4627 any time (checks are normally done only when you call `free' or `realloc'
4630 * It is now possible to easily cross-compile the C library, building on one
4631 system a library to run on another machine and/or operating system. All
4632 you need to do is set the variable `HOST_CC' in `configparms' to the
4633 native compiler for programs to run on the machine you are building on (a
4634 few generator programs are used on Unix systems); set `CC' to the
4637 * The new function `fexecve' (only implemented on the GNU system) executes
4638 a program file given a file descriptor already open on the file.
4642 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4643 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4645 * You can set the variable `libprefix' in `configparms' to specify a prefix
4646 to be prepended to installed library files; this makes it easy to install
4647 the GNU C library to be linked as `-lgnuc' or whatever.
4649 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4650 copies a limited number of characters from a string, and returns the
4651 address of the last character written.
4653 * You no longer need to check for whether the installed `stddef.h' is
4654 compatible with the GNU C library. configure now checks for you.
4656 * You can now define a per-stream `fileno' function to convert the
4657 stream's cookie into an integral file descriptor.
4659 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4660 allocates zero bytes of storage, and returns a unique pointer which
4661 you can pass to `realloc' or `free'. The behavior is undefined if
4662 you dereference this pointer.
4664 * The C library now runs on Sony NEWS m68k machines running either
4665 NewsOS 3 or NewsOS 4.
4667 * The new `syscall' function is a system-dependent primitive function
4668 for invoking system calls. It has the canonical behavior on Unix
4669 systems, including unreliable return values for some calls (such as
4670 `pipe', `fork' and `getppid').
4672 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4673 to `EAGAIN', which is the preferred name. On systems whose kernels
4674 use two distinct codes, the C library now translates EWOULDBLOCK to
4675 EAGAIN in every system call function.
4679 * The GNU C Library Reference Manual is now distributed with the library.
4680 `make dvi' will produce a DVI file of the printed manual.
4681 `make info' will produce Info files that you can read on line using C-h i
4682 in Emacs or the `info' program.
4683 Please send comments on the manual to bug-glibc-manual@gnu.org.
4685 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4687 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4689 * Jason Merrill has contributed a port to the Sequent Symmetry running
4690 Dynix version 3 (i386-sequent-dynix).
4692 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4693 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4695 * New function `memory_warnings' lets you arrange to get warnings when
4696 malloc is running out of memory to allocate, like Emacs gives you.
4698 * The C library now contains the relocating allocator used in Emacs 19 for
4699 its editing buffers. This allocator (ralloc) minimizes allocation
4700 overhead and fragmentation by moving allocated regions around whenever it
4701 needs to. You always refer to a ralloc'd region with a "handle" (a
4702 pointer to a pointer--an object of type `void **').
4704 * There is a new `printf' format: `%m' gives you the string corresponding
4705 to the error code in `errno'.
4707 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4708 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4709 pass, the `a' modifier says to fill in a `char **' you pass with a
4712 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4713 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4714 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4716 * `mkstemp' is a traditional Unix function to atomically create and open a
4717 uniquely-named temporary file.
4721 * The standard location for the file that says what the local timezone is
4722 has changed again. It is now `/usr/local/etc/localtime' (or more
4723 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4725 * The distribution no longer contains any files with names longer than 14
4728 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4729 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4731 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4735 * The configuration process has changed quite a bit. The `configure'
4736 script is now used just like the configuration scripts for other GNU
4737 packages. The `sysdeps' directory hierarchy is much rearranged.
4738 The file `INSTALL' explains the new scheme in detail.
4740 * The header files no longer need to be processed into ANSI C and
4741 traditional C versions. There is just one set of files to install, and
4742 it will work with ANSI or old C compilers (including `gcc -traditional').
4744 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4745 MIPS DECStation running Ultrix 4.
4747 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4748 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4749 made itself into a shared library.
4751 * Yet further improved support for the i386, running 4.3 BSD-like systems
4752 (such as Mach 3 with the Unix single-server), or System V.
4754 * New function `strncasecmp' to do case-insensitive string comparison
4755 with limited length.
4757 * New function `strsep' is a reentrant alternative to `strtok'.
4759 * New functions `scandir' and `alphasort' for searching directories.
4761 * New function `setenv' is a better interface to `putenv'.
4763 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4764 function for traversing a directory tree.
4766 * The GNU obstack package is now also part of the C library.
4767 The new function `open_obstack_stream' creates a stdio stream that
4768 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4769 formatted output directly to an obstack.
4771 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4772 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4774 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4776 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4777 things to your strings.
4779 * There are some new test programs: test-fseek, testmb, and testrand.
4781 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4782 These ports are not finished, but are a good starting place for really
4783 supporting those systems.
4785 * `/etc/localtime' is now the standard location for the file that says what
4786 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4787 This follows the general principle that `/etc' is the place for all local
4788 configuration files.
4790 * The C library header files now use `extern "C"' when used by the C++
4791 compiler, so the C library should now work with C++ code.
4793 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4794 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4797 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4798 function which uses the `merge sort' algorithm, and is said to be
4799 significantly faster than the old GNU `qsort' function. Merge sort is
4800 now the standard `qsort' function. The new algorithm can require a lot
4801 of temporary storage; so, the old sorting function is called when the
4802 required storage is not available.
4804 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4805 provides the Unix `crypt' function, plus some other entry points.
4807 * The code and header files taken from 4.4 BSD have been updated with the
4808 latest files released from Berkeley.
4810 ----------------------------------------------------------------------
4811 Copyright information:
4813 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4815 Permission is granted to anyone to make or distribute verbatim copies
4816 of this document as received, in any medium, provided that the
4817 copyright notice and this permission notice are preserved,
4818 thus giving the recipient permission to redistribute in turn.
4820 Permission is granted to distribute modified versions
4821 of this document, or of portions of it,
4822 under the above conditions, provided also that they
4823 carry prominent notices stating who last changed them.
4826 version-control: never