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 Deprecated and removed features, and other changes affecting compatibility:
20 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
21 installed. Software that was using either header should be updated to
22 use standard <stdio.h> interfaces instead.
24 * The stdio.h functions 'getc' and 'putc' are no longer defined as macros.
25 This was never required by the C standard, and the macros just expanded
26 to call alternative names for the same functions. If you hoped getc and
27 putc would provide performance improvements over fgetc and fputc, instead
28 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
29 necessary, flockfile and funlockfile.
31 * All stdio functions now treat end-of-file as a sticky condition. If you
32 read from a file until EOF, and then the file is enlarged by another
33 process, you must call clearerr or another function with the same effect
34 (e.g. fseek, rewind) before you can read the additional data. This
35 corrects a longstanding C99 conformance bug. It is most likely to affect
36 programs that use stdio to read interactive input from a terminal.
39 * The macros 'major', 'minor', and 'makedev' are now only available from
40 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
41 headers that happen to include <sys/types.h>. These macros are rarely
42 used, not part of POSIX nor XSI, and their names frequently collide with
43 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
46 <sys/sysmacros.h> is a GNU extension. Portable programs that require
47 these macros should first include <sys/types.h>, and then include
48 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
50 Changes to build and runtime requirements:
52 [Add changes to build and runtime requirements here]
54 Security related changes:
56 [Add security related changes here]
58 The following bugs are resolved with this release:
60 [The release manager will add the list generated by
61 scripts/list-fixed-bugs.py just before the release.]
68 * The GNU C Library can now be compiled with support for building static
69 PIE executables (See --enable-static-pie in INSTALL). These static PIE
70 executables are like static executables but can be loaded at any address
71 and provide additional security hardening benefits at the cost of some
72 memory and performance. When the library is built with --enable-static-pie
73 the resulting libc.a is usable with GCC 8 and above to create static PIE
74 executables using the GCC option '-static-pie'. This feature is currently
75 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
76 aarch64 with binutils 2.30 or later.
78 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
79 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
82 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
84 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
86 * In order to support faster and safer process termination the malloc API
87 family of functions will no longer print a failure address and stack
88 backtrace after detecting heap corruption. The goal is to minimize the
89 amount of work done after corruption is detected and to avoid potential
90 security issues in continued process execution. Reducing shutdown time
91 leads to lower overall process restart latency, so there is benefit both
92 from a security and performance perspective.
94 * The abort function terminates the process immediately, without flushing
95 stdio streams. Previous glibc versions used to flush streams, resulting
96 in deadlocks and further data corruption. This change also affects
97 process aborts as the result of assertion failures.
99 * On platforms where long double has the IEEE binary128 format (aarch64,
100 alpha, mips64, riscv, s390 and sparc), the math library now implements
101 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
102 These are the same interfaces added in version 2.26 for some platforms where
103 this format is supported but is not the format of long double.
105 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
106 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
107 implements interfaces for that type, as defined by ISO/IEC TS
108 18661-3:2015. These are corresponding interfaces to those supported for
111 * The math library now implements interfaces for the _Float32, _Float64 and
112 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
113 corresponding interfaces to those supported for _Float128.
115 * glibc now implements the memfd_create and mlock2 functions on Linux.
117 * Support for memory protection keys was added. The <sys/mman.h> header now
118 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
121 * The copy_file_range function was added.
123 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
125 * The ldconfig utility now processes `include' directives using the C/POSIX
126 collation ordering. Previous glibc versions used locale-specific
127 ordering, the change might break systems that relied on that.
129 * Support for two grammatical forms of month names has been added.
130 In a call to strftime, the "%B" and "%b" format specifiers will now
131 produce the grammatical form required when the month is used as part
132 of a complete date. New "%OB" and "%Ob" specifiers produce the form
133 required when the month is named by itself. For instance, in Greek
134 and in many Slavic and Baltic languages, "%B" will produce the month
135 in genitive case, and "%OB" will produce the month in nominative case.
137 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
138 are all valid and will all accept any known form of month
139 name---standalone or complete, abbreviated or full. In a call to
140 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
141 the strings used by "%B" and "%b", respectively. New query
142 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
143 used by "%OB" and "%Ob", respectively.
145 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
146 define the strings for %OB and %Ob, respectively; these have the
147 same syntax as "mon" and "abmon". These arrays are optional; if they
148 are not provided then they have the same content as "mon" and "abmon",
151 These features are provided for locales which define "alt_mon" and/or
152 "ab_alt_mon" in their locale source data. This release includes such
153 alternative month name data for the following languages: Belarusian,
154 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
156 This feature is currently a GNU extension, but it is expected to
157 be added to the next revision of POSIX, and it is also already
158 available on some BSD-derived operating systems.
160 This feature will cause existing statically compiled applications
161 to fail to load locales and fall back to the builtin C/POSIX locales.
162 See notes below for other changes affecting compatibility.
164 * Support for the RISC-V ISA running on Linux has been added. This port
165 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
166 for the following ISA and ABI pairs:
172 Deprecated and removed features, and other changes affecting compatibility:
174 * Statically compiled applications attempting to load locales compiled for the
175 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
176 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
177 support for two grammatical forms of the month names, also extends the locale
178 data binary format. Static applications needing locale support must be
179 recompiled to match the runtime and data they are deployed with. In some
180 distributions there is an upgrade window where dynamically linked applications
181 may use a new library but the old locale data and also fall back to the
182 builtin C/POSIX locales; restarting the application process is sufficient to
185 * Support for statically linked applications which call dlopen is deprecated
186 and will be removed in a future version of glibc. Applications which call
187 dlopen need to be linked dynamically instead.
189 * Support for old programs which use internal stdio data structures and
190 functions is deprecated. This includes programs which use the C++ streams
191 provided by libstdc++ in GCC 2.95. Programs which use the internal
192 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
193 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
194 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
195 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
196 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
197 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
198 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
199 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
200 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
201 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
202 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
203 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
204 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
205 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
206 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
207 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
208 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
209 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
210 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
211 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
212 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
213 future version of glibc. Unlike other symbol removals, these old
214 applications will not be supported using compatibility symbols.
216 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
217 defined by <sys/ptrace.h>.
219 * libm no longer supports SVID error handling (calling a user-provided
220 matherr function on error) or the _LIB_VERSION variable to control error
221 handling. (SVID error handling and the _LIB_VERSION variable still work
222 for binaries linked against older versions of the GNU C Library.) The
223 libieee.a library is no longer provided. math.h no longer defines struct
224 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
225 TLOSS, PLOSS and HUGE.
227 * The libm functions pow10, pow10f and pow10l are no longer supported for
228 new programs. Programs should use the standard names exp10, exp10f and
229 exp10l for these functions instead.
231 * The mcontext_t type is no longer the same as struct sigcontext. On
232 platforms where it was previously the same, this changes the C++ name
233 mangling for interfaces involving this type.
235 * The add-ons mechanism for building additional packages at the same time as
236 glibc has been removed. The --enable-add-ons configure option is now
239 * The --without-fp configure option is now ignored. Whether hardware
240 floating-point instructions are used is now configured based on whether
241 the compiler used at configure time (without any options implied by a
242 --with-cpu= configure option) uses such instructions.
244 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
245 the specified string can be parsed as a domain name.
247 * In the malloc_info output, the <heap> element may contain another <aspace>
248 element, "subheaps", which contains the number of sub-heaps.
250 * The libresolv function p_secstodate is no longer supported for new
253 * The tilepro-*-linux-gnu configuration is no longer supported.
255 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
256 and will be removed in a future release. Software that is still using
257 either header should be updated to use standard <stdio.h> interfaces
260 libio.h was originally the header for a set of supported GNU extensions,
261 but they have not been maintained as such in many years, they are now
262 standing in the way of improvements to stdio, and we don't think there are
263 any remaining external users. _G_config.h was never intended for public
264 use, but predates the bits convention.
266 Changes to build and runtime requirements:
268 * bison version 2.7 or later is required to generate code in the 'intl'
271 Security related changes:
273 CVE-2009-5064: The ldd script would sometimes run the program under
274 examination directly, without preventing code execution through the
275 dynamic linker. (The glibc project disputes that this is a security
276 vulnerability; only trusted binaries must be examined using the ldd
279 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
280 suffered from a one-byte overflow during ~ operator processing (either
281 on the stack or the heap, depending on the length of the user name).
282 Reported by Tim Rühsen.
284 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
285 would sometimes fail to free memory allocated during ~ operator
286 processing, leading to a memory leak and, potentially, to a denial
289 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
290 without GLOB_NOESCAPE, could write past the end of a buffer while
291 unescaping user names. Reported by Tim Rühsen.
293 CVE-2017-17426: The malloc function, when called with an object size near
294 the value SIZE_MAX, would return a pointer to a buffer which is too small,
295 instead of NULL. This was a regression introduced with the new malloc
296 thread cache in glibc 2.26. Reported by Iain Buclaw.
298 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
299 to the allocation of too much memory. (This is not a security bug per se,
300 it is mentioned here only because of the CVE assignment.) Reported by
303 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
304 of the number of search path components. (This is not a security
305 vulnerability per se because no trust boundary is crossed if the fix for
306 CVE-2017-1000366 has been applied, but it is mentioned here only because
307 of the CVE assignment.) Reported by Qualys.
309 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
310 for AT_SECURE or SUID binaries could be used to load libraries from the
313 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
314 succeeds without returning an absolute path due to unexpected behaviour
315 of the Linux kernel getcwd syscall. Reported by halfdog.
317 CVE-2018-6485: The posix_memalign and memalign functions, when called with
318 an object size near the value of SIZE_MAX, would return a pointer to a
319 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
321 CVE-2018-6551: The malloc function, when called with an object size near
322 the value of SIZE_MAX, would return a pointer to a buffer which is too
323 small, instead of NULL.
325 The following bugs are resolved with this release:
327 [866] glob: glob should match dangling symlinks
328 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
329 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
331 [5997] math: Very slow execution of sinf function
332 [10580] localedata: hr_HR: updated locale
333 [10871] locale: 'mon' array should contain both nominative and genitive
335 [12349] localedata: eu_ES: incorrect thousands separator
336 [13605] localedata: shn_MM: new Shan locale
337 [13805] localedata: ru_RU: currency should use ',' as radix point
338 [13953] localedata: km_KH: locale update
339 [13994] localedata: mjw_IN: new locale
340 [14121] build: make writes .mo files in po directory
341 [14333] libc: Fix the race between atexit() and exit()
342 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
343 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
345 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
346 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
347 width Latin characters
348 [15332] localedata: es_CU: locale update
349 [15436] stdio: Don't close or flush stdio streams on abort
350 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
352 [16148] localedata: ca_ES: incorrect thousands separator
353 [16750] dynamic-link: ldd should not try to execute the binaries
355 [16777] localedata: pl_PL: incorrect thousands separator in locale
356 [16905] localedata: hanzi: new collation
357 [17563] localedata: cmn_TW: add hanzi collation
358 [17750] localedata: wrong collation order of diacritics in most locales
359 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
361 [17956] build: Build fails on missing definitions from header file
362 nss/nss.h when Mozilla NSS is used for cryptography
363 [18203] libc: realpath() does not handle unreachable paths correctly
364 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
365 [18812] localedata: kab_DZ: new Kabyle Algeria locale
366 [18822] libc: Internal functions are called via PLT
367 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
368 [19170] libc: __gmon_start__ defined in hppa in crtn.S
369 [19574] libc: glibc should support building static PIE binaries
370 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
372 [19971] glob: glob: Do not skip entries with zero d_ino values
373 [19982] localedata: fr.po: spelling mistake for error code EXDEV
374 [20008] localedata: km_KH: convert to translit_neutral
375 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
376 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
377 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
378 [20482] localedata: de_CH: abbreviated weekdays should be two letters
379 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
381 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
382 they're set, wrong errors returned by gaih_inet when lookup functions
384 [20756] localedata: [PATCH] Use Unicode wise thousands separator
385 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
387 [20952] localedata: yuw_PG: new locale
388 [21084] localedata: charmaps/IBM858: new codepage
389 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
390 [21242] libc: assert gives pedantic warning in old gcc versions
391 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
392 __regcall calling convention
393 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
394 [21326] libc: C99 functions are not declared for C++11 and later if
395 _GNU_SOURCE is not predefined by g++
396 [21457] libc: sys/ucontext.h namespace
397 [21530] libc: tmpfile() should be implemented using O_TMPFILE
398 [21660] math: GCC fails to compile a formula with tgmath.h
399 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
400 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
401 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
402 [21684] math: tgmath.h handling of complex integers
403 [21685] math: tgmath.h handling of bit-fields
404 [21686] math: tgmath.h handling of __int128
405 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
407 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
409 [21750] localedata: column width of characters incompatible with classical
411 [21754] malloc: malloc: Perform as little work as possible after heap
412 consistency check failures
413 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
414 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
415 [21791] string: Unused XXX_chk_XXX functions in libc.a
416 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
418 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
420 [21845] localedata: Added new Locale bho_NP
421 [21853] localedata: Fix abday Which looks same as day in zh_SG
422 [21854] localedata: Added New Locale en_SC
423 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
424 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
425 _dl_runtime_resolve_avx_slow
426 [21885] network: getaddrinfo: gethosts does not release resolver context
427 on memory allocation failure
428 [21899] libc: XPG4.2 sigaction namespace
429 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
431 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
433 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
434 errno value, causing getaddrinfo to fail
435 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
436 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
437 instead of EAI_NODATA
438 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
439 constant PTRACE_SEIZE_DEVEL
440 [21930] math: C-only gcc builtins used in <math.h> isinf
441 [21932] network: Unpaired __resolv_context_get in generic get*_r
443 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
444 [21944] libc: sigval namespace
445 [21951] localedata: Update hanzi collation by stroke
446 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
447 sysdeps/x86_64/fpu/e_expf.S
448 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
449 allocation in syscall loops)
450 [21959] localedata: Fix Country name for xh_ZA
451 [21960] localedata: Fix abmon for bem_ZM
452 [21966] math: AVX2 mathvec functions use FMA without checking
453 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
454 [21971] localedata: Added New Locale for mfe_MU
455 [21972] libc: assert macro requires operator== (int) for its argument type
456 [21973] math: [sparc] libm missing sqrtl compat symbol
457 [21974] libc: Remove __bb_init_func and __bb_exit_func
458 [21982] string: stratcliff.c: error: assuming signed overflow does not
460 [21986] stdio: __guess_grouping is called incorrectly
461 [21987] math: [sparc32] wrong bits/long-double.h installed
462 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
464 [22022] localedata: Missing country_name for mni_IN
465 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
467 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
469 [22026] locale: iconv_open: heap overflow on gconv_init failure
470 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
471 [22035] math: [m68k] bits/math-inline.h macro namespace
472 [22038] localedata: Fix abbreviated weeks and months for Somali
473 [22044] localedata: Remove redundant data for Limburgish Language
474 [22050] malloc: Linking with -lmcheck does not hook
475 __malloc_initialize_hook correctly
476 [22051] libc: zero terminator in the middle of glibc's .eh_frame
477 [22052] malloc: malloc failed to compile with GCC 7 and -O3
478 [22070] localedata: charmaps/UTF-8: wcwidth for
479 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
480 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
481 Jungseong and Jongseong) should be 0
482 [22078] nss: nss_files performance issue in multi mode
483 [22082] math: bits/math-finite.h exp10 condition
484 [22086] libc: pcprofiledump incorrect cross-endian condition
485 [22093] dynamic-link: ld.so no longer searches in .../x86_64
486 [22095] network: Name server address allocation memory leak in resolv.conf
488 [22096] network: __resolv_conf_attach can incorrectly free passed conf
490 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
491 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
493 [22111] malloc: malloc: per thread cache is not returned when thread exits
494 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
495 [22134] libc: [linux] implement fexecve with execveat
496 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
498 [22145] libc: ttyname() gives up too early in the face of namespaces
499 [22146] math: C++ build issue with float128 on x86_64
500 [22153] nptl: nptl: save error code before process termination
501 [22156] libc: [hppa,ia64,microblaze] Executable stack default
502 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
503 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
504 [22165] libc: [hppa] Text relocations in libc.so
505 [22180] libc: destructor registered via __cxa_atexit is called twice
506 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
507 symlinks") cause "make" segfaults
508 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
510 [22207] libc: FAIL: stdlib/test-atexit-race
511 [22225] math: nearbyint arithmetic moved before feholdexcept
512 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
513 [22235] math: iscanonical in C++ and float128
514 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
515 the svid compat wrapper
516 [22244] math: ynf and yn are wrong without the svid compat wrapper
517 [22273] libc: Improper assert in Linux posix_spawn implementation
518 [22284] libc: -pg -pie doesn't work
519 [22292] locale: localedef exits with error 4 when it should be error 1
520 [22294] locale: Allow "" for int_currency_symbol definition in locales.
521 [22295] locale: Don't warn on non-symbolic characters in locale sources in
523 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
524 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
526 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
527 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
528 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
529 [22322] libc: [mips64] wrong bits/long-double.h installed
530 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
531 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
533 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
534 [22343] malloc: Integer overflow in posix_memalign
535 [22347] libc: getrandom() returns the number of bytes that were copied to
536 the buffer even though the comments say "Return 0 on success and -1 on
538 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
539 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
541 [22370] dynamic-link: Incorrect note padding check
542 [22375] libc: malloc returns pointer from tcache_get when should return
543 NULL (CVE-2017-17426)
544 [22377] math: iseqsig, float128 and C++
545 [22382] localedata: Error in tpi_PG locale
546 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
547 inside the ASCII printable range
548 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
550 [22403] localedata: Slash needs escaping in some locales
551 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
552 [22409] network: res_hnok does not accept some host names used on the
554 [22412] network: res_dnok, res_hnok should perform syntax checks
555 [22413] network: ns_name_pton ignores syntactically invalid trailing
557 [22415] stdio: setvbuf can lead to invalid free/segfault
558 [22432] build: Non-deterministic build
559 [22439] malloc: malloc_info should compute summary statistics for all sub-
561 [22442] network: if_nametoindex could report index for the wrong
563 [22446] build: aliasing violation calling readlink in handle_request
564 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
565 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
566 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
567 __nscd_hash/__nss_hash
568 [22463] network: p_secstodate overflow handling
569 [22469] localedata: pl_PL LC_COLLATE does not use i18n
570 [22478] libc: sigwait can fail with EINTR
571 [22505] libc: ldconfig processes include directive in locale-specific
573 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
574 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
575 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
576 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
577 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
578 [22534] localedata: Collation rules for Serbian and Bosnian should be the
580 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
581 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
582 [22577] libc: missing newline after "cannot allocate TLS data structures
584 [22588] manual: manual/conf.texi: missing underscore in front of
586 [22593] math: nextafter and nexttoward are declared with const attribute
587 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
588 [22603] string: ia64 memchr overflows internal pointer check
589 [22605] libc: SH clone does not set the exit code correctly
590 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
592 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
593 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
594 modified in case of success
595 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
596 [22615] manual: manual: ambiguous wording about errno value in case of
598 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
600 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
601 binaries or if /proc is not mounted (CVE-2017-16997)
602 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
603 [22630] build: $(no-pie-ldflag) is no longer effective
604 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
605 [22635] nptl: pthread_self returns NULL before libpthread is loaded
606 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
607 [22637] nptl: guard size is subtracted from thread stack size instead of
609 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
610 [22657] localedata: hu_HU: Avoid double space in date
611 [22660] math: fmax, fmin sNaN handling on alpha
612 [22664] libc: New warning of GCC8
613 [22665] math: alpha: ceil and floor raise inexact exceptions
614 [22666] math: alpha: trunc raise inexact exceptions
615 [22667] libc: makecontext lacks stack alignment on i386
616 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
617 [22679] libc: getcwd(3) can succeed without returning an absolute path
619 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
621 [22687] math: [powerpc-nofpu] complex long double functions spurious
623 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
624 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
625 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
627 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
628 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
630 [22701] nis: Incomplete removal of libnsl
631 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
632 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
633 [22715] dynamic-link: FAIL: elf/tst-audit10
634 [22719] libc: Backtrace tests fail on hppa
635 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
636 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
637 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
638 before the first call to crypt_r () in crypt/badsalttest.c
645 * A per-thread cache has been added to malloc. Access to the cache requires
646 no locks and therefore significantly accelerates the fast path to allocate
647 and free small amounts of memory. Refilling an empty cache requires locking
648 the underlying arena. Performance measurements show significant gains in a
649 wide variety of user workloads. Workloads were captured using a special
650 instrumented malloc and analyzed with a malloc simulator. Contributed by
651 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
653 * Unicode 10.0.0 Support: Character encoding, character type info, and
654 transliteration tables are all updated to Unicode 10.0.0, using
655 generator scripts contributed by Mike FABIAN (Red Hat).
656 These updates cause user visible changes, especially the changes in
657 wcwidth for many emoji characters cause problems when emoji sequences
658 are rendered with pango, see for example:
659 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
661 * Collation of Hungarian has been overhauled and is now consistent with "The
662 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
665 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
667 - The GNU C Library will now detect when /etc/resolv.conf has been
668 modified and reload the changed configuration. The new resolver option
669 “no-reload” (RES_NORELOAD) disables this behavior.
671 - The GNU C Library now supports an arbitrary number of search domains
672 (configured using the “search” directive in /etc/resolv.conf);
673 previously, there was a hard limit of six domains. For backward
674 compatibility, applications that directly modify the ‘_res’ global
675 object are still limited to six search domains.
677 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
678 Library will now randomly pick a name server from the configuration as a
679 starting point. (Previously, the second name server was always used.)
681 * The tunables feature is now enabled by default. This allows users to tweak
682 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
684 * New function reallocarray, which resizes an allocated block (like realloc)
685 to the product of two sizes, with a guaranteed clean failure upon integer
686 overflow in the multiplication. Originally from OpenBSD, contributed by
687 Dennis Wölfing and Rüdiger Sonderfeld.
689 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
690 These are extended versions of preadv and pwritev, respectively, taking an
691 additional flags argument. The set of supported flags depends on the
692 running kernel; full support currently requires kernel 4.7 or later.
694 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
695 create a new session ID for the spawned process. This feature is
696 scheduled to be added to the next major revision of POSIX; for the time
697 being, it is available under _GNU_SOURCE.
699 * errno.h is now safe to use from C-preprocessed assembly language on all
700 supported operating systems. In this context, it will only define the
701 Exxxx constants, as preprocessor macros expanding to integer literals.
703 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
704 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
705 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
706 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
708 To compile programs that use this feature, the compiler must support
709 128-bit floating point with the type name _Float128 (as defined by TS
710 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
711 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
712 must be defined to make the new interfaces visible.
714 The new functions and macros correspond to those present for other
715 floating-point types (except for a few obsolescent interfaces not
716 supported for the new type), with F128 or f128 suffixes; for example,
717 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
718 printf or scanf formats for the new type; the strfromf128 and strtof128
719 interfaces should be used instead.
721 Deprecated and removed features, and other changes affecting compatibility:
723 * The synchronization that pthread_spin_unlock performs has been changed to
724 now be equivalent to a C11 atomic store with release memory order to the
725 spin lock's memory location. Previously, several (but not all)
726 architectures used stronger synchronization (e.g., containing what is
727 often called a full barrier). This change can improve performance, but
728 may affect odd fringe uses of spin locks that depend on the previous
729 behavior (e.g., using spin locks as atomic variables to try to implement
730 Dekker's mutual exclusion algorithm).
732 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
735 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
736 will only be built and installed when the GNU C Library is configured with
737 --enable-obsolete-rpc. This allows alternative RPC implementations, such
738 as TIRPC or rpcsvc-proto, to be used.
740 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
741 libnss_compat, are deprecated, and will not be built or installed by
744 The NIS(+) support library, libnsl, is also deprecated. By default, a
745 compatibility shared library will be built and installed, but not headers
746 or development libraries. Only a few NIS-related programs require this
747 library. (In particular, the GNU C Library has never required programs
748 that use 'gethostbyname' to be linked with libnsl.)
750 Replacement implementations based on TIRPC, which additionally support
751 IPv6, are available from <https://github.com/thkukuk/>. The configure
752 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
753 name service modules, to be built and installed.
755 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
756 support is enabled, the configured recursive resolver must support EDNS.
757 (Responding to EDNS-enabled queries with responses which are not
758 EDNS-enabled is fine, but FORMERR responses are not.)
760 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
761 servers have not supported this opcode for a long time.
763 * The _res_opcodes variable has been removed from libresolv. It had been
764 exported by accident.
766 * <string.h> no longer includes inline versions of any string functions,
767 as this kind of optimization is better done by the compiler. The macros
768 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
770 * The nonstandard header <xlocale.h> has been removed. Most programs should
771 use <locale.h> instead. If you have a specific need for the definition of
772 locale_t with no other declarations, please contact
773 libc-alpha@sourceware.org and explain.
775 * The obsolete header <sys/ultrasound.h> has been removed.
777 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
779 * The obsolete function cfree has been removed. Applications should use
782 * The stack_t type no longer has the name struct sigaltstack. This changes
783 the C++ name mangling for interfaces involving this type.
785 * The ucontext_t type no longer has the name struct ucontext. This changes
786 the C++ name mangling for interfaces involving this type.
788 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
789 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
790 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
791 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
792 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
793 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
794 fpregset_t type no longer has the name struct fpu. This changes the C++
795 name mangling for interfaces involving those types.
797 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
798 synced with the kernel:
800 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
801 are not supported on this architecture and have been removed.
803 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
804 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
805 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
807 Programs that assume the GET/SETREGS ptrace requests are universally
808 available will now fail to build, instead of malfunctioning at runtime.
810 Changes to build and runtime requirements:
812 * Linux kernel 3.2 or later is required at runtime, on all architectures
813 supported by that kernel. (This is a change from version 2.25 only for
816 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
818 * On most architectures, GCC 4.9 or later is required to build the GNU C
819 Library. On powerpc64le, GCC 6.2 or later is required.
821 Older GCC versions and non-GNU compilers are still supported when
822 compiling programs that use the GNU C Library. (We do not know exactly
823 how old, and some GNU extensions to C may be _de facto_ required. If you
824 are interested in helping us make this statement less vague, please
825 contact libc-alpha@sourceware.org.)
827 Security related changes:
829 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
830 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
832 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
833 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
835 * Avoid printing a backtrace from the __stack_chk_fail function since it is
836 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
839 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
840 fixed (CVE-2017-12133).
842 The following bugs are resolved with this release:
844 [984] network: Respond to changed resolv.conf in gethostbyname
845 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
846 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
848 [12189] libc: __stack_chk_fail should not attempt a backtrace
850 [14096] time: Race condition on timezone/tst-timezone.out
851 [14172] localedata: az_IR: new locale
852 [14995] build: glibc fails to build if gold is the default linker, even if
854 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
855 [16637] network: inet_pton function is accepting IPv6 with bad format
856 [16640] string: string/strtok.c: undefined behaviour inconsistent between
857 x86 and other generic code
858 [16875] localedata: ko_KR: fix lang_name
859 [17225] localedata: ar_SY: localized month names for May and June are
861 [17297] localedata: da_DK: wrong date_fmt string
862 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
863 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
864 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
865 [19066] localedata: ar_SA abbreviated day and month names are in English
866 [19569] network: resolv: Support an arbitrary number of search domains
867 [19570] network: Implement random DNS server selection in the stub
869 [19838] locale: localedef fails on PA-RISC
870 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
871 order of 0D36 and 0D37
872 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
874 [20098] libc: FAIL: debug/backtrace-tst on hppa
875 [20257] network: sunrpc: clntudp_call does not enforce timeout when
877 [20275] localedata: locale day/abday/mon/abmon should not have trailing
879 [20313] localedata: Update locale data to Unicode 9.0
880 [20424] manual: Document how to provide a malloc replacement
881 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
883 [20686] locale: Add el_GR@euro to SUPPORTED.
884 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
885 failures consistently
886 [21015] dynamic-link: Document and fix --enable-bind-now
887 [21016] nptl: pthread_cond support is broken on hppa
888 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
889 frame-pointer on i386
890 [21049] libc: segfault in longjmp_chk() due to clobbered processor
892 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
893 [21088] libc: Build fails with --enable-static-nss
894 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
895 generic c code is used
896 [21109] libc: Tunables broken on big-endian
897 [21112] math: powf has large ulp errors with base close to 1 and exponent
899 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
901 [21120] malloc: glibc malloc is incompatible with GCC 7
902 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
904 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
905 (-0.0) when linking with -lieee
906 [21171] math: log10, log2 and lgamma return incorrect results
907 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
908 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
909 [21207] localedata: ce_RU: update weekdays from CLDR
910 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
911 [21217] localedata: Update months from CLDR-31
912 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
913 [21243] libc: support_delete_temp_file should issue warning for failed
915 [21244] libc: support resolv_test_start() socket fd close should be
917 [21253] libc: localedef randomly segfaults when using -fstack-check due to
918 new posix_spawn implementation
919 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
920 leads to lower CPU frequency
921 [21259] libc: [alpha] termios.h missing IXANY for POSIX
922 [21261] libc: [sparc64] bits/setjmp.h namespace
923 [21267] network: [mips] bits/socket.h IOC* namespace
924 [21268] libc: [alpha] termios.h NL2, NL3 namespace
925 [21270] libc: mmap64 silently truncates large offset values
926 [21275] libc: posix_spawn always crashes on ia64 now
927 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
928 [21280] math: [powerpc] logbl for POWER7 return incorrect results
929 [21289] libc: Incorrect declaration for 32-bit platforms with
930 _FILE_OFFSET_BITS=64 causes build error
931 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
932 not support gethostbyname4_r
933 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
935 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
937 [21340] libc: Support POSIX_SPAWN_SETSID
938 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
939 [21359] network: ns_name_pack needs additional byte in destination buffer
940 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
941 against fragmentation attacks (CVE-2017-12132)
942 [21369] network: resolv: Remove EDNS fallback
943 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
945 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
946 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
947 [21393] stdio: Missing dup3 error check in freopen, freopen64
948 [21396] libc: Use AVX2 memcpy/memset on Skylake server
949 [21399] localedata: Bad description for U00EC in
950 localedata/charmaps/CP1254
951 [21411] malloc: realloc documentation error
952 [21426] network: sys/socket.h uio.h namespace
953 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
954 [21445] libc: signal.h bsd_signal namespace
955 [21455] network: Network headers stdint.h namespace
956 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
958 [21475] network: resolv: Overlong search path is truncated mid-label
959 [21511] libc: sigstack namespace
960 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
961 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
962 bits/syscall.h: No such file or directory
963 [21517] libc: struct sigaltstack namespace
964 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
965 [21533] localedata: Update locale data to Unicode 10.0
967 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
968 junk at end of line, first unrecognized character is `@'
969 [21538] libc: SIG_HOLD missing for XPG4
970 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
971 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
972 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
973 [21543] libc: sigevent namespace
974 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
975 size for general purpose registers in mcontext_t structure
976 [21550] libc: sigwait namespace
977 [21552] libc: XPG4 bsd_signal namespace
978 [21554] libc: sigpause namespace
979 [21560] libc: sys/wait.h signal.h namespace
980 [21561] libc: waitid namespace
981 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
982 '/usr/include/stdlib.h': Permission denied
983 [21575] libc: sys/wait.h missing struct rusage definition
984 [21584] libc: sigaltstack etc namespace
985 [21597] libc: siginterrupt namespace
986 [21607] math: hppa: FAIL: math/test-tgmath
987 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
989 [21622] libc: [tile] missing SA_* for POSIX.1:2008
990 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
991 stack and heap (CVE-2017-1000366)
992 [21625] libc: wait3 namespace
993 [21654] nss: Incorrect pointer alignment in NSS group merge result
995 [21657] network: Parse interface zone id for node-local multicast
996 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
998 [21666] libc: .symver is used on common symbol
999 [21668] network: resolv: res_init cross-thread broadcast introduces race
1001 [21687] math: tgmath.h totalorder, totalordermag return type
1002 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1004 [21696] libc: Incorrect assumption of of __cpu_mask in
1005 posix/sched_cpucount.c
1006 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1007 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1009 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1010 leading to relocation crash
1011 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1012 `resolv_conf_matches (resp, conf)' failed.
1013 [21710] localedata: Added Samoan language locale for Samoa
1014 [21711] localedata: Pashto yesstr/nostr locale are missing
1015 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1016 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1017 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1018 [21723] localedata: yesstr/nostr missing for Chinese language locale
1019 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1021 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1023 [21728] localedata: New Locale for Tongan language
1024 [21729] localedata: incorrect LC_NAME fields for hi_IN
1025 [21733] localedata: yesstr and nostr are missing for zh_HK
1026 [21734] localedata: Missing yesstr and nostr are for kw_GB
1027 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1028 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1029 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1030 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1032 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1034 [21749] localedata: Wrong abbreviated day name (“abday”) for
1036 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1037 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1038 [21759] localedata: missing yesstr and nostr for Tigrinya
1039 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1040 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1041 [21767] localedata: Missing Bislama locales
1042 [21768] localedata: Missing yesstr and nostr for aa_ET
1043 [21770] localedata: Missing Field in li_NL
1044 [21778] nptl: Robust mutex may deadlock
1045 [21779] libc: MicroBlaze segfaults when loading libpthread
1046 [21783] localedata: Fix int_select international_call_prefixes
1047 [21784] localedata: Inconsistency in country_isbn
1048 [21788] localedata: Missing Country Postal Abbreviations
1049 [21794] localedata: Added-country_isbn-for-Italy
1050 [21795] localedata: Add/Fix country_isbn for France
1051 [21796] localedata: Added country_isbn for Republic of Korea
1052 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1053 [21799] localedata: Added int_select international_call_prefixes
1054 [21801] localedata: Added int_select international_call_prefixes
1055 [21804] nptl: Double semicolon in thread-shared-types.h
1056 [21807] localedata: LC_ADDRESS fix for pap_CW
1057 [21808] localedata: Fix LC_ADDRESS for pap_AW
1058 [21821] localedata: Added country_name in mai_IN
1059 [21822] localedata: Fix LC_TIME for mai_IN
1060 [21823] localedata: missing yesstr, nostr for sa_IN
1061 [21825] localedata: Fix name_mrs for mag_IN
1062 [21828] localedata: 2.26 changelog should mention user visible changes
1064 [21835] localedata: Added Maithili language locale for Nepal
1065 [21838] localedata: Removed redundant data for the_NP
1066 [21839] localedata: Fix LC_MONETARY for ta_LK
1067 [21844] localedata: Fix Latin characters and Months Sequence.
1068 [21848] localedata: Fix mai_NP Title Name
1073 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1074 24731-2:2010, is supported to enable declarations of functions from that
1075 TR. Note that not all functions from that TR are supported by the GNU C
1078 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1079 18661-1:2014, is supported to enable declarations of functions and macros
1080 from that TS. Note that not all features from that TS are supported by
1083 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1084 18661-4:2015, is supported to enable declarations of functions and macros
1085 from that TS. Note that most features from that TS are not supported by
1088 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1089 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1090 Since the GNU C Library defaults to a much newer revision of POSIX, this
1091 will only affect programs that specifically request an old conformance
1092 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1093 see a change in the visible declarations, but a program compiled with
1094 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1097 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1098 defined by all multithreaded code, but glibc has not required this for
1101 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1102 means that in a future release, the macros “major”, “minor”, and “makedev”
1103 will only be available from <sys/sysmacros.h>.
1105 These macros are not part of POSIX nor XSI, and their names frequently
1106 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1107 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1108 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1111 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1112 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1113 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1115 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1116 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1117 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1118 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1119 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1120 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1121 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1122 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1123 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1124 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1125 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1128 * New <math.h> features are added from TS 18661-1:2014:
1130 - Signaling NaN macros: SNANF, SNAN, SNANL.
1132 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1133 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1134 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1136 - llogb functions: the llogb, llogbf and llogbl functions, and the
1137 FP_LLOGB0 and FP_LLOGBNAN macros.
1139 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1142 - Comparison macros: iseqsig.
1144 - Classification macros: iscanonical, issubnormal, iszero.
1146 - Total order functions: totalorder, totalorderf, totalorderl,
1147 totalordermag, totalordermagf, totalordermagl.
1149 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1151 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1152 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1154 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1155 are added to libc. They convert a floating-point number into string.
1157 * Most of glibc can now be built with the stack smashing protector enabled.
1158 It is recommended to build glibc with --enable-stack-protector=strong.
1159 Implemented by Nick Alcock (Oracle).
1161 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1162 intended to be used instead of memset() to erase sensitive data after use;
1163 the compiler will not optimize out calls to explicit_bzero even if they
1164 are "unnecessary" (in the sense that no _correct_ program can observe the
1165 effects of the memory clear).
1167 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1168 to float instead of double. This does not affect the ABI of any libraries
1169 that are part of the GNU C Library, but may affect the ABI of other
1170 libraries that use this type in their interfaces.
1172 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1173 float_t and double_t types are now defined to long double instead of float
1174 and double. These options are not the default, and this does not affect
1175 the ABI of any libraries that are part of the GNU C Library, but it may
1176 affect the ABI of other libraries that use this type in their interfaces,
1177 if they are compiled or used with those options.
1179 * The getentropy and getrandom functions, and the <sys/random.h> header file
1182 * The buffer size for byte-oriented stdio streams is now limited to 8192
1183 bytes by default. Previously, on Linux, the default buffer size on most
1184 file systems was 4096 bytes (and thus remains unchanged), except on
1185 network file systems, where the buffer size was unpredictable and could be
1186 as large as several megabytes.
1188 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1189 for the Linux quota interface which predates kernel version 2.4.22 has
1192 * The malloc_get_state and malloc_set_state functions have been removed.
1193 Already-existing binaries that dynamically link to these functions will
1194 get a hidden implementation in which malloc_get_state is a stub. As far
1195 as we know, these functions are used only by GNU Emacs and this change
1196 will not adversely affect already-built Emacs executables. Any undumped
1197 Emacs executables, which normally exist only during an Emacs build, should
1198 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1200 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1201 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1202 “no-ip6-dotint” had already been the default, and support for the
1203 “ip6-dotint” option was removed from the Internet in 2006.
1205 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1206 flag from <resolv.h> have been removed. The option relied on a
1207 backwards-incompatible DNS extension which was never deployed on the
1210 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1211 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1212 They were already unimplemented.
1214 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1215 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1216 removed again from the IETF name lookup interface specification in RFC
1217 2553. Applications should use getaddrinfo instead.
1219 * DNSSEC-related declarations and definitions have been removed from the
1220 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1221 decode the data part of DNSSEC record types. Previous versions of glibc
1222 only implemented minimal support for the previous version of DNSSEC, which
1223 is incompatible with the currently deployed version.
1225 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1226 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1227 <arpa/nameser.h> header file because the distinction between RR types and
1228 meta-RR types is not officially standardized, subject to revision, and
1229 thus not suitable for encoding in a macro.
1231 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1232 and rhook members of the res_state type in <resolv.h> have been removed.
1233 The glibc stub resolver did not support these hooks, but the header file
1234 did not reflect that.
1236 * For multi-arch support it is recommended to use a GCC which has
1237 been built with support for GNU indirect functions. This ensures
1238 that correct debugging information is generated for functions
1239 selected by IFUNC resolvers. This support can either be enabled by
1240 configuring GCC with '--enable-gnu-indirect-function', or by
1241 enabling it by default by setting 'default_gnu_indirect_function'
1242 variable for a particular architecture in the GCC source file
1245 * GDB pretty printers have been added for mutex and condition variable
1246 structures in POSIX Threads. When installed and loaded in gdb these pretty
1247 printers show various pthread variables in human-readable form when read
1248 using the 'print' or 'display' commands in gdb.
1250 * Tunables feature added to allow tweaking of the runtime for an application
1251 program. This feature can be enabled with the '--enable-tunables' configure
1252 flag. The GNU C Library manual has details on usage and README.tunables has
1253 instructions on adding new tunables to the library.
1255 * A new version of condition variables functions have been implemented in
1256 the NPTL implementation of POSIX Threads to provide stronger ordering
1259 * A new version of pthread_rwlock functions have been implemented to use a more
1260 scalable algorithm primarily through not using a critical section anymore to
1263 Security related changes:
1265 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1266 have been created with makecontext could fail to terminate due to a
1267 missing .cantunwind annotation. This has been observed to lead to a hang
1268 (denial of service) in some Go applications compiled with gccgo. Reported
1269 by Andreas Schwab. (CVE-2016-6323)
1271 * The DNS stub resolver functions would crash due to a NULL pointer
1272 dereference when processing a query with a valid DNS question type which
1273 was used internally in the implementation. The stub resolver now uses a
1274 question type which is outside the range of valid question type values.
1277 The following bugs are resolved with this release:
1279 [4099] stdio: Overly agressive caching by stream i/o functions.
1280 [7065] build: Support building glibc with -fstack-protector or -fstack-
1282 [9842] localedata: en_CA: incorrect date format
1283 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1284 before it started waiting
1285 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1286 [15765] nptl: sem_open is wrongly a cancellation point
1287 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1288 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1290 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1291 library linked with pthread
1292 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1293 "FIXME: Ingo" issue)
1294 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1295 [17252] libc: getrandom and getentropy syscall
1296 [17730] malloc: thread-local storage is sometimes improperly free()'d
1297 after being __libc_memalign()'d
1298 [18241] stdio: failed fseek on memstream does not set errno and can fail
1300 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1302 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1303 PTHREAD_PRIO_INHERIT mutex on ARM
1304 [18784] network: res_query and related function crash for special record
1305 type queries (CVE-2015-5180)
1306 [19380] math: strtod does not raise "inexact"
1307 [19387] string: Integer overflow in memchr
1308 [19390] string: Integer overflow in strncat
1309 [19398] build: linknamespace tests fail with massively parallel build
1310 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1312 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1313 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1314 [19514] libc: [PATCH] Fix spelling errors in spelling
1315 "implement"/"implementation" in several places
1316 [19582] network: Deprecate RES_USE_INET6
1317 [19673] manual: clog10 docs appear to be erroneous
1318 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1319 causes a segmentation fault
1320 [19826] libc: invalid pointer returned from __tls_get_addr with static
1322 [20016] network: resolv: Remove hooks support from the API
1323 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1325 [20033] math: [x86_64] vectorized math function don't call the __finite
1327 [20116] nptl: use after free in pthread_create
1328 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1329 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1331 [20311] nptl: please install proc_service.h
1332 [20366] libc: Compilation errors in installed headers in strict-compliance
1334 [20370] malloc: malloc: Arena free list management is still racy
1335 (incorrect fix in bug 19243)
1336 [20386] libc: assert (X = 0) does not result in GCC warning
1337 [20432] malloc: malloc: Minimize interface required for interposition
1338 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1339 in _Unwind_Backtrace (CVE-2016-6323)
1340 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1341 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1343 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1344 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1345 [20477] network: resolv: incorrect double-checked locking related to
1347 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1349 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1351 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1352 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1353 cause transition penalty
1354 [20517] math: sparcv9 missing fdiml compat symbol
1355 [20524] manual: strverscmp is inconsistent
1356 [20525] libc: <sys/quota.h> should be based on kernel headers
1357 [20539] math: GCC 7 -static -lm fails to link at -O3
1358 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1360 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1362 [20591] network: Remove obsolete DNSSEC support
1363 [20592] network: DNS resource record type classification macros in
1364 <arpa/nameser.h> are incorrect
1365 [20593] network: Update DNS RR type definitions
1366 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1367 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1369 [20629] network: libresolv: Remove support for bitstring labels
1371 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1372 assertion: "maxidx >= 2"
1373 [20660] math: [arm] Use VSQRT
1374 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1375 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1376 [20689] libc: Test for FMA should also check for AVX.
1377 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1378 [20715] math: iszero macro breaks existing code
1379 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1380 [20728] libc: powerpc: Missing TOC stub in clone
1381 [20729] build: glibc-2.24 fails to build for i486 with -Os
1382 [20750] build: Build fails with default PIE enabled gcc-6.x
1383 [20768] math: [alpha] sqrt fegetenv namespace
1384 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1386 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1387 __FLT_EVAL_METHOD__ is 2
1388 [20790] network: rpcgen buffer overrun in get_prog_declaration
1389 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1390 [20829] libc: crypt snprintf namespace
1391 [20847] libc: tst-vfork3 failure
1392 [20855] math: Default bits/mathdef.h has inappropriate float_t
1393 [20859] math: [sh4] FP_ILOGB0 invalid
1394 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1395 U+20AC), not same as GBK
1396 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1397 [20916] math: pow handling of sNaN arguments
1398 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1399 [20919] math: Bad pow (qNaN, 0) result with -lieee
1400 [20940] math: hypot sNaN handling
1401 [20947] math: fmax, fmin sNaN handling
1402 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1403 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1405 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1406 [20973] nptl: robust mutexes: Lost wake-ups
1407 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1408 [20978] nis: strlen on null pointer in nss_nisplus
1409 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1411 [21014] string: i686 memchr overflows internal pointer check
1412 [21019] libc: [mips] n32 lseek incorrect on overflow
1413 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1414 [21026] libc: [MIPS] readahead syscall is broken on n64
1415 [21028] math: Fallback fesetexceptflag should always succeed
1416 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1417 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1419 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1420 [21061] librt: [microblaze] librt lost clock_* exports
1421 [21073] libc: tunables: insecure environment variables passed to
1422 subprocesses with AT_SECURE
1423 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1427 * The minimum Linux kernel version that this version of the GNU C Library
1428 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1429 kernel version 2.6.32 or later suffices (on architectures that already
1430 required kernel versions more recent than 3.2, those requirements remain
1431 unchanged). Linux 3.2 or later kernel headers are required on all
1434 * The pap_AN locale has been deleted. This has been deprecated for a long
1435 time. It has been replaced by pap_AW & pap_CW, both of which have long
1436 been included in previous releases.
1438 * The readdir_r and readdir64_r functions have been deprecated. It is
1439 recommended to use readdir and readdir64 instead.
1441 * The type “union wait” has been removed. It was deprecated in the early
1442 1990s and never part of POSIX. Application code should use the int type
1443 instead of “union wait”.
1445 * A new NSS action is added to facilitate large distributed system
1446 administration. The action, MERGE, allows remote user stores like LDAP
1447 to be merged into local user stores like /etc/groups in order to provide
1448 easy to use, updated, and managed sets of merged credentials. The new
1449 action can be used by configuring it in /etc/nsswitch.conf:
1450 group: files [SUCCESS=merge] nis
1451 Implemented by Stephen Gallagher (Red Hat).
1453 * The deprecated __malloc_initialize_hook variable has been removed from the
1456 * The long unused localedef --old-style option has been removed. It hasn't
1457 done anything in over 16 years. Scripts using this option can safely
1460 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1461 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1462 return the next representable value in the direction of positive infinity
1463 and the nextdown functions return the next representable value in the
1464 direction of negative infinity. These are currently enabled as GNU
1467 Security related changes:
1469 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1470 could result in a stack overflow when getnetbyname was called with an
1471 overly long name. (CVE-2016-3075)
1473 * Previously, getaddrinfo copied large amounts of address data to the stack,
1474 even after the fix for CVE-2013-4458 has been applied, potentially
1475 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1476 instead. Reported by Michael Petlan. (CVE-2016-3706)
1478 * The glob function suffered from a stack-based buffer overflow when it was
1479 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1480 Reported by Alexander Cherepanov. (CVE-2016-1234)
1482 * The Sun RPC UDP client could exhaust all available stack space when
1483 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1484 alloca plugin for GCC. (CVE-2016-4429)
1486 * The IPv6 name server management code in libresolv could result in a memory
1487 leak for each thread which is created, performs a failing naming lookup,
1488 and exits. Over time, this could result in a denial of service due to
1489 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1491 The following bugs are resolved with this release:
1493 [1170] localedata: ne_NP: update Nepali locale definition file
1494 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1496 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1497 [6796] math: fdim() does not set errno on overflow
1498 [10354] libc: posix_spawn should use vfork() in more cases than presently
1499 [11213] localedata: localedata: add copyright disclaimer to locale files
1500 [12143] localedata: chr_US: new Cherokee locale
1501 [12450] localedata: sgs_LT: new locale
1502 [12676] localedata: ln_CD: new locale
1503 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1505 [13304] math: fma, fmaf, fmal produce wrong results
1506 [14259] build: --localedir arg to configure is ignored
1507 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1508 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1509 [14934] localedata: es_CL: wrong first weekday chilean locale
1510 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1512 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1514 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1515 [15368] nptl: raise() is not async-signal-safe
1516 [15479] math: ceil, floor, round and trunc raise inexact exception
1517 [15578] localedata: kk_KZ: various updates
1518 [16003] localedata: pap_AN: punt old locale
1519 [16137] localedata: iw_IL: punt old locale
1520 [16190] localedata: eo: new esperanto locale
1521 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1522 [16742] malloc: race condition: pthread_atfork() called before first
1523 malloc() results in unexpected locking behaviour/deadlocks
1524 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1526 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1527 [17565] localedata: pt_PT: wrong (work-)week start
1528 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1529 [17950] build: Build fails with -msse
1530 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1531 [18433] libc: posix_spawn does not return correctly upon failure to
1533 [18453] localedata: charmaps/IBM875: incorrect codes
1534 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1536 [18896] localedata: he_IL: improvements for currency
1537 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1538 Romanian locale data
1539 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1541 [19056] libc: Deprecate readdir_r
1542 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1544 [19198] localedata: nl_NL: small improvements for Dutch locales
1545 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1546 nameservers (CVE-2016-5417)
1547 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1549 [19400] locale: Language missing in "iso-639.def", trivial fix in
1551 [19431] malloc: Deadlock between fflush, getdelim, and fork
1552 [19505] libc: Incorrect file descriptor validity checks in
1553 posix_spawn_file_actions_add{open,close,dup2}
1554 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1555 when using RTLD_NEXT
1556 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1557 `intl/{gettextP,loadinfo}.h'
1558 [19534] libc: execle, execlp may use malloc
1559 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1560 [19573] network: res_nclose and __res_maybe_init disagree about name
1561 server initialization, breaking Hesiod
1562 [19575] localedata: Status of GB18030 tables
1563 [19581] localedata: sr_* date_fmt string contains additional newline
1564 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1566 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1567 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1568 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1569 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1571 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1573 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1575 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1576 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1577 [19633] locale: strfmon_l applies global locale to number formatting
1578 [19642] network: Memory leak in getnameinfo
1579 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1580 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1582 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1583 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1585 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1586 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1588 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1590 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1592 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1593 pointers and lengths in error-case.
1594 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1595 report errors on UTF-16 surrogates values.
1596 [19755] nscd: nscd assertion failure in gc
1597 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1598 [19759] libc: mempcpy shouldn't be inlined
1599 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1600 [19765] libc: s390 needs an optimized mempcpy
1601 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1602 NAME_MAX limit assumption (CVE-2016-1234)
1603 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1604 [19787] network: Missing and incorrect truncation checks in getnameinfo
1605 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1607 [19791] network: Assertion failure in res_query.c with un-connectable name
1609 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1610 [19822] math: libm.so install clobbers old version
1611 [19825] network: resolv: send_vc can return uninitialized data in second
1612 response to getaddrinfo
1613 [19830] network: nss_dns: should check RDATA length against buffer length
1614 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1615 confronted with A/AAAA records of invalid size
1616 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1617 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1618 [19853] stdio: Printing IBM long double in decimal with high precision is
1620 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1621 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1622 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1624 [19865] network: Assertion failure or memory leak in
1625 _nss_dns_getcanonname_r
1626 [19868] network: nss_dns: netent code does not skip over non-PTR records
1627 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1629 [19881] string: Improve x86-64 memset
1630 [19907] string: Incorrect memcpy tests
1631 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1633 [19925] libc: termios.h XCASE namespace
1634 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1635 [19929] libc: limits.h NL_NMAX namespace
1636 [19931] stdio: Memory leak in vfprintf
1637 [19957] libc: clone(CLONE_VM) access invalid parent memory
1638 [19963] localedata: en_IL: New locale
1639 [19989] stdio: stdio.h cuserid namespace
1640 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1642 [19996] locale: langinfo.h nl_langinfo_l namespace
1643 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1644 [20010] network: getaddrinfo: Stack overflow in hostent translation
1646 [20012] stdio: libio: fmemopen append mode failure
1647 [20014] stdio: stdio.h namespace for pre-threads POSIX
1648 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1649 [20023] libc: fcntl.h timespec namespace
1650 [20024] math: [x86_64] vectorized sincos trashes the stack
1651 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1652 [20041] time: sys/time.h timespec namespace
1653 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1654 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1655 [20051] libc: ttyslot in wrong header under wrong conditions
1656 [20054] libc: gethostname not declared for XPG4
1657 [20055] libc: termios.h missing tcgetsid for XPG4
1658 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1660 [20073] libc: sys/stat.h fchmod namespace
1661 [20074] libc: stdlib.h rand_r namespace
1662 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1663 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1665 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1666 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1668 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1669 [20119] libc: Wrong mask for processors level type from CPUID
1670 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1671 AS not supporting AVX512
1672 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1674 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1675 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1677 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1678 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1679 [20178] libc: posix_spawn{p} should not call exit
1680 [20191] stdio: libio: vtables hardening
1681 [20195] string: FMA4 detection requires CPUID execution with register
1683 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1684 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1686 [20212] math: acos (sNaN) returns sNaN
1687 [20213] math: asin (sNaN) returns sNaN
1688 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1689 [20218] math: [i386] asinhl (sNaN) returns sNaN
1690 [20219] math: [i386] atanhl (sNaN) returns sNaN
1691 [20222] stdio: fopencookie: Mangle function pointers
1692 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1693 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1694 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1696 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1697 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1698 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1699 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1700 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1701 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1702 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1703 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1704 [20240] math: modf (sNaN) returns sNaN
1705 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1706 [20250] math: frexp (sNaN) returns sNaN
1707 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1708 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1710 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1711 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1712 subscript is below array bounds [-Werror=array-bounds]
1713 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1714 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1716 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1717 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1719 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1720 "invalid" exceptions
1721 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1722 [20316] localedata: id_ID: Februari instead of Pebruari
1723 [20327] string: POWER8 strcasecmp returns incorrect result
1724 [20347] math: Failure: Test: j0_downward (0xap+0)
1725 [20348] libc: FAIL: misc/tst-preadvwritev64
1726 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1727 [20350] libc: There is no test for p{read,write}64
1728 [20357] math: Incorrect cos result for 1.5174239687223976
1729 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1733 * Unicode 8.0.0 Support: Character encoding, character type info, and
1734 transliteration tables are all updated to Unicode 8.0.0, using new
1735 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1736 These updates cause user visible changes, such as the fixes for bugs
1737 89, 16061, and 18568.
1739 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1740 kernel-internal CPU set size. This means that requests that change the
1741 CPU affinity which failed before (for example, an all-ones CPU mask) will
1742 now succeed. Applications that need to determine the effective CPU
1743 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1744 setting it because the kernel can adjust it (and the previous size check
1745 would not detect this in the majority of cases).
1747 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1748 the following new symbols are used: fts64_children, fts64_close,
1749 fts64_open, fts64_read and fts64_set.
1751 * getaddrinfo now detects certain invalid responses on an internal netlink
1752 socket. If such responses are received, an affected process will
1753 terminate with an error message of "Unexpected error <number> on netlink
1754 descriptor <number>" or "Unexpected netlink response of size <number> on
1755 descriptor <number>". The most likely cause for these errors is a
1756 multi-threaded application which erroneously closes and reuses the netlink
1757 file descriptor while it is used by getaddrinfo.
1759 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1760 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1761 unnecessary serialization of memory allocation requests across threads.
1762 The defect is now corrected. Users should see a substantial increase in
1763 the concurent throughput of allocation requests for applications which
1764 trigger this bug. Affected applications typically create and
1765 destroy threads frequently. (Bug 19048 was reported and analyzed by
1768 * There is now a --disable-timezone-tools configure option for disabling the
1769 building and installing of the timezone related utilities (zic, zdump, and
1770 tzselect). This is useful for people who build the timezone data and code
1771 independent of the GNU C Library.
1773 * The obsolete header <regexp.h> has been removed. Programs that require
1774 this header must be updated to use <regex.h> instead.
1776 * The obsolete functions bdflush, create_module, get_kernel_syms,
1777 query_module and uselib are no longer available to newly linked binaries;
1778 the header <sys/kdaemon.h> has been removed. These functions and header
1779 were specific to systems using the Linux kernel and could not usefully be
1780 used with the GNU C Library on systems with version 2.6 or later of the
1783 * Optimized string, wcsmbs and memory functions for IBM z13.
1784 Implemented by Stefan Liebler.
1786 * Newly linked programs that define a variable called signgam will no longer
1787 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1788 require signgam to be set by those functions must ensure that they use the
1789 variable provided by the GNU C Library and declared in <math.h>, without
1790 defining their own copy.
1792 * The minimum GCC version that can be used to build this version of the GNU
1793 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1794 still be used to compile programs using the GNU C Library.
1796 Security related changes:
1798 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1799 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1801 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1802 the pointer guard feature. It is always enabled. Previously,
1803 LD_POINTER_GUARD could be used to disable security hardening in binaries
1804 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1807 * An integer overflow in hcreate and hcreate_r could lead to an
1808 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1810 * The catopen function no longer has unbounded stack usage. Reported by
1811 Max. (CVE-2015-8779)
1813 * The nan, nanf and nanl functions no longer have unbounded stack usage
1814 depending on the length of the string passed as an argument to the
1815 functions. Reported by Joseph Myers. (CVE-2014-9761)
1817 * A stack-based buffer overflow was found in libresolv when invoked from
1818 libnss_dns, allowing specially crafted DNS responses to seize control
1819 of execution flow in the DNS client. The buffer overflow occurs in
1820 the functions send_dg (send datagram) and send_vc (send TCP) for the
1821 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1822 family. The use of AF_UNSPEC triggers the low-level resolver code to
1823 send out two parallel queries for A and AAAA. A mismanagement of the
1824 buffers used for those queries could result in the response of a query
1825 writing beyond the alloca allocated buffer created by
1826 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1827 the overflow. Thanks to the Google Security Team and Red Hat for
1828 reporting the security impact of this issue, and Robert Holiday of
1829 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1831 The following bugs are resolved with this release:
1833 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1834 [887] math: Math library function "logb" and "nextafter" inconsistent
1835 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1836 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1837 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1838 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1840 [4404] localedata: German translation of "Alarm clock" is misleading
1841 [6799] math: nextafter() and nexttoward() doen't set errno on
1842 overflow/underflow errors
1843 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1845 [10432] nis: _nss_nis_setnetgrent assertion failure
1846 [11460] libc: fts has no LFS support
1847 [12926] network: getaddrinfo()/make_request() may spin forever
1848 [13065] nptl: Race condition in pthread barriers
1849 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1850 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1852 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1853 [14912] libc: Rename non-installed bits/*.h headers
1854 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1855 [15367] math: Let gcc use __builtin_isinf
1856 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1857 [15421] math: lgamma wrongly sets signgam for ISO C
1858 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1859 FE_INVALID with argument out of range
1860 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1862 [15786] dynamic-link: ifunc resolver functions can smash function
1864 [15918] math: Unnecessary check for equality in hypotf()
1865 [16061] localedata: Review / update transliteration data
1866 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1868 [16141] time: strptime %z offset restriction
1869 [16171] math: drem should be alias of remainder
1870 [16296] math: fegetround is pure?
1871 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1872 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1874 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1875 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1877 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1878 "Exception "Inexact" set" on ppc32
1879 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1880 [16517] math: Missing underflow exception from tanf/tan/tanl
1881 [16519] math: Missing underflow exception from sinhf
1882 [16520] math: Missing underflow exception from tanhf
1883 [16521] math: Missing underflow exception from exp2
1884 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1886 [16734] stdio: fopen calls mmap to allocate its buffer
1887 [16961] math: nan function incorrect handling of bad sequences
1888 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1889 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1890 [16985] locale: localedef: confusing error message when opening output
1892 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1893 [17197] locale: Redundant shift character in iconv conversion output at
1895 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1896 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1897 [17250] dynamic-link: static linking breaks nss loading
1898 (getaddrinfo/getpwnam/etc...)
1899 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1901 [17441] math: isnan() should use __builtin_isnan() in GCC
1902 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1903 (related to lock elision)
1904 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1905 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1906 [17887] time: strptime should be able to parse "+01:00" style timezones
1907 [17905] libc: catopen() Multiple unbounded stack allocations
1909 [18084] libc: backtrace (..., 0) dumps core on x86
1910 [18086] libc: nice() sets errno to 0 on success
1911 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1912 is too large (CVE-2015-8778)
1913 [18251] dynamic-link: SONAME missing when audit modules provides path
1914 [18265] libc: add attributes for wchar string and memory functions
1915 [18370] math: csqrt missing underflows
1916 [18421] libc: [hppa] read-only segment has dynamic relocations
1917 [18472] libc: Obsolete syscall wrappers should be compat symbols
1918 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1919 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1920 [18525] localedata: Remove locale timezone information
1921 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1922 [18568] localedata: Update locale data to Unicode 8.0
1923 [18589] locale: sort-test.sh fails at random
1924 [18595] math: ctan, ctanh missing underflows
1925 [18604] libc: assert macro-expands its argument
1926 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1927 contains a vector instruction exception.
1928 [18611] math: j1, jn missing errno setting on underflow
1929 [18618] localedata: sync Chechen locale definitions with other *_RU
1931 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1932 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1933 [18665] network: In send_dg, the recvfrom function is NOT always using the
1934 buffer size of a newly created buffer (CVE-2015-7547)
1935 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1937 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1939 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1940 [18699] math: tilegx cproj() for various complex infinities does not yield
1942 [18724] libc: Harden put*ent functions against data injection
1943 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1944 [18755] build: build errors with -DNDEBUG
1945 [18757] stdio: fmemopen fails to set errno on failure
1946 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1948 [18781] libc: openat64 lacks O_LARGEFILE
1949 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1950 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1951 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1952 [18790] math: [ldbl-128ibm] tanhl inaccurate
1953 [18795] libc: stpncpy fortification misses buffer lengths that are
1954 statically too large
1955 [18796] build: build fails for --disable-mathvec
1956 [18803] math: hypot missing underflows
1957 [18820] stdio: fmemopen may leak memory on failure
1958 [18823] math: csqrt spurious underflows
1959 [18824] math: fma spurious underflows
1960 [18825] math: pow missing underflows
1961 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1962 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1963 [18870] build: sem_open.c fails to compile with missing symbol
1965 [18872] stdio: Fix memory leak in printf_positional
1966 [18873] libc: posix_fallocate overflow check ineffective
1967 [18875] math: Excess precision leads incorrect libm
1968 [18877] libc: arm: mmap offset regression
1969 [18887] libc: memory corruption when using getmntent on blank lines
1970 [18918] localedata: hu_HU: change time to HH:MM:SS format
1971 [18921] libc: Regression: extraneous stat() and fstat() performed by
1973 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1974 binaries (CVE-2015-8777)
1975 [18951] math: tgamma missing underflows
1976 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1978 [18953] localedata: lt_LT: change currency symbol to the euro
1979 [18956] math: powf inaccuracy
1980 [18961] math: [i386] exp missing underflows
1981 [18966] math: [i386] exp10 missing underflows
1982 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1983 [18969] build: multiple string test failures due to missing locale
1985 [18970] libc: Reference of pthread_setcancelstate in libc.a
1986 [18977] math: float / long double Bessel functions not in XSI POSIX
1987 [18980] math: i386 libm functions return with excess range and precision
1988 [18981] math: i386 scalb*, ldexp return with excess range and precision
1989 [18982] stdio: va_list and vprintf
1990 [18985] time: Passing out of range data to strftime() causes a segfault
1992 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1993 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1994 [19012] locale: iconv_open leaks memory on error path
1995 [19016] math: clog, clog10 inaccuracy
1996 [19018] nptl: Mangle function pointers in tls_dtor_list
1997 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1998 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1999 [19048] malloc: malloc: arena free list can become cyclic, increasing
2001 [19049] math: [powerpc] erfc incorrect zero sign
2002 [19050] math: [powerpc] log* incorrect zero sign
2003 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2004 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2005 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2006 [19074] network: Data race in _res_hconf_reorder_addrs
2007 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2008 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2009 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2011 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2013 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2014 [19086] manual: posix_fallocate64 documented argument order is wrong.
2015 [19088] math: lround, llround missing exceptions close to overflow
2017 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2018 [19095] math: dbl-64 lrint incorrect for 64-bit long
2019 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2020 [19124] dynamic-link: ld.so failed to build with older assmebler
2021 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2022 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2023 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2024 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2025 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2026 pthread_setaffinity_np
2027 [19156] math: [ldbl-128] j0l spurious underflows
2028 [19164] nptl: tst-getcpu fails with many possible CPUs
2029 [19168] math: math/test-ildoubl and math/test-ldouble failure
2030 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2031 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2033 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2034 (FE_NOMASK_ENV) do not clear SSE exceptions
2035 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2036 _int_new_arena/reused_arena
2037 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2038 [19201] math: dbl-64 remainder incorrect sign of zero result
2039 [19205] math: bits/math-finite.h conditions do not match math.h and
2041 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2042 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2043 for C99-based standards
2044 [19212] libc: features.h not -Wundef clean
2045 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2047 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2048 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2049 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2051 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2052 "inexact" exceptions
2053 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2055 [19242] libc: strtol incorrect in Turkish locales
2056 [19243] malloc: reused_arena can pick an arena on the free list, leading
2057 to an assertion failure and reference count corruption
2058 [19253] time: tzset() ineffective when temporary TZ did not include DST
2060 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2061 [19270] math: [hppa] Shared libm missing __isnanl
2062 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2063 MAP_HUGETLB and MAP_STACK defines
2064 [19313] nptl: Wrong __cpu_mask for x32
2065 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2066 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2067 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2068 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2069 [19363] time: x32: times() return value wrongly truncates/sign extends
2071 [19367] dynamic-link: Improve branch prediction on Silvermont
2072 [19369] network: Default domain name not reset by res_ninit when "search"
2073 / "domain" entry is removed from resolv.conf
2074 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2076 [19385] localedata: bg_BG: time separator should be colon, not comma
2077 [19408] libc: linux personality syscall wrapper may erroneously return an
2078 error on 32-bit architectures
2079 [19415] libc: dladdr returns wrong names on hppa
2080 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2081 IBM905, IBM907, and IBM909
2082 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2083 [19443] build: build failures with -DDEBUG
2084 [19451] build: Make check fails on test-double-vlen2
2085 [19462] libc: Glibc failed to build with -Os
2086 [19465] math: Wrong code with -Os
2087 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2089 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2091 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2092 [19486] math: S390: Math tests fail with "Exception Inexact set".
2093 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2094 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2096 [19590] math: Fail to build shared objects that use libmvec.so functions.
2100 * The following bugs are resolved with this release:
2102 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2103 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2104 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2105 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2106 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2107 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2108 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2109 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2110 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2111 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2112 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2113 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2114 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2115 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2116 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2117 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2118 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2119 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2120 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2121 18657, 18676, 18694, 18696, 18887.
2123 * Cache information can be queried via sysconf() function on s390 e.g. with
2124 _SC_LEVEL1_ICACHE_SIZE as argument.
2126 * A buffer overflow in gethostbyname_r and related functions performing DNS
2127 requests has been fixed. If the NSS functions were called with a
2128 misaligned buffer, the buffer length change due to pointer alignment was
2129 not taken into account. This could result in application crashes or,
2130 potentially arbitrary code execution, using crafted, but syntactically
2131 valid DNS responses. (CVE-2015-1781)
2133 * The time zone file parser has been made more robust against crafted time
2134 zone files, avoiding heap buffer overflows related to the processing of
2135 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2136 large time zone data files. Overly long time zone specifiers in the TZ
2137 variable no longer result in stack overflows and crashes.
2139 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2140 for LD and GD on x86 and x86-64, has been implemented. You will need
2141 binutils-2.24 or later to enable this optimization.
2143 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2144 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2145 Hat). These updates cause user visible changes, such as the fix for bug
2148 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2149 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2150 condition in some applications.
2152 * Added vector math library named libmvec with the following vectorized x86_64
2153 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2155 The library can be disabled with --disable-mathvec. Use of the functions is
2156 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2157 Shared library libmvec.so is linked in as needed when using -lm (no need to
2158 specify -lmvec explicitly for not static builds).
2159 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2161 * A new fmemopen implementation has been added with the goal of POSIX
2162 compliance. The new implementation fixes the following long-standing
2163 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2164 old implementation is still present for use be by existing binaries.
2166 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2167 releases. It has been fixed to match 2.19 and older, but binaries built
2168 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2170 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2171 Contributed by Roland McGrath (Google).
2173 * The header <regexp.h> is deprecated, and will be removed in a future
2174 release. Use of this header will trigger a deprecation warning.
2175 Application developers should update their code to use <regex.h> instead.
2177 This header was formerly part of SUS, but was deprecated in 1994 and
2178 removed from the standard in 2001. Also, the glibc implementation
2179 leaks memory. See BZ#18681 for more details.
2183 * The following bugs are resolved with this release:
2185 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2186 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2187 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2188 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2189 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2190 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2191 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2192 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2193 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2194 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2197 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2198 too little memory for the to-be-scanned arguments and overflow the
2199 allocated buffer. The implementation now correctly computes the required
2200 buffer size when using malloc, and switches to malloc from alloca as
2203 * A new semaphore algorithm has been implemented in generic C code for all
2204 machines. Previous custom assembly implementations of semaphore were
2205 difficult to reason about or ensure that they were safe. The new version
2206 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2207 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2208 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2210 * Port to Altera Nios II has been contributed by Mentor Graphics.
2212 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2213 implementations for powerpc64/powerpc64le.
2214 Implemented by Adhemerval Zanella (IBM).
2216 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2217 and powerpc64le. This may improve lock scaling of existing programs on
2218 HTM capable systems. The lock elision code is only enabled with
2219 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2220 powerpc will issue a transaction abort on every syscall to avoid side
2221 effects being visible outside transactions.
2223 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2224 AArch64. Contributed by ARM Ltd.
2226 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2228 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2229 under certain input conditions resulting in the execution of a shell for
2230 command substitution when the application did not request it. The
2231 implementation now checks WRDE_NOCMD immediately before executing the
2232 shell and returns the error WRDE_CMDSUB as expected.
2234 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2235 processing format strings with a large number of format specifiers.
2237 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2238 infinite loop if the DNS response contained a PTR record of an unexpected
2241 * The minimum GCC version that can be used to build this version of the GNU
2242 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2243 still be used to compile programs using the GNU C Library.
2245 * The GNU C Library is now built with -Werror by default. This can be
2246 disabled by configuring with --disable-werror.
2248 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2250 * The obsolete sigvec function has been removed. This was the original
2251 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2252 programs have been using instead for about 25 years. Of course, ABI
2253 compatibility for old binaries using sigvec remains intact.
2255 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2256 with newer versions of bison.
2258 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2259 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2260 registers overlay two consecutive single-precision registers. MIPS32R2
2261 introduced a new FPU mode (FR=1) where double-precision registers extend the
2262 corresponding single-precision registers which is incompatible with the
2263 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2264 require the use of FR=1 making a transition necessary. New o32 ABI
2265 extensions enable users to migrate over time from the original o32 ABI
2266 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2267 tracks the ABI of any loaded object and verifies that new objects are
2268 compatible. Mode transitions will also be requested as required and
2269 unsupportable objects will be rejected. The ABI checks include both soft and
2270 hard float ABIs for o32, n32 and n64.
2272 GCC 5 with GNU binutils 2.25 onwards:
2273 It is strongly recommended that all o32 system libraries are built using the
2274 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2275 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2276 compiler using --with-fp-32=xx to set this by default.
2280 * The following bugs are resolved with this release:
2282 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2283 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2284 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2285 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2286 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2287 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2288 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2289 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2290 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2291 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2292 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2293 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2294 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2295 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2296 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2298 * Reverted change of ABI data structures for s390 and s390x:
2299 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2300 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2301 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2302 are preserved pointing straight to the same implementation as the old ones.
2303 Given that, new callers will simply provide a too-big buffer to these
2304 functions. Any applications/libraries out there that embed jmp_buf or
2305 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2306 against 2.19 headers will have to rebuilt again. This is necessary in any
2307 case to revert the breakage in their ABI caused by the glibc change.
2309 * Support for file description locks is added to systems running the
2310 Linux kernel. The standard file locking interfaces are extended to
2311 operate on file descriptions, not file descriptors, via the use of
2312 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2313 are associated with an open file instead of a process.
2315 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2317 * The minimum Linux kernel version that this version of the GNU C Library
2318 can be used with is 2.6.32.
2320 * Running the testsuite no longer terminates as soon as a test fails.
2321 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2322 with PASS or FAIL lines for individual tests. A summary of the results is
2323 printed, including a list of failing lists, and "make check" exits with
2324 error status if there were any unexpected failures. "make check
2325 stop-on-test-failure=y" may be used to keep the old behavior.
2327 * The am33 port, which had not worked for several years, has been removed
2330 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2331 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2332 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2333 conflicted with POSIX (support for which was removed in 2.19), the
2334 interfaces those macros enabled remain available when compiling with
2335 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2336 test macros defined.
2338 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2340 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2341 This may improve lock scaling of existing programs on TX capable systems.
2342 The lock elision code is only built with --enable-lock-elision=yes and
2343 then requires a GCC version supporting the TX builtins. With lock elision
2344 default mutexes are elided via __builtin_tbegin, if the cpu supports
2345 transactions. By default lock elision is not enabled and the elision code
2348 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2349 copy the path argument. This allowed programs to cause posix_spawn to
2350 deference a dangling pointer, or use an unexpected pathname argument if
2351 the string was modified after the posix_spawn_file_actions_addopen
2354 * All supported architectures now use the main glibc sysdeps directory
2355 instead of some being in a separate "ports" directory (which was
2356 distributed separately before glibc 2.17).
2358 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2359 On configurations that support it (all Linux configurations), it's now
2360 used regardless of the --enable-add-ons switch to configure. It is no
2361 longer possible to build such configurations without pthreads support.
2363 * Locale names, including those obtained from environment variables (LANG
2364 and the LC_* variables), are more tightly checked for proper syntax.
2365 setlocale will now fail (with EINVAL) for locale names that are overly
2366 long, contain slashes without starting with a slash, or contain ".." path
2367 components. (CVE-2014-0475) Previously, some valid locale names were
2368 silently replaced with the "C" locale when running in AT_SECURE mode
2369 (e.g., in a SUID program). This is no longer necessary because of the
2372 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2373 with application code using Intel MPX instructions. (With all previous
2374 versions, the MPX register state could be clobbered when making calls
2375 into or out of a shared library.) Note that while the new dynamic
2376 linker is compatible with all known x86 hardware whether or not it
2377 supports Intel MPX, some x86 instruction-set emulators might fail to
2378 handle the new instruction encodings. This is known to affect Valgrind
2379 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2380 release), and might affect other tools that do instruction emulation.
2382 * Support for loadable gconv transliteration modules has been removed.
2383 The support for transliteration modules has been non-functional for
2384 over a decade, and the removal is prompted by security defects. The
2385 normal gconv conversion modules are still supported. Transliteration
2386 with //TRANSLIT is still possible, and the //IGNORE specifier
2387 continues to be supported. (CVE-2014-5119)
2389 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2390 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2391 resulting a denial-of-service security vulnerability in applications which
2392 use functions related to iconv. (CVE-2014-6040)
2396 * The following bugs are resolved with this release:
2398 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2399 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2400 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2401 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2402 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2403 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2404 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2405 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2406 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2407 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2408 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2409 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2410 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2411 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2412 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2413 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2414 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2415 16431, 16453, 16474, 16506, 16510, 16529
2417 * Slovenian translations for glibc messages have been contributed by the
2418 Translation Project's Slovenian team of translators.
2420 * The public headers no longer use __unused nor __block. This change is to
2421 support compiling programs that are derived from BSD sources and use
2422 __unused internally, and to support compiling with Clang's -fblock
2423 extension which uses __block.
2425 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2426 large collation sequences to optimize multiple passes. This cache
2427 computation may overflow for large collation sequences and may cause a
2428 stack or buffer overflow. This is now fixed to use a slower algorithm
2429 which does not use a cache if there is an integer overflow.
2431 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2432 rules for large collation sequences to optimize multiple passes and falls
2433 back to alloca if malloc fails, resulting in a possible stack overflow.
2434 The implementation now falls back to an uncached collation sequence lookup
2437 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2438 initialized for static applications resulting in the security feature
2439 being disabled. The pointer guard is now correctly initialized to a
2440 random value for static applications. Existing static applications need
2441 to be recompiled to take advantage of the fix (bug 15754).
2443 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2444 to the d_name member of struct dirent, or omit the terminating NUL
2445 character. (Bugzilla #14699).
2447 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2448 aligned_alloc functions could allocate too few bytes or corrupt the
2449 heap when passed very large allocation size values (Bugzilla #15855,
2452 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2453 for AF_INET6 has been fixed (Bugzilla #16072).
2455 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2456 pap_CW, quz_PE, the_NP.
2458 * Substantially revised locales: gd_GB, ht_HT
2460 * The LC_ADDRESS field was updated to support country_car for almost all
2463 * ISO 1427 definitions were updated.
2465 * ISO 3166 definitions were updated.
2467 * The localedef utility now supports --big-endian and --little-endian
2468 command-line options to generate locales for a different system from that
2469 for which the C library was built.
2471 * Binary locale files now only depend on the endianness of the system for
2472 which they are generated and not on other properties of that system. As a
2473 consequence, binary files generated with new localedef may be incompatible
2474 with old versions of the GNU C Library, and binary files generated with
2475 old localedef may be incompatible with this version of the GNU C Library,
2476 in the following circumstances:
2478 + Locale files may be incompatible on m68k systems.
2480 + Locale archive files (but not separate files for individual locales) may
2481 be incompatible on systems where plain "char" is signed.
2483 * The configure option --disable-versioning has been removed. Builds with
2484 --disable-versioning had not worked for several years.
2486 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2487 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2489 * SystemTap probes for malloc have been introduced.
2491 * SystemTap probes for slow multiple precision fallback paths of
2492 transcendental functions have been introduced.
2494 * Support for powerpc64le has been added.
2496 * The soft-float powerpc port now supports e500 processors.
2498 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2500 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2501 set of header declarations that are enabled by default, even when other
2502 feature test macros or compiler options such as -std=c99 would otherwise
2503 disable some of those declarations.
2505 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2506 conflict with POSIX. The libbsd-compat library (which was a dummy library
2507 that did nothing) has also been removed.
2509 * Preliminary documentation about Multi-Thread, Async-Signal and
2510 Async-Cancel Safety has been added.
2512 * Change of ABI data structures for s390 and s390x:
2513 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2514 allow for future hardware extensions. All library functions that accept or
2515 return these structures were versioned in order to provide backwards
2516 compatibility with existing code. However, not all cases can be handled
2517 with interface versioning. If an application embeds either structure into
2518 another structure and that structure is passed to another compilation unit
2519 which expects the newer larger-sized object then any access to the new
2520 fields will result in undefined behaviour. Similarly any access to fields
2521 that were moved by the enlarging of the structures will cause undefined
2522 behaviour. To fix the undefined behaviour all objects that are part of the
2523 final in-memory image must agree on the size of structures, and this may
2524 require recompilation.
2528 * The following bugs are resolved with this release:
2530 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2531 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2532 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2533 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2534 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2535 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2536 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2537 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2538 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2539 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2540 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2541 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2542 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2545 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2546 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2547 Distributions can re-enable building and using pt_chown via the new configure
2548 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2549 considerable security risks and should only be used if the distribution
2550 understands and accepts the risks.
2552 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2555 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2556 fixed (Bugzilla #15330).
2558 * Add support for calling C++11 thread_local object destructors on thread
2559 and program exit. This needs compiler support for offloading C++11
2560 destructor calls to glibc.
2562 * Improved worst case performance of libm functions with double inputs and
2565 * Support for priority inherited mutexes in pthread condition variables on
2566 non-x86 architectures.
2568 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2570 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2572 * Optimized string functions for ARM. Implemented by Will Newton and
2575 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2577 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2580 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2581 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2583 * Added a benchmark framework to track performance of functions in glibc.
2585 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2586 It is based on draft TS 18661 and currently enabled as a GNU extension.
2588 * On Linux, the clock function now uses the clock_gettime system call
2589 for improved precision, rather than old times system call.
2591 * Added support for version-3 tz format files. This is needed when using
2592 the tz database release 2013e or later, and affects a few unusual cases --
2593 currently only TZ='America/Godthab' for time stamps after 2037.
2595 * Added new API functions pthread_getattr_default_np and
2596 pthread_setattr_default_np to get and set the default pthread
2597 attributes of a process.
2599 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2600 This may improve lock scaling of existing programs on TSX capable systems.
2601 When the --enable-lock-elision=yes parameter is specified at configure
2602 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2605 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2606 availability). Implemented by Ryan S. Arnold.
2608 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2612 * The following bugs are resolved with this release:
2614 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2615 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2616 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2617 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2618 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2619 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2620 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2621 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2622 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2623 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2624 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2625 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2626 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2627 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2628 14879, 14889, 14893, 14898, 14914.
2630 * Optimization of memcpy for MIPS.
2632 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2633 EMFILE has been fixed (Bugzilla #14889).
2635 * The add-on ports collection is now distributed in the "ports" subdirectory
2636 of the main GNU C Library distribution, rather than separately.
2638 * Port to ARM AArch64 contributed by Linaro.
2640 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2641 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2643 Implemented by Andreas Krebbel.
2645 * The new function secure_getenv allows secure access to the environment,
2646 returning NULL if running in a SUID/SGID process. This function replaces
2647 the internal function __secure_getenv.
2649 * SystemTap static probes have been added into the dynamic linker.
2650 Implemented by Gary Benson.
2652 * Optimizations of string functions strstr, strcasestr and memmem.
2653 Implemented by Maxim Kuvyrkov.
2655 * The minimum Linux kernel version that this version of the GNU C Library
2656 can be used with is 2.6.16.
2658 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2659 powerpc POWER7. Implemented by Will Schmidt.
2661 * New configure option --disable-nscd builds the C library such that it
2662 never attempts to contact the Name Service Caching Daemon (nscd).
2663 New configure option --disable-build-nscd avoids building nscd itself;
2664 this is the default if --disable-nscd is used.
2666 * Improved support for cross-compilation, including cross-testing and
2667 bootstrap builds without a previously built glibc.
2669 * Several testsuite tests are now able to test multiple IFUNC variants of an
2670 interface, rather than just testing the one that would be chooen by
2673 * New configure options --with-bugurl and --with-pkgversion, for
2674 distributors to use to embed their bug-reporting and package version
2675 information in --help and --version output.
2677 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2678 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2679 allows creation of chroots without the procfs mounted on /proc.
2681 * The `crypt' function now fails if passed salt bytes that violate the
2682 specification for those values. On Linux, the `crypt' function will
2683 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2684 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2685 when the mode is enabled.
2687 * The `clock_*' suite of functions (declared in <time.h>) is now available
2688 directly in the main C library. Previously it was necessary to link with
2689 -lrt to use these functions. This change has the effect that a
2690 single-threaded program that uses a function such as `clock_gettime' (and
2691 is not linked with -lrt) will no longer implicitly load the pthreads
2692 library at runtime and so will not suffer the overheads associated with
2693 multi-thread support in other code such as the C++ runtime library.
2695 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2700 * The following bugs are resolved with this release:
2702 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2703 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2704 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2705 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2706 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2707 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2708 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2709 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2710 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2711 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2712 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2713 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2714 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2715 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2716 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2717 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2718 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2719 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2720 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2721 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2722 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2723 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2726 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2727 configuring glibc with:
2728 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2729 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2730 Implemented by H.J. Lu.
2734 + define static_assert
2736 + do not declare gets
2738 + declare at_quick_exit and quick_exit also for ISO C11
2740 + aligned_alloc. NB: The code is deliberately allows the size parameter
2741 to not be a multiple of the alignment. This is a moronic requirement
2742 in the standard but it is only a requirement on the caller, not the
2745 + timespec_get added
2747 + uchar.h support added
2749 + CMPLX, CMPLXF, CMPLXL added
2751 Implemented by Ulrich Drepper.
2753 * Support for the IA-64 has been moved to ports.
2755 * Remove support for anything but ELF binary format
2757 * Checking versions of poll, ppoll added.
2758 Implemented by Ulrich Drepper.
2760 * More generic and 64-bit performance optimizations to math functions.
2761 Implemented by Ulrich Drepper.
2763 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2764 headers and functions available at compile time as they were before
2765 version 2.14. This option will be removed at some time in the future
2766 after the TI-RPC library becomes fully sufficient for the needs of
2767 existing applications.
2769 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2770 Note that glibc is not expected to work with any Linux kernel version
2773 * New header <sys/auxv.h> and function getauxval allowing easy access to
2774 the AT_* key-value pairs passed from the Linux kernel. The header also
2775 defines the HWCAP_* bits associated with the AT_HWCAP key.
2777 * New locales: mag_IN
2779 * New configure option --enable-systemtap builds SystemTap static probes
2780 into libc for setjmp and longjmp and into libpthread for various operations.
2781 So far the setjmp/longjmp probes and some of the libpthread probes are
2782 provided only for i*86 and x86_64.
2783 Implemented by Roland McGrath and Rayson Ho.
2785 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2787 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2790 * More optimized functions for SPARC. Implemented by David S. Miller.
2792 * Improved support for cross-compilation, especially bootstrap builds
2793 without a previously built glibc.
2795 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2796 Chris Metcalf from Tilera.
2798 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2799 now supported for ARM processors.
2801 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2802 as the name of the dynamic linker, to distinguish it from the
2803 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2805 * Support for CRIS has been removed from ports.
2807 * A new class of installed header has been documented for low-level
2808 platform-specific functionality. PowerPC added the first instance with a
2809 function to provide time base register access. Contributed by Tulio
2810 Magno Quites Machado Filho.
2812 * ix86 configurations now install header files that are consistent with
2813 what x86-64 configurations install. These same header files can be used
2814 for -m32, -m64, or -mx32 builds.
2815 Contributed by H.J. Lu.
2817 * Math library bug fixes. A thorough audit of all open math library bugs was
2818 conducted by Joseph Myers. Significant progress was made on many math
2819 library bugs resulting in more accurate exceptions and function results.
2820 Many thanks to all those that contributed including Andreas Jaeger for his
2821 patch review and work on the x87 trigonometric instruction issues.
2823 * Timezone data is no longer installed. Timezone-related binaries and scripts
2824 will continue to be installed. Users should obtain their timezone data from
2825 their distribution provider or from the tzdata package at
2826 <ftp://munnari.oz.au/pub/>.
2830 * The following bugs are resolved with this release:
2832 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2833 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2834 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2835 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2836 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2837 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2838 13484, 13506, 13515, 13523, 13524, 13538, 13540
2840 * New program pldd to list loaded object of a process
2841 Implemented by Ulrich Drepper.
2843 * Add nss_db support back to glibc. No more dependency on Berkeley db
2844 and support for initgroups lookups.
2845 Implemented by Ulrich Drepper.
2847 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2848 Contributed by HJ Lu.
2850 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2851 Contributed by HJ Lu.
2853 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2854 on x86-32 and x86-64.
2855 Contributed by Liubov Dmitrieva.
2857 * Optimized strchr and strrchr for SSE on x86-32.
2858 Contributed by Liubov Dmitrieva.
2860 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2861 for x86-64 and x86-32.
2862 Contributed by Liubov Dmitrieva.
2864 * New interfaces: scandirat, scandirat64
2865 Implemented by Ulrich Drepper.
2867 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2868 Implemented by Ulrich Drepper.
2870 * nscd now also caches the netgroup database.
2871 Implemented by Ulrich Drepper.
2873 * Integrate libm with gcc's -ffinite-math-only option.
2874 Implemented by Ulrich Drepper.
2876 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2877 to math functions. Implemented by Ulrich Drepper.
2879 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2880 Implemented by Ulrich Drepper.
2882 * New Linux interfaces: process_vm_readv, process_vm_writev
2884 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2885 Implemented by Ulrich Drepper.
2887 * Optimized nearbyint and strcasecmp for PPC.
2888 Implemented by Adhemerval Zanella.
2890 * New locales: bho_IN, unm_US, es_CU, ta_LK
2894 * The following bugs are resolved with this release:
2896 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2897 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2898 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2899 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2900 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2901 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2902 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2903 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2904 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2905 12795, 12811, 12813, 12814, 12841
2907 * The RPC implementation in libc is obsoleted. Old programs keep working
2908 but new programs cannot be linked with the routines in libc anymore.
2909 Programs in need of RPC functionality must be linked against TI-RPC.
2910 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2912 Visible changes of this change include (obviously) the inability to link
2913 programs using RPC functions without referencing the TI-RPC library and the
2914 removal of the RPC headers from the glibc headers.
2915 Implemented by Ulrich Drepper.
2917 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2918 syncfs, setns, sendmmsg
2920 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2921 yue_HK, lij_IT, mhr_RU
2923 * New encodings: CP770, CP771, CP772, CP773, CP774
2925 * New program sotruss to trace calls through PLTs
2926 Implemented by Ulrich Drepper.
2928 * The malloc hook implementation is marked deprecated and will be removed
2929 from the default implementation in the next version. The design never
2930 worked ever since the introduction of threads. Even programs which do
2931 not create threads themselves can use multiple threads created internally.
2935 * The following bugs are resolved with this release:
2937 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2938 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2939 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2940 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2943 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2945 * POWER7 optimizations: memset, memcmp, strncmp
2947 * New optimized string functions for x86-64: strnlen (SSE2),
2948 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2949 Implemented by Ulrich Drepper.
2953 * The following bugs are resolved with this release:
2955 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2956 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2957 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2958 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2959 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2960 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2961 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2962 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2964 * New interfaces: pthread_getname_np, pthread_setname_np
2966 * New Linux interface: recvmmsg
2968 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2970 * The dynamic linker now recognizes supported ABI versions from the
2971 EI_ABIVERSION field in the ELF header.
2972 Implemented by Ulrich Drepper.
2974 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2975 will not be used to fill in password fields in the passwd.byname replies.
2976 Instead it is used to synthesize the shadow.byname table, should it be
2977 missing. This is a useful mode in some installations involving Solaris.
2978 Implemented by Ulrich Drepper.
2980 * New locales: kok_IN, sq_MK, cv_RU
2984 * The following bugs are resolved with this release:
2986 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2987 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2988 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2989 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2990 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2991 10789, 10817, 10823, 10840, 10847
2993 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2995 Implemented by Ulrich Drepper.
2997 * Checking version of longjmp added that fails if an uninitialized stack
2998 frame would be created. Implemented by Ulrich Drepper.
3000 * STT_GNU_IFUNC is now supported in static executables.
3001 Implemented by H.J. Lu.
3003 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3005 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3006 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3007 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3008 Contributed by H.J. Lu.
3010 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3011 Implemented by Ulrich Drepper.
3013 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3015 Contributed by H.J. Lu.
3017 * Support for fma instruction in AVX on x86-64.
3018 Implemented by H.J. Lu and Ulrich Drepper.
3020 * AVX support in x86-64 auditing support in ld.so.
3021 Implemented by H.J. Lu.
3023 * STB_GNU_UNIQUE support added.
3024 Implemented by Ulrich Drepper.
3026 * Implement second fallback mode for optimized DNS lookup for even more
3027 broken environments. If two requests from the same port are not
3028 handled correctly close the socket and open a new one before sending
3029 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3030 can be used to select this mode right away, instead of rediscovering the
3031 necessity is every process again.
3032 Implemented by Ulrich Drepper.
3034 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3035 Implemented by Adam Tkac.
3037 * Optimized iconv conversions for S390x.
3038 Implemented by Andreas Krebbel.
3040 * Using condvars with PI mutexes is now more efficient due to kernel
3041 support for requeueing to PI futexes. NPTL support added for x86-64.
3042 Implemented by Ulrich Drepper.
3044 * New locales: ps_AF, my_MM
3048 * The following bugs are resolved with this release:
3050 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3051 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3052 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3053 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3054 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3056 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3057 Implemented by Ulrich Drepper.
3059 * New Linux interfaces: accept4, fallocate, fallocate64.
3060 Implemented by Ulrich Drepper.
3062 * Correct declarations of string function when used in C++ code. This
3063 could lead to compile errors for invalid C++ code.
3065 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3068 * New POSIX 2008 interface: psiginfo
3069 Implemented by Ulrich Drepper.
3071 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3072 Implemented by Ulrich Drepper.
3074 * Support for selecting between multiple function definitions at runtime
3075 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3077 * The libcrypt library can now use the hash function implementations in
3078 NSS. Implemented by Ulrich Drepper.
3080 * The malloc implementation can be compiled to be less memory efficient
3081 but higher performing in multi-threaded programs.
3082 Implemented by Ulrich Drepper.
3084 * New locales: nan_TW@latin, ks_IN
3086 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3087 Implemented by Ulrich Drepper.
3089 * Extended printf hook support. It is possible to use user-defined types
3090 and extend existing format specifiers.
3091 Implemented by Ulrich Drepper.
3093 * Handling for group shadow files has been added.
3094 Implemented by Ulrich Drepper.
3096 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3097 servers (the case, e.g., for some people using the built-in DNS
3098 server in ADSL modems/routers). There is a once-per-process timeout
3099 in case of a broken server. To avoid it, users can run nscd or put
3100 'options single-request' in /etc/resolv.conf.
3101 Implemented by Ulrich Drepper.
3105 * The following bugs are resolved with this release:
3107 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3108 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3109 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3110 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3111 6968, 6974, 6980, 6995, 7008, 7009, 7029
3113 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3114 up at the same time. Implemented by Ulrich Drepper.
3116 * TLS descriptors for LD and GD on x86 and x86-64.
3117 Implemented by Alexandre Oliva.
3119 * getaddrinfo now handles DCCP and UDPlite.
3120 Implemented by Ulrich Drepper.
3122 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3123 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3124 Implemented by Ulrich Drepper.
3126 * New implementation of memmem, strstr, and strcasestr which is O(n).
3127 Implemented by Eric Blake.
3129 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3131 * Implement "e" option for popen to open file descriptor with the
3132 close-on-exec flag set. Implemented by Ulrich Drepper.
3134 * Implement "b" mode for fmemopen. In this mode writes writes don't
3135 implicitly add a NUL byte and seeks from the end of the buffer really
3136 use the buffer end, not the string length as the basis.
3137 Implemented by Ulrich Drepper.
3139 * Many functions, exported and internal, now atomically set the close-on-exec
3140 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3142 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3144 Implemented by Pravin Satpute.
3146 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3150 * The following bugs are resolved with this release:
3152 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3153 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3154 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3155 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3156 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3157 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3158 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3160 * New locales: bo_CN, bo_IN, shs_CA.
3162 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3164 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3165 Implemented by Pravin Satpute.
3167 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3169 * nscd caches now all timeouts for DNS entries
3170 Implemented by Ulrich Drepper.
3172 * nscd is more efficient and wakes up less often.
3173 Implemented by Ulrich Drepper.
3175 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3176 vdprintf, and obstack_vprintf.
3177 Implemented by Jakub Jelinek.
3179 * Faster memset for x86-64.
3180 Implemented by Harsha Jagasia and H.J. Lu.
3182 * Faster memcpy on x86.
3183 Implemented by Ulrich Drepper.
3185 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3186 Implemented by Ulrich Drepper.
3188 * Faster sqrt and sqrtf implemention for some PPC variants.
3189 Implemented by Steven Munroe.
3193 * The following bugs are resolved with this release:
3195 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3196 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3197 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3198 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3199 5071, 5103, 5104, 5112, 5113, 5184, 5186
3201 * More checking functions: fread, fread_unlocked, open*, mq_open.
3202 Implemented by Jakub Jelinek and Ulrich Drepper.
3204 * Extend fortification to C++. Implemented by Jakub Jelinek.
3206 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3207 by not recognizing 'a' as a modifier when those specs are requested.
3208 Implemented by Jakub Jelinek.
3210 * PPC optimizations to math and string functions.
3211 Implemented by Steven Munroe.
3213 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3214 options to be passed. Implemented by Ulrich Drepper.
3216 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3218 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3219 Implemented by Ulrich Drepper.
3221 * Handle private futexes in the NPTL implementation.
3222 Implemented by Jakub Jelinek and Ulrich Drepper.
3224 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3225 Implemented by Roland McGrath and Ulrich Drepper.
3227 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3229 * SHA-256 and SHA-512 based password encryption.
3230 Implemented by Ulrich Drepper.
3232 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3233 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3236 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3237 Implemented by Ulrich Drepper.
3241 * The following bugs are resolved with this release:
3243 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3244 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3245 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3246 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3247 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3248 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3249 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3252 * New Linux interfaces: epoll_pwait, sched_getcpu.
3254 * New generic interfaces: strerror_l.
3256 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3260 * The following bugs are resolved with this release:
3262 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3263 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3264 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3265 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3266 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3267 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3268 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3269 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3270 3225, 3251, 3252, 3253, 3273
3272 * For Linux, the sorting of addresses returned by getaddrinfo now also
3273 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3274 Implemented by Ulrich Drepper.
3276 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3277 Implemented by Ulrich Drepper.
3279 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3281 * New iconv module for MIK. Contributed by Alexander Shopov.
3283 * For sites with broken group and/or passwd database, the auto-propagate
3284 option of nscd can prevent creating ID lookup entries from the results
3285 of a name lookup and vice versa. This usually is no problem but some
3286 site might have problems with the default behavior.
3287 Implemented by Ulrich Drepper.
3289 * Iterating over entire database in NIS can be slow. With the
3290 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3291 to trade time for memory. The entire database will be read at once.
3292 Implemented by Ulrich Drepper.
3294 * The interfaces introduced in RFC 3542 have been implemented by
3297 * Support for the new ELF hash table format was added by Ulrich Drepper.
3299 * Support for priority inheritance mutexes added by Jakub Jelinek and
3302 * Support for priority protected mutexes added by Jakub Jelinek.
3304 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3305 or_IN, csb_PL, fy_NL, sr_ME.
3309 * More overflow detection functions.
3311 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3312 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3314 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3315 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3316 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3317 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3318 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3319 by Masahide Washizawa.
3321 * It is now possible to install audit libraries for the dynamic linker using
3322 LD_AUDIT. Implemented by Ulrich Drepper.
3324 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3325 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3326 For a libc and libpthread that works well on Linux 2.4 kernels, we
3327 recommend using the stable 2.3 branch.
3329 * The new function open_wmemstream acts like open_memstream,
3330 but uses a wchar_t wide character buffer.
3332 * The new function ppoll is an improved version of poll, similar to pselect.
3334 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3335 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3336 renameat, unlinkat, linkat, symlinkat, readlinkat.
3338 * New Linux kernel system calls: unshare,
3339 inotify_init, inotify_add_watch, inotify_rm_watch.
3341 * The euidaccess function is now also known by the alias eaccess,
3342 for compatibility with some other systems.
3344 * Timezone data updated to 2006b version.
3348 * The following bugs are resolved with this release:
3350 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3351 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3352 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3353 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3354 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3355 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3357 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3359 * As of this release, GCC 4 can be used to compile the C Library.
3361 * Timezone data updated to 2005m version.
3365 * The following bugs are resolved with this release:
3367 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3368 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3369 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3370 777, 787, 821, 822, 823, 825
3372 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3376 * Support for RFC 3678. Real implementations exist only for Linux so far.
3377 Implemented by Ulrich Drepper.
3379 * nscd can now cache entries persistently. Expiring entries are reloaded.
3380 For speedups the cache can be shared in memory with client processes.
3381 Implemented by Ulrich Drepper.
3383 * nscd can now perform SELinux checks.
3384 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3386 * getaddrinfo queries are now cached. Canonical name lookup is performed
3388 Implemented by Ulrich Drepper.
3390 * The nothrow function attribute is used when headers are used by gcc when
3391 compiling C code. This can avoid significant amounts of exception
3394 * The malloc functions perform more error checking and are stricter when
3395 it comes to reacting on errors. The default action is to terminate
3396 the process after showing an error message. Implemented by Ulrich Drepper.
3398 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3399 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3401 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3402 namespaces using the new function dlmopen(). This feature is of course,
3403 like most other dynamic loading functionality, not available in statically
3404 linked applications. Implemented by Ulrich Drepper.
3406 * Low-overhead boundary checking variants of string and some stdio functions
3407 were added. These are to be used in conjunction with a gcc patch by
3408 Jakub Jelinek which adds calls to these functions if possible.
3409 Implemented by Jakub Jelinek and Ulrich Drepper.
3411 * Old code for several operating systems and machine architectures that
3412 have not been in working condition in a long time have been removed from
3413 the main source tree maintained by the GNU C Library's maintainers.
3414 These files are now reside in the separate `ports' source module
3415 that is usable as an add-on when building the library.
3419 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3420 interrogate the dynamic linker, compatible with the Solaris interface.
3422 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3423 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3425 * getifaddrs now uses the netlink interface on Linux to get its information.
3426 Implemented by Thorsten Kukuk.
3428 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3429 Implemented by Ulrich Drepper.
3431 * support for non-executable stacks on x86 has been added. Changes mostly
3434 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3437 * getaddrinfo now performs destination address selection according to
3442 * Thread-safe interfaces for many functions that access locale data
3443 were added in version 2.3, but these features were omitted from NEWS.
3444 Many functions have variants with an `_l' suffix that take a `locale_t'
3445 object as a parameter rather than consulting the current locale.
3446 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3447 create and maintain `locale_t' objects. Additionally, the new function
3448 `uselocale' sets "the current locale" (as used by functions not so
3449 parameterized) set for an individual thread. These features were added
3450 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3452 * The functions getresuid, getresgid, setresuid, and setresgid, which
3453 have long been available on Linux, are now declared in <unistd.h>
3454 and are now also available on the Hurd.
3456 * ELF thread-local storage support (TLS) now works on x86-64.
3458 * The new dynamic string token $LIB is expanded in shared library names.
3459 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3461 * Aldy Hernandez contributed complete software floating point support for
3462 PowerPC machines with no FPU.
3464 * fexecve is implemented on Linux.
3466 * The `btowc' function should work at least twice as fast due to
3467 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3469 * With approriate thread add-ons cancelable functions are now implemented
3470 in libc.so as well. No need to call the function in libpthread. This
3471 change allowed to finally disable the incorrect and expensive handling
3472 of weak definition in ld.so.
3474 * Yet more PLT entries in libc.so have been removed. We finally arrived
3475 at the bare minimum. Startup times improved appropriately.
3477 * Support for the new Linux/x86 system call interface was added. The
3478 AT_SYSINFO auxiliary vector entry is recognized and handled.
3482 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3485 * iconv (the program and the interface) now accepts empty names (excluding
3486 options like //TRANSLIT) to mean "use charset of current locale".
3488 * localedef can now transliterate characters in strings which are not in
3489 the provided charmap. The information from the input locale is used.
3491 * Prelinking support was added for ELF targets. This requires additional
3492 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3494 * Read-only stdio streams now use mmap to speed up operation by eliminating
3495 copying and buffer underflows. To use add 'm' to the mode string of
3496 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3498 * The malloc functions were completely rewritten by Wolfram Gloger based
3499 on Doug Lea's malloc-2.7.0.c.
3501 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3502 implementation of regex.
3504 * Bruno Haible upgraded the iconv and locale implementation to support
3507 * Contents of the LC_* and LANG environment variables in the CEN style are
3508 not recognized anymore. It never was used. Change by Ulrich Drepper.
3510 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3511 thread-local storage (TLS) ABI on some platforms.
3512 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3514 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3515 EUC-JISX0213, and TSCII.
3517 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3518 BSD-compatible interface for getting all network interface addresses.
3519 Implementation for IPv4 by Roland McGrath.
3521 * Loading of locale data is faster due to the introduction of a locale
3522 archive. Implemented by Roland McGrath and Ulrich Drepper.
3524 * Startup times are significantly reduced by not using exported functions
3525 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3528 * Steven Munroe contributed a port to PowerPC64/Linux.
3532 * The Hurd now uses the GNU libio implementation of stdio rather than the
3533 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3535 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3536 with the same behavior as the Linux system call of the same name.
3540 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3541 128-bit long double format.
3543 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3544 IBM1161, and IBM1162 charsets.
3546 * Andreas Jaeger contributed a port to x86-64/Linux.
3548 * Peter Bruin contributed a port to PowerPC/Hurd.
3550 * libc functions using I/O streams now can handle wide-oriented streams
3553 * optimizations in the dynamic linker. Binaries created by recent binutils
3554 versions start up quicker due to reduced time spent on relocations.
3556 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3560 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3561 asinh, atanh, j0 for the 128-bit long double format.
3563 * Bruno Haible updated all the code handling Unicode in some form to
3564 support Unicode 3.1.
3566 * Speed of regex for single-byte locales is back to previous levels.
3567 Patch by Isamu Hasegawa.
3569 * Alpha, SPARC, and IA-64 now also using floating stacks.
3571 * Startup time of internationalized applications greatly improved through
3572 iconv cache. Use iconvconfig to generate the cache file.
3573 Contributed by Ulrich Drepper.
3575 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3576 severe performance problems. Patch by David Mosberger.
3578 * The Hurd port got a lot more functionality like AIO, various stdio
3579 extensions, etc. Mainly done by Roland McGrath.
3581 * mtrace can now lookup symbols in shared libraries.
3585 * Intel's IA-64 math library is largely integrated. It provides fast and
3586 accurate implementatations for most basic and standard math functions
3587 in float, double, and long double format.
3589 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3590 and asin for the 96-bit long double format and asin, log, tan for the
3591 128-bit long double format.
3593 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3594 The basic double functions exist today. Contributed by Abraham Ziv
3595 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3596 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3598 * An asynchronous name lookup library was added. The interface is designed
3599 after POSIX AIO. The proposal was circulated beforehand to get comments.
3600 No negative ones came in. Implemented by Ulrich Drepper.
3602 * Port to S390/64bit contributed by Martin Schwidefsky
3603 <schwidefsky@de.ibm.com>.
3605 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3606 of functions for Linux/IA-64.
3608 * The RPC code is now thread safe. Threads can now use the same service
3609 of different services at the same time. Patch by Eric Norum
3610 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3612 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3613 family of functions for Linux/S390.
3615 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3616 of functions for Linux/x86.
3618 * Port to Linux/CRIS contributed by Axis Communications.
3622 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3623 we can now check for namespace violations and missing declarations. The
3624 result is that almost all headers are now Unix-compliant (as defined in
3625 the upcoming XPG6). The negative side is that some programs might need
3626 corrections, too, if they depend on the incorrect form of the headers in
3627 previous versions which defined too many symbols and included too many
3630 * regex now handles multibyte character sets correctly.
3631 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3633 * iconv (the program) does now conform to the upcoming XPG6 and handles
3634 charmaps. Instead of the charset names the path of charmaps can be
3635 provided and the conversion happens based on this data.
3636 Contributed by Ulrich Drepper.
3638 * The locale program now provides detailed information about the installed
3643 only lists the names of the supported locales
3647 provides details such as country, language, and codeset name.
3648 Contributed by Ulrich Drepper.
3652 * The gencat program now parses the input file according to the charset
3653 selected by the LC_CTYPE category. This is important for stateful
3654 character sets. To make generating catalogs easier there is a way
3655 to overwrite the charset selected by the locale: before the first
3656 message or $ quote line the catalog can contain a line like
3658 $ codeset=ISO-8859-2
3660 to select the charset (ISO-8859-2 in this case).
3662 Implemented by Shinya Hanataka and Ulrich Drepper.
3664 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3665 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3666 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3668 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3669 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3671 * The LANGUAGE environment variable is now ignored unless the locale is
3672 changed from the default "C" locale.
3674 * The usual bug fixes.
3678 * Greg McGary added runtime support for bounds checking using gcc's
3679 new -fbounded-pointers option. ix86 target is complete. PowerPC
3682 * Thorsten Kukuk added secure mode support to nscd.
3684 * The Berkeley db libraries have been removed.
3686 Related, the nss_db module is now in a separate package since it
3687 obviously requires a database library being available.
3689 * Wide character I/O streams implemented by Ulrich Drepper.
3691 * Functions from the extended socket API added by Ulrich Drepper.
3693 * Functions feenableexcept and fedisableexcept to control the
3694 behaviour of individual exceptions have been added by Andreas Jaeger.
3696 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3698 * The resolver code has been updated from bind 8.2.3-T5B which supports
3699 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3702 This change could in some situations effect backward compatibility. Since
3703 now `_res' is a thread-local instead of a global variable, modifying it
3704 in one thread does not have any effect in other threads.
3706 The resolver library was also extended to allow IPv6 as the transport
3707 protocol for the requests. This work was done by Stig Venaas.
3709 * Compatibility code for K&R C compilers has been removed from the
3710 header files. A ISO C compiler is needed to use the library
3711 (conforming to either C89 or C99 standard).
3713 * Complete rewrite of the localedef program to support multibyte character
3714 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3715 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3716 Implemented by Ulrich Drepper.
3718 Bruno Haible significantly improved the generation and use of the data
3719 structures for the wide character tables.
3721 * Plural handling in gettext implemented by Ulrich Drepper.
3723 * The utmp daemon has been removed.
3725 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3727 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3730 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3732 * POSIX barriers implemented by Kaz Kylheku.
3734 * POSIX spawn function family implemented by Ulrich Drepper.
3736 * POSIX spinlocks are now available.
3738 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3740 * the configure option --enable-kernel=X.Y.Z allows to strip out
3741 compatibility for kernel versions before X.Y.Z. This is currently only
3742 implemented for Linux.
3744 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3745 field sin6_scope_id. This means that all programs using IPv6 should be
3746 recompiled. Don't expect binary compatibility with previous glibc
3749 * various conversion modules for IBM character sets contributed by
3752 * IA-64 port by Jes Sorensen and HJ Lu.
3766 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3768 * New xdr functions are added; some rpc functions are now 64bit clean.
3770 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3772 * Fixed known incompatibilities with glibc 2.0.
3774 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3776 * Optimized a number of functions (especially the ELF dynamic loader).
3778 * Update timezone data files.
3780 * lots of charmaps corrections
3782 * some new locale definitions and charmaps
3787 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3788 conform to POSIX member type requirements. Symbol versions have been
3789 adjusted as needed within the library, and for direct use by applications,
3790 but there is potential for problems if third-party libraries use
3791 struct timeval as part of their interface. This does not present
3792 a problem for X and other "essential" system libraries.
3794 * An additional locale model to support C++ Standard Library locale
3795 model and probably more was implemented by Ulrich Drepper.
3797 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3800 * Miles Bader provided the `argp' function family to support hierachical
3801 command line argument parsing, layered on top of getopt.
3803 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3805 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3808 * scanf recognizes the %a and %A format for scanning floating point numbers.
3810 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3811 information and interfaces for the available integer types.
3813 * about 130 new math functions were added to implement the ISO C9x math
3816 * the new header <complex.h> contains definitions of the complex math
3817 functions from ISO C 9X.
3819 * the new header <tgmath.h> defines generic macros to use complex or
3820 real valued functions.
3822 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3824 * Andreas Jaeger provided a test suite for the math library.
3826 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3828 * Ulrich Drepper added character set conversion functions (iconv).
3830 * Optimized string functions have been added.
3832 * The localedata addon is now part of glibc.
3834 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3836 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3837 daemon for NSS (nscd).
3839 Missing a better place here are some numbers on improvements. Under
3840 Linux 2.1.125 un-tar-ing the kernel sources takes
3844 using local files 12.19s 6.88s 22.91s
3846 using NIS 13.92s 8.91s 26.34s
3848 using NIS & nscd 10.37s 7.34s 25.30s
3850 using NIS+ 27.57s 30.37s 640.46s
3852 using NIS+ & nscd 10.25s 7.83s 26.51s
3854 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3856 Keep in mind that non-namelookup related operations dominate above times.
3857 It was just a common complain that using NIS+ unpacking the kernel is
3860 [1] The old nscd implementation is not available anymore since it was
3861 distributed with glibc up to version 2.0.98 and thus is now replaced.
3863 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3865 * Mark Kettenis provided a Hesiod NSS module.
3867 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3868 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3870 * rcmd can now handle netgroups (Dick Streefland).
3872 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3875 * Support for the IPv6 protocol has been added to the socket API, as per the
3876 latest draft standards.
3878 * Support for Linux 2.2 has been added.
3880 * Interface changes relative to the latest 2.0.x release:
3881 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3882 addseverity NEW: Unix98
3883 alphasort64 NEW: LFS
3884 argp_err_exit_status NEW: argp, GNU ext
3885 argp_error NEW: argp, GNU ext
3886 argp_failure NEW: argp, GNU ext
3887 argp_help NEW: argp, GNU ext
3888 argp_parse NEW: argp, GNU ext
3889 argp_program_bug_address NEW: argp, GNU ext
3890 argp_program_version NEW: argp, GNU ext
3891 argp_program_version_hook NEW: argp, GNU ext
3892 argp_state_help NEW: argp, GNU ext
3893 argp_usage NEW: argp, GNU ext
3894 authdes_create NEW: Secure RPC
3895 authdes_getucred NEW: Secure RPC
3896 authdes_pk_create NEW: Secure RPC
3897 backtrace NEW: GNU ext.
3898 backtrace_symbols NEW: GNU ext.
3899 backtrace_symbols_fd NEW: GNU ext.
3901 cacosf NEW: ISO C 9x
3902 cacosh NEW: ISO C 9x
3903 cacoshf NEW: ISO C 9x
3904 cacoshl NEW: ISO C 9x
3905 cacosl NEW: ISO C 9x
3912 casinf NEW: ISO C 9x
3913 casinh NEW: ISO C 9x
3914 casinhf NEW: ISO C 9x
3915 casinhl NEW: ISO C 9x
3916 casinl NEW: ISO C 9x
3918 catanf NEW: ISO C 9x
3919 catanh NEW: ISO C 9x
3920 catanhf NEW: ISO C 9x
3921 catanhl NEW: ISO C 9x
3922 catanl NEW: ISO C 9x
3923 cbc_crypt NEW: Secure RPC
3927 ccoshf NEW: ISO C 9x
3928 ccoshl NEW: ISO C 9x
3934 cimagf NEW: ISO C 9x
3935 cimagl NEW: ISO C 9x
3936 clearerr_locked REMOVED
3937 clntunix_create NEW: sunrpc ext
3939 clog10 NEW: ISO C 9x
3940 clog10f NEW: ISO C 9x
3941 clog10l NEW: ISO C 9x
3951 cprojf NEW: ISO C 9x
3952 cprojl NEW: ISO C 9x
3954 crealf NEW: ISO C 9x
3955 creall NEW: ISO C 9x
3960 csinhf NEW: ISO C 9x
3961 csinhl NEW: ISO C 9x
3964 csqrtf NEW: ISO C 9x
3965 csqrtl NEW: ISO C 9x
3969 ctanhf NEW: ISO C 9x
3970 ctanhl NEW: ISO C 9x
3972 des_setparity NEW: Secure RPC
3973 ecb_crypt NEW: Secure RPC
3974 endutxent NEW: Unix98
3976 exp10f NEW: ISO C 9x
3977 exp10l NEW: ISO C 9x
3981 fattach NEW: STREAMS
3982 fdetach NEW: STREAMS
3986 feclearexcept NEW: ISO C 9x
3987 fegetenv NEW: ISO C 9x
3988 fegetexceptflag NEW: ISO C 9x
3989 fegetround NEW: ISO C 9x
3990 feholdexcept NEW: ISO C 9x
3992 feraiseexcept NEW: ISO C 9x
3993 ferror_locked REMOVED
3994 fesetenv NEW: ISO C 9x
3995 fesetexceptflag NEW: ISO C 9x
3996 fesetround NEW: ISO C 9x
3997 fetestexcept NEW: ISO C 9x
3998 feupdateenv NEW: ISO C 9x
3999 fflush_locked REMOVED
4003 fgets_unlocked NEW: GNU ext.
4004 fileno_locked REMOVED
4016 fputc_locked REMOVED
4017 fputs_unlocked NEW: GNU ext.
4018 fread_unlocked NEW: GNU ext.
4023 fstatvfs NEW: Unix98
4027 ftruncate64 NEW: LFS
4029 fwrite_unlocked NEW: GNU ext.
4030 gai_strerror NEW: IPv6
4034 getchar_locked REMOVED
4036 getdate_err NEW: Unix98
4037 getdate_r NEW: GNU ext.
4039 getnameinfo NEW: IPv6
4040 getnetname NEW: Secure RPC
4041 getpmsg NEW: STREAMS
4042 getpt NEW: Unix98 PTY
4043 getrlimit64 NEW: LFS
4044 getutxent NEW: Unix98
4045 getutxid NEW: Unix98
4046 getutxline NEW: Unix98
4047 glob64 NEW: GNU ext.
4048 globfree64 NEW: GNU ext.
4049 gnu_get_libc_release NEW: GNU ext.
4050 gnu_get_libc_version NEW: GNU ext.
4051 grantpt NEW: Unix98 PTY
4052 host2netname NEW: Secure RPC
4054 iconv_close NEW: iconv
4055 iconv_open NEW: iconv
4056 if_freenameindex NEW: IPv6
4057 if_indextoname NEW: IPv6
4058 if_nameindex NEW: IPv6
4059 if_nametoindex NEW: IPv6
4060 in6addr_any NEW: IPv6
4061 in6addr_loopback NEW: IPv6
4062 inet6_isipv4mapped NEW: IPv6
4063 isastream NEW: STREAMS
4064 iswblank NEW: GNU ext.
4065 key_decryptsession NEW: Secure RPC
4066 key_decryptsession_pk NEW: Secure RPC
4067 key_encryptsession NEW: Secure RPC
4068 key_encryptsession_pk NEW: Secure RPC
4069 key_gendes NEW: Secure RPC
4070 key_get_conv NEW: Secure RPC
4071 key_secretkey_is_set NEW: Secure RPC
4072 key_setnet NEW: Secure RPC
4073 key_setsecret NEW: Secure RPC
4074 llrint NEW: ISO C 9x
4075 llrintf NEW: ISO C 9x
4076 llrintl NEW: ISO C 9x
4077 llround NEW: ISO C 9x
4078 llroundf NEW: ISO C 9x
4079 llroundl NEW: ISO C 9x
4084 lrintf NEW: ISO C 9x
4085 lrintl NEW: ISO C 9x
4086 lround NEW: ISO C 9x
4087 lroundf NEW: ISO C 9x
4088 lroundl NEW: ISO C 9x
4090 makecontext NEW: Unix98
4091 mempcpy NEW: GNU ext.
4094 modify_ldt NEW: kernel
4098 nearbyint NEW: ISO C 9x
4099 nearbyintf NEW: ISO C 9x
4100 nearbyintl NEW: ISO C 9x
4101 netname2host NEW: Secure RPC
4102 netname2user NEW: Secure RPC
4103 nexttoward NEW: ISO C 9x
4104 nexttowardf NEW: ISO C 9x
4105 nexttowardl NEW: ISO C 9x
4109 passwd2des NEW: Secure RPC
4111 pow10f NEW: GNU ext.
4112 pow10l NEW: GNU ext.
4115 printf_size NEW: GNU ext.
4116 printf_size_info NEW: GNU ext.
4117 profil_counter REMOVED
4118 pthread_mutexattr_getkind_np REPLACED
4119 pthread_mutexattr_setkind_np REPLACED
4120 ptsname NEW: Unix98 PTY
4121 ptsname_r NEW: Unix98 PTY
4123 putchar_locked REMOVED
4124 putgrent NEW: GNU ext.
4126 putpmsg NEW: STREAMS
4127 pututxline NEW: Unix98
4131 readdir64_r NEW: LFS
4132 remquo NEW: ISO C 9x
4133 remquof NEW: ISO C 9x
4134 remquol NEW: ISO C 9x
4136 roundf NEW: ISO C 9x
4137 roundl NEW: ISO C 9x
4139 scalbln NEW: ISO C 9x
4140 scalblnf NEW: ISO C 9x
4141 scalblnl NEW: ISO C 9x
4143 sendfile NEW: kernel
4144 setcontext NEW: Unix98
4145 setrlimit64 NEW: LFS
4146 setutxent NEW: Unix98
4148 sigignore NEW: Unix98
4149 sigqueue NEW: POSIX.1b
4150 sigrelse NEW: Unix98
4151 sigset NEW: POSIX.1b
4152 sigtimedwait NEW: POSIX.1b
4153 sigwaitinfo NEW: POSIX.1b
4154 sincos NEW: GNU ext.
4155 sincosf NEW: GNU ext.
4156 sincosl NEW: GNU ext.
4160 strcasestr NEW: GNU ext.
4161 strtoimax NEW: ISO C 9x
4162 strtoumax NEW: ISO C 9x
4163 strverscmp NEW: GNU ext.
4164 svcauthdes_stats NEW: Secure RPC
4165 svcunix_create NEW: sunrpc etx
4166 svcunixfd_create NEW: sunrpc ext
4167 swapcontext NEW: Unix98
4168 tcgetsid NEW: Unix98 PTY
4169 tdestroy NEW: GNU ext.
4170 tgamma NEW: ISO C 9x
4171 tgammaf NEW: ISO C 9x
4172 tgammal NEW: ISO C 9x
4176 truncf NEW: ISO C 9x
4177 truncl NEW: ISO C 9x
4179 unlockpt NEW: Unix98 PTY
4180 updwtmpx NEW: Unix98
4181 user2netname NEW: Secure RPC
4182 utmpxname NEW: Unix98
4183 versionsort NEW: GNU ext.
4184 versionsort64 NEW: GNU ext.
4186 wcscasecmp NEW: GNU ext.
4187 wcsncasecmp NEW: GNU ext.
4188 wcsnlen NEW: GNU ext.
4189 wcstoimax NEW: ISO C 9x
4190 wcstoll NEW: ISO C 9x
4191 wcstoull NEW: ISO C 9x
4192 wcstoumax NEW: ISO C 9x
4194 wordexp NEW: POSIX.2
4195 wordfree NEW: POSIX.2
4196 write_profiling REMOVED
4197 xdecrypt NEW: Secure RPC
4198 xdr_authdes_cred NEW: Secure RPC
4199 xdr_authdes_verf NEW: Secure RPC
4200 xdr_cryptkeyarg NEW: Secure RPC
4201 xdr_cryptkeyarg2 NEW: Secure RPC
4202 xdr_cryptkeyres NEW: Secure RPC
4203 xdr_getcredres NEW: Secure RPC
4204 xdr_key_netstarg NEW: Secure RPC
4205 xdr_key_netstres NEW: Secure RPC
4206 xdr_keybuf NEW: Secure RPC
4207 xdr_keystatus NEW: Secure RPC
4208 xdr_netnamestr NEW: Secure RPC
4209 xdr_sizeof NEW: Secure RPC
4210 xdr_unixcred NEW: sunrpc ext
4211 xencrypt NEW: Secure RPC
4212 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4223 * inet_ntoa is thread-safe
4225 * updwtmp is moved from libutil to libc
4227 * rewrite of cbrt function
4229 * update of timezone data
4243 * add atoll function
4245 * fix complex problems in Berkeley DB code
4247 * fix math functions
4251 * fixed lots of header problems (especially Linux/GNU specific)
4253 * dynamic loader preserves all registers
4255 * Roland McGrath provided support for handling of auxiliary objects in
4256 the ELF dynamic loader.
4258 * support for parallel builds is improved
4262 * GNU extensions are no longer declared by default. To enable them you
4263 must define the macro `_GNU_SOURCE' in your program or compile with
4266 * The library has changed from using GNU ld symbol aliases to using weak
4267 symbols where available. The ELF object file format supports weak
4268 symbols; GNU ld also supports weak symbols in the a.out format. (There
4269 is also now support for other GNU ld extensions in ELF. Use the
4270 `--with-elf' option to configure to indicate you have ELF, and
4271 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4272 of many files which contained only symbol aliases, reducing the size of
4273 the source and the compiled library; many other files were renamed to
4274 less cryptic names previously occupied by the symbol alias files.
4275 There is a new header file <elf.h> for programs which operate on
4276 files in the ELF format.
4278 * Converted to Autoconf version 2, so `configure' has more options.
4279 Run `configure --help' to see the details.
4281 * The library can now be configured to build profiling, highly-optimized
4282 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4283 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4284 `configure' enable building these extra libraries. The shared library is
4285 built by default when using both ELF and GNU ld. When shared libraries
4286 are enabled, the new library `-ldl' is available for arbitrary run-time
4287 loading of shared objects; its interface is defined in <dlfcn.h>. The
4288 new header file <link.h> gives access to the internals of the run-time
4289 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4290 application of same name on other systems and it provides information
4291 about dynamically linked binaries.
4293 * The C library now provides the run-time support code for profiling
4294 executables compiled with `-pg'. Programs can control the profiling code
4295 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4296 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4297 the support for this file format was contributed by David Mosberger-Tang.
4299 * The math code has been replaced with a math library based on fdlibm from
4300 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4301 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4302 functions. The math functions now reside in a separate library, so
4303 programs using them will need to use `-lm' their linking commands.
4305 * John C. Bowman contributed optimized ix87 assembler inline functions.
4307 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4308 mechanism similar to that found in Solaris 2. This is now used for the
4309 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4310 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4311 file controls what services are used for each individual database. This
4312 works by loading shared libraries with names specified in `nsswitch.conf',
4313 so service modules can be changed or added at any time without even
4314 relinking any program. Currently there are the file, db, and NIS based
4315 NSS services available.
4317 * The new functions `strtoq' and `strtouq' parse integer values from
4318 strings, like `strtol' and `strtoul', but they return `long long int' and
4319 `unsigned long long int' values, respectively (64-bit quantities).
4321 * The new functions `strtof' and `strtold' parse floating-point values from
4322 strings, like `strtod', but they return `float' and `long double' values,
4323 respectively (on some machines `double' and `long double' are the same).
4325 * Ulrich Drepper has contributed new implementations of the floating-point
4326 printing and reading code used in the `printf' family of functions and
4327 `strtod', `strtof', and `strtold'. These new functions are perfectly
4328 accurate, and much faster than the old ones.
4330 * The implementation of the POSIX locale model was completely rewritten by
4331 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4332 to compile the POSIX locale definition.
4334 * The former dummy implementations of the strcoll and strxfrm function are
4335 now replaced by fully functional code contributed by Ulrich Drepper. The
4336 collation information comes from the POSIX locale definitions.
4338 * The new header <langinfo.h> defines an interface for accessing
4339 various locale-dependent data (using the locale chosen with `setlocale').
4341 * Ulrich Drepper has contributed a new suite of functions for operation on
4342 wide-character and multibyte-character strings, in <wchar.h>;
4343 and classification and case conversion of wide characters, in <wctype.h>.
4344 These new functions are conforming to the ISO C, Amendement 1 specification.
4346 * There is now a second implementation of the standard I/O library available.
4347 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4348 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4350 * You can now use positional parameter specifications in format strings
4351 for the `printf' and `scanf' families of functions. For example,
4352 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4353 ``Number 6, Mr Jones''. This is mainly useful when providing different
4354 format strings for different languages, whose grammars may dictate
4355 different orderings of the values being printed. To support this
4356 feature, the interface for `register_printf_handler' has changed; see
4357 the header file <printf.h> for details.
4359 * The `printf' and `scanf' families of functions now understand a new
4360 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4361 to group numbers as indicated by the locale; for `scanf' and friends, this
4362 says to accept as valid only a number with all the proper grouping
4363 separators in the right places. In the default "C" locale, numbers are
4364 not grouped; but locales for specific countries will define the usual
4365 conventions (i.e. separate thousands with `,' in the US locale).
4367 * The pgrp functions have been regularized, slightly incompatibly but much
4368 less confusingly. The core functions are now `getpgid' and `setpgid',
4369 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4370 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4371 provided for compatibility. There is no longer an incompatible `getpgrp'
4372 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4374 * The new header file <fts.h> and suite of functions simplify programs that
4375 operate on directory trees. This code comes from 4.4 BSD.
4377 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4378 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4379 NSS scheme used in glibc.
4381 * There is a new malloc debugging hook `__memalign_hook'.
4383 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4384 `unsigned int' in <sys/types.h>. These are for compatibility only and
4385 their use is discouraged.
4387 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4388 done differently, so that it works even without GNU ld.
4390 * New function `euidaccess' checks allowed access to a file like `access',
4391 but using the effective IDs instead of the real IDs.
4393 * The time zone data files have been updated for the latest and greatest
4394 local time conventions of the countries of the world.
4396 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4399 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4400 for formatting floating-point numbers. They are provided only for
4401 compatibility; new programs should use `sprintf' instead. There are
4402 also equivalent function for the `long double' floating-point type and
4403 all functions also exist in a reentrant form.
4405 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4406 functions for maintaining the login-record files (primarily of use to
4407 system programs such as `login'), and convenient functions for
4408 allocating and initializing a pseudo-terminal (pty) device.
4410 * Ulrich Drepper has contributed new support for System V style
4411 shared memory and IPC on systems that support it.
4413 * Ulrich Drepper has contributed several miscellaneous new functions found
4414 in System V: The `hsearch' family of functions provide an effective
4415 implementation of hash tables; `a64l' and `l64a' provide a very simple
4416 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4419 * Ulrich Drepper has contributed new reentrant counterparts for the
4420 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4422 * Ulrich Drepper has contributed new, highly-optimized versions of several
4423 string functions for the i486/Pentium family of processors.
4425 * Ulrich Drepper has updated the Linux-specific code, based largely
4426 on work done in Hongjiu Lu's version of GNU libc for Linux.
4427 The GNU library now supports Linux versions 2.0.10 and later,
4428 using the ELF object file format (i[3456]86-*-linux).
4430 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4432 * David Mosberger-Tang and Richard Henderson have ported the C library
4433 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4434 dynamic linking support for ELF/Alpha.
4436 * Richard Henderson contributed several Alpha optimized assembler function
4437 for arithmetic and string handling.
4439 * Ulrich Drepper has contributed a new set of message catalog functions to
4440 support multiple languages using the <libintl.h> interface, for use with
4441 his new package GNU gettext. Translation volunteers have contributed
4442 catalogs of the library's messages in Spanish, German, and Korean.
4444 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4445 program and the `catgets' function for reading the catalog files it
4446 creates. (The <libintl.h> interface is preferred; we include the
4447 <nl_types.h> interface using `catgets' only for source compatibility with
4448 programs already written to use it.)
4450 * New header file <values.h> gives SVID-compatible names for <limits.h>
4453 * Various new macros, declarations, and small header files for compatibility
4456 * New function `group_member' is a convenient way to check if a process has
4457 a given effective group ID.
4459 * When using GCC 2.7 and later, the socket functions are now declared in a
4460 special way so that passing an argument of type `struct sockaddr_in *',
4461 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4462 `struct sockaddr *' type, does not generate a type-clash warning.
4464 * New function `error' declared in header file <error.h> is a convenient
4465 function for printing error messages and optionally exiting; this is the
4466 canonical function used in GNU programs. The new functions `err', `warn',
4467 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4468 doing the same thing.
4470 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4471 POSIX.2 `glob' function to do ~ and {...} expansion.
4473 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4474 `clearenv' which is used in POSIX.9 is also available.
4476 * New function `getsid' returns session ID number on systems that support it.
4478 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4479 files <db.h> and <mpool.h> provide a rich set of functions for several
4480 types of simple databases stored in memory and in files, and <ndbm.h> is
4481 an old `ndbm'-compatible interface using the `db' functions. Link with
4482 `-ldb' to get these functions.
4484 * New macro `strdupa' copies a string like `strdup', but uses local stack
4485 space from `alloca' instead of dynamic heap space from `malloc'.
4487 * New function `strnlen' is like `strlen' but searches only a given maximum
4488 number of characters for the null terminator. `stpncpy', `strndup' and
4489 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4492 * New function `statfs' in header <sys/statfs.h>.
4494 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4495 provide convenient functions for operating on blocks of null-terminated
4498 * A new suite of functions in <utmp.h> handle all the details of reading
4499 and writing the utmp file.
4501 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4504 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4505 completely POSIX compliant and also implemented the extended functionality
4506 to handle alternate digit representation and alternate era date formats.
4508 * Ulrich Drepper provided an implementation of the `strptime' function
4509 defined in XPG4.2 which transforms a string into a `struct tm' value.
4511 * Paul Eggert provided the tzselect shell script as part of the timezone
4512 code. The shell script makes it easy to select the correct timezone
4515 * The implementation of the malloc family of functions is completely replaced
4516 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4517 The implementation uses the mmap function (if available) and it is
4518 optimized for the use in multi threaded programs.
4520 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4521 functions. This new functionality is usable by specifying a special
4522 salt string and it is compatible with implementation on *BSD systems.
4524 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4525 `getsubopt' to handle second level command line options, `bsd_signal'
4526 to access BSD style `signal' functionality, the obsolete `regexp' style
4529 * the `lchown' function is available on system which support this
4532 * The implementation of the shadow password handling function was contributed
4535 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4537 * POSIX.1g support was added. The <sys/select.h> header is available,
4538 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4539 implementation of `getaddrinfo'.
4543 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4545 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4546 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4548 * The new function `daemon' from 4.4 BSD is useful for server programs that
4549 want to put themselves in the background.
4551 * Joel Sherrill has contributed support for several standalone boards that
4552 run without an operating system.
4554 * `printf', `scanf' and friends now accept a `q' type modifier for long
4555 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4557 * All of the code taken from BSD (notably most of the math and networking
4558 routines) has been updated from the BSD 4.4-Lite release.
4560 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4562 * The new functions `getdomainname' and `setdomainname' fetch or change the
4563 YP/NIS domain name. These are system calls which exist on systems which
4566 * The time zone data files have been updated for the latest international
4569 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4570 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4574 * The C library now includes support for Sun RPC, from Sun's free
4575 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4576 are included. (There is still no support for YP.)
4578 * Tom Quinn has contributed a port of the C library to SGI machines running
4579 Irix 4 (mips-sgi-irix4).
4581 * The new `lockf' function is a simplified interface to the locking
4582 facilities of `fcntl', included for compatibility.
4584 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4586 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4588 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4591 * The error message printed by `assert' for a failed assertion now includes
4592 the name of the program (if using GNU ld) and the name of the calling
4593 function (with versions of GCC that support this).
4595 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4597 * The library now includes the <sys/mman.h> header file and memory
4598 management functions `mmap', `munmap', `mprotect', `msync', and
4599 `madvise', on systems that support those facilities.
4601 * The interface for `mcheck' has changed slightly: the function called to
4602 abort the program when an allocation inconsistency is detected now takes
4603 an argument that indicates the type of failure. The new function
4604 `mprobe' lets you request a consistency check for a particular block at
4605 any time (checks are normally done only when you call `free' or `realloc'
4608 * It is now possible to easily cross-compile the C library, building on one
4609 system a library to run on another machine and/or operating system. All
4610 you need to do is set the variable `HOST_CC' in `configparms' to the
4611 native compiler for programs to run on the machine you are building on (a
4612 few generator programs are used on Unix systems); set `CC' to the
4615 * The new function `fexecve' (only implemented on the GNU system) executes
4616 a program file given a file descriptor already open on the file.
4620 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4621 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4623 * You can set the variable `libprefix' in `configparms' to specify a prefix
4624 to be prepended to installed library files; this makes it easy to install
4625 the GNU C library to be linked as `-lgnuc' or whatever.
4627 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4628 copies a limited number of characters from a string, and returns the
4629 address of the last character written.
4631 * You no longer need to check for whether the installed `stddef.h' is
4632 compatible with the GNU C library. configure now checks for you.
4634 * You can now define a per-stream `fileno' function to convert the
4635 stream's cookie into an integral file descriptor.
4637 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4638 allocates zero bytes of storage, and returns a unique pointer which
4639 you can pass to `realloc' or `free'. The behavior is undefined if
4640 you dereference this pointer.
4642 * The C library now runs on Sony NEWS m68k machines running either
4643 NewsOS 3 or NewsOS 4.
4645 * The new `syscall' function is a system-dependent primitive function
4646 for invoking system calls. It has the canonical behavior on Unix
4647 systems, including unreliable return values for some calls (such as
4648 `pipe', `fork' and `getppid').
4650 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4651 to `EAGAIN', which is the preferred name. On systems whose kernels
4652 use two distinct codes, the C library now translates EWOULDBLOCK to
4653 EAGAIN in every system call function.
4657 * The GNU C Library Reference Manual is now distributed with the library.
4658 `make dvi' will produce a DVI file of the printed manual.
4659 `make info' will produce Info files that you can read on line using C-h i
4660 in Emacs or the `info' program.
4661 Please send comments on the manual to bug-glibc-manual@gnu.org.
4663 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4665 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4667 * Jason Merrill has contributed a port to the Sequent Symmetry running
4668 Dynix version 3 (i386-sequent-dynix).
4670 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4671 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4673 * New function `memory_warnings' lets you arrange to get warnings when
4674 malloc is running out of memory to allocate, like Emacs gives you.
4676 * The C library now contains the relocating allocator used in Emacs 19 for
4677 its editing buffers. This allocator (ralloc) minimizes allocation
4678 overhead and fragmentation by moving allocated regions around whenever it
4679 needs to. You always refer to a ralloc'd region with a "handle" (a
4680 pointer to a pointer--an object of type `void **').
4682 * There is a new `printf' format: `%m' gives you the string corresponding
4683 to the error code in `errno'.
4685 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4686 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4687 pass, the `a' modifier says to fill in a `char **' you pass with a
4690 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4691 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4692 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4694 * `mkstemp' is a traditional Unix function to atomically create and open a
4695 uniquely-named temporary file.
4699 * The standard location for the file that says what the local timezone is
4700 has changed again. It is now `/usr/local/etc/localtime' (or more
4701 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4703 * The distribution no longer contains any files with names longer than 14
4706 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4707 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4709 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4713 * The configuration process has changed quite a bit. The `configure'
4714 script is now used just like the configuration scripts for other GNU
4715 packages. The `sysdeps' directory hierarchy is much rearranged.
4716 The file `INSTALL' explains the new scheme in detail.
4718 * The header files no longer need to be processed into ANSI C and
4719 traditional C versions. There is just one set of files to install, and
4720 it will work with ANSI or old C compilers (including `gcc -traditional').
4722 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4723 MIPS DECStation running Ultrix 4.
4725 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4726 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4727 made itself into a shared library.
4729 * Yet further improved support for the i386, running 4.3 BSD-like systems
4730 (such as Mach 3 with the Unix single-server), or System V.
4732 * New function `strncasecmp' to do case-insensitive string comparison
4733 with limited length.
4735 * New function `strsep' is a reentrant alternative to `strtok'.
4737 * New functions `scandir' and `alphasort' for searching directories.
4739 * New function `setenv' is a better interface to `putenv'.
4741 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4742 function for traversing a directory tree.
4744 * The GNU obstack package is now also part of the C library.
4745 The new function `open_obstack_stream' creates a stdio stream that
4746 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4747 formatted output directly to an obstack.
4749 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4750 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4752 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4754 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4755 things to your strings.
4757 * There are some new test programs: test-fseek, testmb, and testrand.
4759 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4760 These ports are not finished, but are a good starting place for really
4761 supporting those systems.
4763 * `/etc/localtime' is now the standard location for the file that says what
4764 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4765 This follows the general principle that `/etc' is the place for all local
4766 configuration files.
4768 * The C library header files now use `extern "C"' when used by the C++
4769 compiler, so the C library should now work with C++ code.
4771 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4772 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4775 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4776 function which uses the `merge sort' algorithm, and is said to be
4777 significantly faster than the old GNU `qsort' function. Merge sort is
4778 now the standard `qsort' function. The new algorithm can require a lot
4779 of temporary storage; so, the old sorting function is called when the
4780 required storage is not available.
4782 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4783 provides the Unix `crypt' function, plus some other entry points.
4785 * The code and header files taken from 4.4 BSD have been updated with the
4786 latest files released from Berkeley.
4788 ----------------------------------------------------------------------
4789 Copyright information:
4791 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4793 Permission is granted to anyone to make or distribute verbatim copies
4794 of this document as received, in any medium, provided that the
4795 copyright notice and this permission notice are preserved,
4796 thus giving the recipient permission to redistribute in turn.
4798 Permission is granted to distribute modified versions
4799 of this document, or of portions of it,
4800 under the above conditions, provided also that they
4801 carry prominent notices stating who last changed them.
4804 version-control: never