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 Changes to build and runtime requirements:
33 [Add changes to build and runtime requirements here]
35 Security related changes:
37 [Add security related changes here]
39 The following bugs are resolved with this release:
41 [The release manager will add the list generated by
42 scripts/list-fixed-bugs.py just before the release.]
49 * The GNU C Library can now be compiled with support for building static
50 PIE executables (See --enable-static-pie in INSTALL). These static PIE
51 executables are like static executables but can be loaded at any address
52 and provide additional security hardening benefits at the cost of some
53 memory and performance. When the library is built with --enable-static-pie
54 the resulting libc.a is usable with GCC 8 and above to create static PIE
55 executables using the GCC option '-static-pie'. This feature is currently
56 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
57 aarch64 with binutils 2.30 or later.
59 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
60 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
63 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
65 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
67 * In order to support faster and safer process termination the malloc API
68 family of functions will no longer print a failure address and stack
69 backtrace after detecting heap corruption. The goal is to minimize the
70 amount of work done after corruption is detected and to avoid potential
71 security issues in continued process execution. Reducing shutdown time
72 leads to lower overall process restart latency, so there is benefit both
73 from a security and performance perspective.
75 * The abort function terminates the process immediately, without flushing
76 stdio streams. Previous glibc versions used to flush streams, resulting
77 in deadlocks and further data corruption. This change also affects
78 process aborts as the result of assertion failures.
80 * On platforms where long double has the IEEE binary128 format (aarch64,
81 alpha, mips64, riscv, s390 and sparc), the math library now implements
82 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
83 These are the same interfaces added in version 2.26 for some platforms where
84 this format is supported but is not the format of long double.
86 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
87 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
88 implements interfaces for that type, as defined by ISO/IEC TS
89 18661-3:2015. These are corresponding interfaces to those supported for
92 * The math library now implements interfaces for the _Float32, _Float64 and
93 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
94 corresponding interfaces to those supported for _Float128.
96 * glibc now implements the memfd_create and mlock2 functions on Linux.
98 * Support for memory protection keys was added. The <sys/mman.h> header now
99 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
102 * The copy_file_range function was added.
104 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
106 * The ldconfig utility now processes `include' directives using the C/POSIX
107 collation ordering. Previous glibc versions used locale-specific
108 ordering, the change might break systems that relied on that.
110 * Support for two grammatical forms of month names has been added.
111 In a call to strftime, the "%B" and "%b" format specifiers will now
112 produce the grammatical form required when the month is used as part
113 of a complete date. New "%OB" and "%Ob" specifiers produce the form
114 required when the month is named by itself. For instance, in Greek
115 and in many Slavic and Baltic languages, "%B" will produce the month
116 in genitive case, and "%OB" will produce the month in nominative case.
118 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
119 are all valid and will all accept any known form of month
120 name---standalone or complete, abbreviated or full. In a call to
121 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
122 the strings used by "%B" and "%b", respectively. New query
123 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
124 used by "%OB" and "%Ob", respectively.
126 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
127 define the strings for %OB and %Ob, respectively; these have the
128 same syntax as "mon" and "abmon". These arrays are optional; if they
129 are not provided then they have the same content as "mon" and "abmon",
132 These features are provided for locales which define "alt_mon" and/or
133 "ab_alt_mon" in their locale source data. This release includes such
134 alternative month name data for the following languages: Belarusian,
135 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
137 This feature is currently a GNU extension, but it is expected to
138 be added to the next revision of POSIX, and it is also already
139 available on some BSD-derived operating systems.
141 This feature will cause existing statically compiled applications
142 to fail to load locales and fall back to the builtin C/POSIX locales.
143 See notes below for other changes affecting compatibility.
145 * Support for the RISC-V ISA running on Linux has been added. This port
146 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
147 for the following ISA and ABI pairs:
153 Deprecated and removed features, and other changes affecting compatibility:
155 * Statically compiled applications attempting to load locales compiled for the
156 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
157 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
158 support for two grammatical forms of the month names, also extends the locale
159 data binary format. Static applications needing locale support must be
160 recompiled to match the runtime and data they are deployed with. In some
161 distributions there is an upgrade window where dynamically linked applications
162 may use a new library but the old locale data and also fall back to the
163 builtin C/POSIX locales; restarting the application process is sufficient to
166 * Support for statically linked applications which call dlopen is deprecated
167 and will be removed in a future version of glibc. Applications which call
168 dlopen need to be linked dynamically instead.
170 * Support for old programs which use internal stdio data structures and
171 functions is deprecated. This includes programs which use the C++ streams
172 provided by libstdc++ in GCC 2.95. Programs which use the internal
173 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
174 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
175 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
176 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
177 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
178 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
179 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
180 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
181 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
182 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
183 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
184 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
185 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
186 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
187 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
188 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
189 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
190 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
191 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
192 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
193 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
194 future version of glibc. Unlike other symbol removals, these old
195 applications will not be supported using compatibility symbols.
197 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
198 defined by <sys/ptrace.h>.
200 * libm no longer supports SVID error handling (calling a user-provided
201 matherr function on error) or the _LIB_VERSION variable to control error
202 handling. (SVID error handling and the _LIB_VERSION variable still work
203 for binaries linked against older versions of the GNU C Library.) The
204 libieee.a library is no longer provided. math.h no longer defines struct
205 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
206 TLOSS, PLOSS and HUGE.
208 * The libm functions pow10, pow10f and pow10l are no longer supported for
209 new programs. Programs should use the standard names exp10, exp10f and
210 exp10l for these functions instead.
212 * The mcontext_t type is no longer the same as struct sigcontext. On
213 platforms where it was previously the same, this changes the C++ name
214 mangling for interfaces involving this type.
216 * The add-ons mechanism for building additional packages at the same time as
217 glibc has been removed. The --enable-add-ons configure option is now
220 * The --without-fp configure option is now ignored. Whether hardware
221 floating-point instructions are used is now configured based on whether
222 the compiler used at configure time (without any options implied by a
223 --with-cpu= configure option) uses such instructions.
225 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
226 the specified string can be parsed as a domain name.
228 * In the malloc_info output, the <heap> element may contain another <aspace>
229 element, "subheaps", which contains the number of sub-heaps.
231 * The libresolv function p_secstodate is no longer supported for new
234 * The tilepro-*-linux-gnu configuration is no longer supported.
236 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
237 and will be removed in a future release. Software that is still using
238 either header should be updated to use standard <stdio.h> interfaces
241 libio.h was originally the header for a set of supported GNU extensions,
242 but they have not been maintained as such in many years, they are now
243 standing in the way of improvements to stdio, and we don't think there are
244 any remaining external users. _G_config.h was never intended for public
245 use, but predates the bits convention.
247 Changes to build and runtime requirements:
249 * bison version 2.7 or later is required to generate code in the 'intl'
252 Security related changes:
254 CVE-2009-5064: The ldd script would sometimes run the program under
255 examination directly, without preventing code execution through the
256 dynamic linker. (The glibc project disputes that this is a security
257 vulnerability; only trusted binaries must be examined using the ldd
260 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
261 suffered from a one-byte overflow during ~ operator processing (either
262 on the stack or the heap, depending on the length of the user name).
263 Reported by Tim Rühsen.
265 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
266 would sometimes fail to free memory allocated during ~ operator
267 processing, leading to a memory leak and, potentially, to a denial
270 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
271 without GLOB_NOESCAPE, could write past the end of a buffer while
272 unescaping user names. Reported by Tim Rühsen.
274 CVE-2017-17426: The malloc function, when called with an object size near
275 the value SIZE_MAX, would return a pointer to a buffer which is too small,
276 instead of NULL. This was a regression introduced with the new malloc
277 thread cache in glibc 2.26. Reported by Iain Buclaw.
279 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
280 to the allocation of too much memory. (This is not a security bug per se,
281 it is mentioned here only because of the CVE assignment.) Reported by
284 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
285 of the number of search path components. (This is not a security
286 vulnerability per se because no trust boundary is crossed if the fix for
287 CVE-2017-1000366 has been applied, but it is mentioned here only because
288 of the CVE assignment.) Reported by Qualys.
290 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
291 for AT_SECURE or SUID binaries could be used to load libraries from the
294 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
295 succeeds without returning an absolute path due to unexpected behaviour
296 of the Linux kernel getcwd syscall. Reported by halfdog.
298 CVE-2018-6485: The posix_memalign and memalign functions, when called with
299 an object size near the value of SIZE_MAX, would return a pointer to a
300 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
302 CVE-2018-6551: The malloc function, when called with an object size near
303 the value of SIZE_MAX, would return a pointer to a buffer which is too
304 small, instead of NULL.
306 The following bugs are resolved with this release:
308 [866] glob: glob should match dangling symlinks
309 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
310 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
312 [5997] math: Very slow execution of sinf function
313 [10580] localedata: hr_HR: updated locale
314 [10871] locale: 'mon' array should contain both nominative and genitive
316 [12349] localedata: eu_ES: incorrect thousands separator
317 [13605] localedata: shn_MM: new Shan locale
318 [13805] localedata: ru_RU: currency should use ',' as radix point
319 [13953] localedata: km_KH: locale update
320 [13994] localedata: mjw_IN: new locale
321 [14121] build: make writes .mo files in po directory
322 [14333] libc: Fix the race between atexit() and exit()
323 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
324 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
326 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
327 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
328 width Latin characters
329 [15332] localedata: es_CU: locale update
330 [15436] stdio: Don't close or flush stdio streams on abort
331 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
333 [16148] localedata: ca_ES: incorrect thousands separator
334 [16750] dynamic-link: ldd should not try to execute the binaries
336 [16777] localedata: pl_PL: incorrect thousands separator in locale
337 [16905] localedata: hanzi: new collation
338 [17563] localedata: cmn_TW: add hanzi collation
339 [17750] localedata: wrong collation order of diacritics in most locales
340 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
342 [17956] build: Build fails on missing definitions from header file
343 nss/nss.h when Mozilla NSS is used for cryptography
344 [18203] libc: realpath() does not handle unreachable paths correctly
345 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
346 [18812] localedata: kab_DZ: new Kabyle Algeria locale
347 [18822] libc: Internal functions are called via PLT
348 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
349 [19170] libc: __gmon_start__ defined in hppa in crtn.S
350 [19574] libc: glibc should support building static PIE binaries
351 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
353 [19971] glob: glob: Do not skip entries with zero d_ino values
354 [19982] localedata: fr.po: spelling mistake for error code EXDEV
355 [20008] localedata: km_KH: convert to translit_neutral
356 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
357 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
358 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
359 [20482] localedata: de_CH: abbreviated weekdays should be two letters
360 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
362 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
363 they're set, wrong errors returned by gaih_inet when lookup functions
365 [20756] localedata: [PATCH] Use Unicode wise thousands separator
366 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
368 [20952] localedata: yuw_PG: new locale
369 [21084] localedata: charmaps/IBM858: new codepage
370 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
371 [21242] libc: assert gives pedantic warning in old gcc versions
372 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
373 __regcall calling convention
374 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
375 [21326] libc: C99 functions are not declared for C++11 and later if
376 _GNU_SOURCE is not predefined by g++
377 [21457] libc: sys/ucontext.h namespace
378 [21530] libc: tmpfile() should be implemented using O_TMPFILE
379 [21660] math: GCC fails to compile a formula with tgmath.h
380 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
381 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
382 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
383 [21684] math: tgmath.h handling of complex integers
384 [21685] math: tgmath.h handling of bit-fields
385 [21686] math: tgmath.h handling of __int128
386 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
388 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
390 [21750] localedata: column width of characters incompatible with classical
392 [21754] malloc: malloc: Perform as little work as possible after heap
393 consistency check failures
394 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
395 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
396 [21791] string: Unused XXX_chk_XXX functions in libc.a
397 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
399 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
401 [21845] localedata: Added new Locale bho_NP
402 [21853] localedata: Fix abday Which looks same as day in zh_SG
403 [21854] localedata: Added New Locale en_SC
404 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
405 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
406 _dl_runtime_resolve_avx_slow
407 [21885] network: getaddrinfo: gethosts does not release resolver context
408 on memory allocation failure
409 [21899] libc: XPG4.2 sigaction namespace
410 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
412 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
414 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
415 errno value, causing getaddrinfo to fail
416 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
417 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
418 instead of EAI_NODATA
419 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
420 constant PTRACE_SEIZE_DEVEL
421 [21930] math: C-only gcc builtins used in <math.h> isinf
422 [21932] network: Unpaired __resolv_context_get in generic get*_r
424 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
425 [21944] libc: sigval namespace
426 [21951] localedata: Update hanzi collation by stroke
427 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
428 sysdeps/x86_64/fpu/e_expf.S
429 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
430 allocation in syscall loops)
431 [21959] localedata: Fix Country name for xh_ZA
432 [21960] localedata: Fix abmon for bem_ZM
433 [21966] math: AVX2 mathvec functions use FMA without checking
434 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
435 [21971] localedata: Added New Locale for mfe_MU
436 [21972] libc: assert macro requires operator== (int) for its argument type
437 [21973] math: [sparc] libm missing sqrtl compat symbol
438 [21974] libc: Remove __bb_init_func and __bb_exit_func
439 [21982] string: stratcliff.c: error: assuming signed overflow does not
441 [21986] stdio: __guess_grouping is called incorrectly
442 [21987] math: [sparc32] wrong bits/long-double.h installed
443 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
445 [22022] localedata: Missing country_name for mni_IN
446 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
448 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
450 [22026] locale: iconv_open: heap overflow on gconv_init failure
451 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
452 [22035] math: [m68k] bits/math-inline.h macro namespace
453 [22038] localedata: Fix abbreviated weeks and months for Somali
454 [22044] localedata: Remove redundant data for Limburgish Language
455 [22050] malloc: Linking with -lmcheck does not hook
456 __malloc_initialize_hook correctly
457 [22051] libc: zero terminator in the middle of glibc's .eh_frame
458 [22052] malloc: malloc failed to compile with GCC 7 and -O3
459 [22070] localedata: charmaps/UTF-8: wcwidth for
460 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
461 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
462 Jungseong and Jongseong) should be 0
463 [22078] nss: nss_files performance issue in multi mode
464 [22082] math: bits/math-finite.h exp10 condition
465 [22086] libc: pcprofiledump incorrect cross-endian condition
466 [22093] dynamic-link: ld.so no longer searches in .../x86_64
467 [22095] network: Name server address allocation memory leak in resolv.conf
469 [22096] network: __resolv_conf_attach can incorrectly free passed conf
471 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
472 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
474 [22111] malloc: malloc: per thread cache is not returned when thread exits
475 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
476 [22134] libc: [linux] implement fexecve with execveat
477 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
479 [22145] libc: ttyname() gives up too early in the face of namespaces
480 [22146] math: C++ build issue with float128 on x86_64
481 [22153] nptl: nptl: save error code before process termination
482 [22156] libc: [hppa,ia64,microblaze] Executable stack default
483 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
484 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
485 [22165] libc: [hppa] Text relocations in libc.so
486 [22180] libc: destructor registered via __cxa_atexit is called twice
487 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
488 symlinks") cause "make" segfaults
489 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
491 [22207] libc: FAIL: stdlib/test-atexit-race
492 [22225] math: nearbyint arithmetic moved before feholdexcept
493 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
494 [22235] math: iscanonical in C++ and float128
495 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
496 the svid compat wrapper
497 [22244] math: ynf and yn are wrong without the svid compat wrapper
498 [22273] libc: Improper assert in Linux posix_spawn implementation
499 [22284] libc: -pg -pie doesn't work
500 [22292] locale: localedef exits with error 4 when it should be error 1
501 [22294] locale: Allow "" for int_currency_symbol definition in locales.
502 [22295] locale: Don't warn on non-symbolic characters in locale sources in
504 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
505 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
507 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
508 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
509 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
510 [22322] libc: [mips64] wrong bits/long-double.h installed
511 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
512 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
514 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
515 [22343] malloc: Integer overflow in posix_memalign
516 [22347] libc: getrandom() returns the number of bytes that were copied to
517 the buffer even though the comments say "Return 0 on success and -1 on
519 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
520 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
522 [22370] dynamic-link: Incorrect note padding check
523 [22375] libc: malloc returns pointer from tcache_get when should return
524 NULL (CVE-2017-17426)
525 [22377] math: iseqsig, float128 and C++
526 [22382] localedata: Error in tpi_PG locale
527 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
528 inside the ASCII printable range
529 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
531 [22403] localedata: Slash needs escaping in some locales
532 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
533 [22409] network: res_hnok does not accept some host names used on the
535 [22412] network: res_dnok, res_hnok should perform syntax checks
536 [22413] network: ns_name_pton ignores syntactically invalid trailing
538 [22415] stdio: setvbuf can lead to invalid free/segfault
539 [22432] build: Non-deterministic build
540 [22439] malloc: malloc_info should compute summary statistics for all sub-
542 [22442] network: if_nametoindex could report index for the wrong
544 [22446] build: aliasing violation calling readlink in handle_request
545 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
546 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
547 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
548 __nscd_hash/__nss_hash
549 [22463] network: p_secstodate overflow handling
550 [22469] localedata: pl_PL LC_COLLATE does not use i18n
551 [22478] libc: sigwait can fail with EINTR
552 [22505] libc: ldconfig processes include directive in locale-specific
554 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
555 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
556 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
557 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
558 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
559 [22534] localedata: Collation rules for Serbian and Bosnian should be the
561 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
562 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
563 [22577] libc: missing newline after "cannot allocate TLS data structures
565 [22588] manual: manual/conf.texi: missing underscore in front of
567 [22593] math: nextafter and nexttoward are declared with const attribute
568 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
569 [22603] string: ia64 memchr overflows internal pointer check
570 [22605] libc: SH clone does not set the exit code correctly
571 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
573 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
574 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
575 modified in case of success
576 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
577 [22615] manual: manual: ambiguous wording about errno value in case of
579 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
581 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
582 binaries or if /proc is not mounted (CVE-2017-16997)
583 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
584 [22630] build: $(no-pie-ldflag) is no longer effective
585 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
586 [22635] nptl: pthread_self returns NULL before libpthread is loaded
587 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
588 [22637] nptl: guard size is subtracted from thread stack size instead of
590 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
591 [22657] localedata: hu_HU: Avoid double space in date
592 [22660] math: fmax, fmin sNaN handling on alpha
593 [22664] libc: New warning of GCC8
594 [22665] math: alpha: ceil and floor raise inexact exceptions
595 [22666] math: alpha: trunc raise inexact exceptions
596 [22667] libc: makecontext lacks stack alignment on i386
597 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
598 [22679] libc: getcwd(3) can succeed without returning an absolute path
600 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
602 [22687] math: [powerpc-nofpu] complex long double functions spurious
604 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
605 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
606 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
608 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
609 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
611 [22701] nis: Incomplete removal of libnsl
612 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
613 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
614 [22715] dynamic-link: FAIL: elf/tst-audit10
615 [22719] libc: Backtrace tests fail on hppa
616 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
617 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
618 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
619 before the first call to crypt_r () in crypt/badsalttest.c
626 * A per-thread cache has been added to malloc. Access to the cache requires
627 no locks and therefore significantly accelerates the fast path to allocate
628 and free small amounts of memory. Refilling an empty cache requires locking
629 the underlying arena. Performance measurements show significant gains in a
630 wide variety of user workloads. Workloads were captured using a special
631 instrumented malloc and analyzed with a malloc simulator. Contributed by
632 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
634 * Unicode 10.0.0 Support: Character encoding, character type info, and
635 transliteration tables are all updated to Unicode 10.0.0, using
636 generator scripts contributed by Mike FABIAN (Red Hat).
637 These updates cause user visible changes, especially the changes in
638 wcwidth for many emoji characters cause problems when emoji sequences
639 are rendered with pango, see for example:
640 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
642 * Collation of Hungarian has been overhauled and is now consistent with "The
643 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
646 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
648 - The GNU C Library will now detect when /etc/resolv.conf has been
649 modified and reload the changed configuration. The new resolver option
650 “no-reload” (RES_NORELOAD) disables this behavior.
652 - The GNU C Library now supports an arbitrary number of search domains
653 (configured using the “search” directive in /etc/resolv.conf);
654 previously, there was a hard limit of six domains. For backward
655 compatibility, applications that directly modify the ‘_res’ global
656 object are still limited to six search domains.
658 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
659 Library will now randomly pick a name server from the configuration as a
660 starting point. (Previously, the second name server was always used.)
662 * The tunables feature is now enabled by default. This allows users to tweak
663 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
665 * New function reallocarray, which resizes an allocated block (like realloc)
666 to the product of two sizes, with a guaranteed clean failure upon integer
667 overflow in the multiplication. Originally from OpenBSD, contributed by
668 Dennis Wölfing and Rüdiger Sonderfeld.
670 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
671 These are extended versions of preadv and pwritev, respectively, taking an
672 additional flags argument. The set of supported flags depends on the
673 running kernel; full support currently requires kernel 4.7 or later.
675 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
676 create a new session ID for the spawned process. This feature is
677 scheduled to be added to the next major revision of POSIX; for the time
678 being, it is available under _GNU_SOURCE.
680 * errno.h is now safe to use from C-preprocessed assembly language on all
681 supported operating systems. In this context, it will only define the
682 Exxxx constants, as preprocessor macros expanding to integer literals.
684 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
685 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
686 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
687 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
689 To compile programs that use this feature, the compiler must support
690 128-bit floating point with the type name _Float128 (as defined by TS
691 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
692 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
693 must be defined to make the new interfaces visible.
695 The new functions and macros correspond to those present for other
696 floating-point types (except for a few obsolescent interfaces not
697 supported for the new type), with F128 or f128 suffixes; for example,
698 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
699 printf or scanf formats for the new type; the strfromf128 and strtof128
700 interfaces should be used instead.
702 Deprecated and removed features, and other changes affecting compatibility:
704 * The synchronization that pthread_spin_unlock performs has been changed to
705 now be equivalent to a C11 atomic store with release memory order to the
706 spin lock's memory location. Previously, several (but not all)
707 architectures used stronger synchronization (e.g., containing what is
708 often called a full barrier). This change can improve performance, but
709 may affect odd fringe uses of spin locks that depend on the previous
710 behavior (e.g., using spin locks as atomic variables to try to implement
711 Dekker's mutual exclusion algorithm).
713 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
716 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
717 will only be built and installed when the GNU C Library is configured with
718 --enable-obsolete-rpc. This allows alternative RPC implementations, such
719 as TIRPC or rpcsvc-proto, to be used.
721 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
722 libnss_compat, are deprecated, and will not be built or installed by
725 The NIS(+) support library, libnsl, is also deprecated. By default, a
726 compatibility shared library will be built and installed, but not headers
727 or development libraries. Only a few NIS-related programs require this
728 library. (In particular, the GNU C Library has never required programs
729 that use 'gethostbyname' to be linked with libnsl.)
731 Replacement implementations based on TIRPC, which additionally support
732 IPv6, are available from <https://github.com/thkukuk/>. The configure
733 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
734 name service modules, to be built and installed.
736 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
737 support is enabled, the configured recursive resolver must support EDNS.
738 (Responding to EDNS-enabled queries with responses which are not
739 EDNS-enabled is fine, but FORMERR responses are not.)
741 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
742 servers have not supported this opcode for a long time.
744 * The _res_opcodes variable has been removed from libresolv. It had been
745 exported by accident.
747 * <string.h> no longer includes inline versions of any string functions,
748 as this kind of optimization is better done by the compiler. The macros
749 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
751 * The nonstandard header <xlocale.h> has been removed. Most programs should
752 use <locale.h> instead. If you have a specific need for the definition of
753 locale_t with no other declarations, please contact
754 libc-alpha@sourceware.org and explain.
756 * The obsolete header <sys/ultrasound.h> has been removed.
758 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
760 * The obsolete function cfree has been removed. Applications should use
763 * The stack_t type no longer has the name struct sigaltstack. This changes
764 the C++ name mangling for interfaces involving this type.
766 * The ucontext_t type no longer has the name struct ucontext. This changes
767 the C++ name mangling for interfaces involving this type.
769 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
770 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
771 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
772 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
773 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
774 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
775 fpregset_t type no longer has the name struct fpu. This changes the C++
776 name mangling for interfaces involving those types.
778 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
779 synced with the kernel:
781 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
782 are not supported on this architecture and have been removed.
784 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
785 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
786 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
788 Programs that assume the GET/SETREGS ptrace requests are universally
789 available will now fail to build, instead of malfunctioning at runtime.
791 Changes to build and runtime requirements:
793 * Linux kernel 3.2 or later is required at runtime, on all architectures
794 supported by that kernel. (This is a change from version 2.25 only for
797 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
799 * On most architectures, GCC 4.9 or later is required to build the GNU C
800 Library. On powerpc64le, GCC 6.2 or later is required.
802 Older GCC versions and non-GNU compilers are still supported when
803 compiling programs that use the GNU C Library. (We do not know exactly
804 how old, and some GNU extensions to C may be _de facto_ required. If you
805 are interested in helping us make this statement less vague, please
806 contact libc-alpha@sourceware.org.)
808 Security related changes:
810 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
811 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
813 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
814 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
816 * Avoid printing a backtrace from the __stack_chk_fail function since it is
817 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
820 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
821 fixed (CVE-2017-12133).
823 The following bugs are resolved with this release:
825 [984] network: Respond to changed resolv.conf in gethostbyname
826 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
827 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
829 [12189] libc: __stack_chk_fail should not attempt a backtrace
831 [14096] time: Race condition on timezone/tst-timezone.out
832 [14172] localedata: az_IR: new locale
833 [14995] build: glibc fails to build if gold is the default linker, even if
835 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
836 [16637] network: inet_pton function is accepting IPv6 with bad format
837 [16640] string: string/strtok.c: undefined behaviour inconsistent between
838 x86 and other generic code
839 [16875] localedata: ko_KR: fix lang_name
840 [17225] localedata: ar_SY: localized month names for May and June are
842 [17297] localedata: da_DK: wrong date_fmt string
843 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
844 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
845 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
846 [19066] localedata: ar_SA abbreviated day and month names are in English
847 [19569] network: resolv: Support an arbitrary number of search domains
848 [19570] network: Implement random DNS server selection in the stub
850 [19838] locale: localedef fails on PA-RISC
851 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
852 order of 0D36 and 0D37
853 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
855 [20098] libc: FAIL: debug/backtrace-tst on hppa
856 [20257] network: sunrpc: clntudp_call does not enforce timeout when
858 [20275] localedata: locale day/abday/mon/abmon should not have trailing
860 [20313] localedata: Update locale data to Unicode 9.0
861 [20424] manual: Document how to provide a malloc replacement
862 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
864 [20686] locale: Add el_GR@euro to SUPPORTED.
865 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
866 failures consistently
867 [21015] dynamic-link: Document and fix --enable-bind-now
868 [21016] nptl: pthread_cond support is broken on hppa
869 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
870 frame-pointer on i386
871 [21049] libc: segfault in longjmp_chk() due to clobbered processor
873 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
874 [21088] libc: Build fails with --enable-static-nss
875 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
876 generic c code is used
877 [21109] libc: Tunables broken on big-endian
878 [21112] math: powf has large ulp errors with base close to 1 and exponent
880 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
882 [21120] malloc: glibc malloc is incompatible with GCC 7
883 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
885 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
886 (-0.0) when linking with -lieee
887 [21171] math: log10, log2 and lgamma return incorrect results
888 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
889 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
890 [21207] localedata: ce_RU: update weekdays from CLDR
891 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
892 [21217] localedata: Update months from CLDR-31
893 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
894 [21243] libc: support_delete_temp_file should issue warning for failed
896 [21244] libc: support resolv_test_start() socket fd close should be
898 [21253] libc: localedef randomly segfaults when using -fstack-check due to
899 new posix_spawn implementation
900 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
901 leads to lower CPU frequency
902 [21259] libc: [alpha] termios.h missing IXANY for POSIX
903 [21261] libc: [sparc64] bits/setjmp.h namespace
904 [21267] network: [mips] bits/socket.h IOC* namespace
905 [21268] libc: [alpha] termios.h NL2, NL3 namespace
906 [21270] libc: mmap64 silently truncates large offset values
907 [21275] libc: posix_spawn always crashes on ia64 now
908 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
909 [21280] math: [powerpc] logbl for POWER7 return incorrect results
910 [21289] libc: Incorrect declaration for 32-bit platforms with
911 _FILE_OFFSET_BITS=64 causes build error
912 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
913 not support gethostbyname4_r
914 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
916 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
918 [21340] libc: Support POSIX_SPAWN_SETSID
919 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
920 [21359] network: ns_name_pack needs additional byte in destination buffer
921 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
922 against fragmentation attacks (CVE-2017-12132)
923 [21369] network: resolv: Remove EDNS fallback
924 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
926 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
927 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
928 [21393] stdio: Missing dup3 error check in freopen, freopen64
929 [21396] libc: Use AVX2 memcpy/memset on Skylake server
930 [21399] localedata: Bad description for U00EC in
931 localedata/charmaps/CP1254
932 [21411] malloc: realloc documentation error
933 [21426] network: sys/socket.h uio.h namespace
934 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
935 [21445] libc: signal.h bsd_signal namespace
936 [21455] network: Network headers stdint.h namespace
937 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
939 [21475] network: resolv: Overlong search path is truncated mid-label
940 [21511] libc: sigstack namespace
941 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
942 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
943 bits/syscall.h: No such file or directory
944 [21517] libc: struct sigaltstack namespace
945 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
946 [21533] localedata: Update locale data to Unicode 10.0
948 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
949 junk at end of line, first unrecognized character is `@'
950 [21538] libc: SIG_HOLD missing for XPG4
951 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
952 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
953 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
954 [21543] libc: sigevent namespace
955 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
956 size for general purpose registers in mcontext_t structure
957 [21550] libc: sigwait namespace
958 [21552] libc: XPG4 bsd_signal namespace
959 [21554] libc: sigpause namespace
960 [21560] libc: sys/wait.h signal.h namespace
961 [21561] libc: waitid namespace
962 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
963 '/usr/include/stdlib.h': Permission denied
964 [21575] libc: sys/wait.h missing struct rusage definition
965 [21584] libc: sigaltstack etc namespace
966 [21597] libc: siginterrupt namespace
967 [21607] math: hppa: FAIL: math/test-tgmath
968 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
970 [21622] libc: [tile] missing SA_* for POSIX.1:2008
971 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
972 stack and heap (CVE-2017-1000366)
973 [21625] libc: wait3 namespace
974 [21654] nss: Incorrect pointer alignment in NSS group merge result
976 [21657] network: Parse interface zone id for node-local multicast
977 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
979 [21666] libc: .symver is used on common symbol
980 [21668] network: resolv: res_init cross-thread broadcast introduces race
982 [21687] math: tgmath.h totalorder, totalordermag return type
983 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
985 [21696] libc: Incorrect assumption of of __cpu_mask in
986 posix/sched_cpucount.c
987 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
988 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
990 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
991 leading to relocation crash
992 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
993 `resolv_conf_matches (resp, conf)' failed.
994 [21710] localedata: Added Samoan language locale for Samoa
995 [21711] localedata: Pashto yesstr/nostr locale are missing
996 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
997 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
998 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
999 [21723] localedata: yesstr/nostr missing for Chinese language locale
1000 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1002 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1004 [21728] localedata: New Locale for Tongan language
1005 [21729] localedata: incorrect LC_NAME fields for hi_IN
1006 [21733] localedata: yesstr and nostr are missing for zh_HK
1007 [21734] localedata: Missing yesstr and nostr are for kw_GB
1008 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1009 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1010 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1011 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1013 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1015 [21749] localedata: Wrong abbreviated day name (“abday”) for
1017 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1018 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1019 [21759] localedata: missing yesstr and nostr for Tigrinya
1020 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1021 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1022 [21767] localedata: Missing Bislama locales
1023 [21768] localedata: Missing yesstr and nostr for aa_ET
1024 [21770] localedata: Missing Field in li_NL
1025 [21778] nptl: Robust mutex may deadlock
1026 [21779] libc: MicroBlaze segfaults when loading libpthread
1027 [21783] localedata: Fix int_select international_call_prefixes
1028 [21784] localedata: Inconsistency in country_isbn
1029 [21788] localedata: Missing Country Postal Abbreviations
1030 [21794] localedata: Added-country_isbn-for-Italy
1031 [21795] localedata: Add/Fix country_isbn for France
1032 [21796] localedata: Added country_isbn for Republic of Korea
1033 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1034 [21799] localedata: Added int_select international_call_prefixes
1035 [21801] localedata: Added int_select international_call_prefixes
1036 [21804] nptl: Double semicolon in thread-shared-types.h
1037 [21807] localedata: LC_ADDRESS fix for pap_CW
1038 [21808] localedata: Fix LC_ADDRESS for pap_AW
1039 [21821] localedata: Added country_name in mai_IN
1040 [21822] localedata: Fix LC_TIME for mai_IN
1041 [21823] localedata: missing yesstr, nostr for sa_IN
1042 [21825] localedata: Fix name_mrs for mag_IN
1043 [21828] localedata: 2.26 changelog should mention user visible changes
1045 [21835] localedata: Added Maithili language locale for Nepal
1046 [21838] localedata: Removed redundant data for the_NP
1047 [21839] localedata: Fix LC_MONETARY for ta_LK
1048 [21844] localedata: Fix Latin characters and Months Sequence.
1049 [21848] localedata: Fix mai_NP Title Name
1054 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1055 24731-2:2010, is supported to enable declarations of functions from that
1056 TR. Note that not all functions from that TR are supported by the GNU C
1059 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1060 18661-1:2014, is supported to enable declarations of functions and macros
1061 from that TS. Note that not all features from that TS are supported by
1064 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1065 18661-4:2015, is supported to enable declarations of functions and macros
1066 from that TS. Note that most features from that TS are not supported by
1069 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1070 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1071 Since the GNU C Library defaults to a much newer revision of POSIX, this
1072 will only affect programs that specifically request an old conformance
1073 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1074 see a change in the visible declarations, but a program compiled with
1075 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1078 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1079 defined by all multithreaded code, but glibc has not required this for
1082 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1083 means that in a future release, the macros “major”, “minor”, and “makedev”
1084 will only be available from <sys/sysmacros.h>.
1086 These macros are not part of POSIX nor XSI, and their names frequently
1087 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1088 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1089 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1092 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1093 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1094 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1096 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1097 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1098 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1099 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1100 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1101 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1102 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1103 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1104 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1105 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1106 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1109 * New <math.h> features are added from TS 18661-1:2014:
1111 - Signaling NaN macros: SNANF, SNAN, SNANL.
1113 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1114 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1115 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1117 - llogb functions: the llogb, llogbf and llogbl functions, and the
1118 FP_LLOGB0 and FP_LLOGBNAN macros.
1120 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1123 - Comparison macros: iseqsig.
1125 - Classification macros: iscanonical, issubnormal, iszero.
1127 - Total order functions: totalorder, totalorderf, totalorderl,
1128 totalordermag, totalordermagf, totalordermagl.
1130 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1132 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1133 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1135 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1136 are added to libc. They convert a floating-point number into string.
1138 * Most of glibc can now be built with the stack smashing protector enabled.
1139 It is recommended to build glibc with --enable-stack-protector=strong.
1140 Implemented by Nick Alcock (Oracle).
1142 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1143 intended to be used instead of memset() to erase sensitive data after use;
1144 the compiler will not optimize out calls to explicit_bzero even if they
1145 are "unnecessary" (in the sense that no _correct_ program can observe the
1146 effects of the memory clear).
1148 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1149 to float instead of double. This does not affect the ABI of any libraries
1150 that are part of the GNU C Library, but may affect the ABI of other
1151 libraries that use this type in their interfaces.
1153 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1154 float_t and double_t types are now defined to long double instead of float
1155 and double. These options are not the default, and this does not affect
1156 the ABI of any libraries that are part of the GNU C Library, but it may
1157 affect the ABI of other libraries that use this type in their interfaces,
1158 if they are compiled or used with those options.
1160 * The getentropy and getrandom functions, and the <sys/random.h> header file
1163 * The buffer size for byte-oriented stdio streams is now limited to 8192
1164 bytes by default. Previously, on Linux, the default buffer size on most
1165 file systems was 4096 bytes (and thus remains unchanged), except on
1166 network file systems, where the buffer size was unpredictable and could be
1167 as large as several megabytes.
1169 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1170 for the Linux quota interface which predates kernel version 2.4.22 has
1173 * The malloc_get_state and malloc_set_state functions have been removed.
1174 Already-existing binaries that dynamically link to these functions will
1175 get a hidden implementation in which malloc_get_state is a stub. As far
1176 as we know, these functions are used only by GNU Emacs and this change
1177 will not adversely affect already-built Emacs executables. Any undumped
1178 Emacs executables, which normally exist only during an Emacs build, should
1179 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1181 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1182 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1183 “no-ip6-dotint” had already been the default, and support for the
1184 “ip6-dotint” option was removed from the Internet in 2006.
1186 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1187 flag from <resolv.h> have been removed. The option relied on a
1188 backwards-incompatible DNS extension which was never deployed on the
1191 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1192 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1193 They were already unimplemented.
1195 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1196 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1197 removed again from the IETF name lookup interface specification in RFC
1198 2553. Applications should use getaddrinfo instead.
1200 * DNSSEC-related declarations and definitions have been removed from the
1201 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1202 decode the data part of DNSSEC record types. Previous versions of glibc
1203 only implemented minimal support for the previous version of DNSSEC, which
1204 is incompatible with the currently deployed version.
1206 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1207 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1208 <arpa/nameser.h> header file because the distinction between RR types and
1209 meta-RR types is not officially standardized, subject to revision, and
1210 thus not suitable for encoding in a macro.
1212 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1213 and rhook members of the res_state type in <resolv.h> have been removed.
1214 The glibc stub resolver did not support these hooks, but the header file
1215 did not reflect that.
1217 * For multi-arch support it is recommended to use a GCC which has
1218 been built with support for GNU indirect functions. This ensures
1219 that correct debugging information is generated for functions
1220 selected by IFUNC resolvers. This support can either be enabled by
1221 configuring GCC with '--enable-gnu-indirect-function', or by
1222 enabling it by default by setting 'default_gnu_indirect_function'
1223 variable for a particular architecture in the GCC source file
1226 * GDB pretty printers have been added for mutex and condition variable
1227 structures in POSIX Threads. When installed and loaded in gdb these pretty
1228 printers show various pthread variables in human-readable form when read
1229 using the 'print' or 'display' commands in gdb.
1231 * Tunables feature added to allow tweaking of the runtime for an application
1232 program. This feature can be enabled with the '--enable-tunables' configure
1233 flag. The GNU C Library manual has details on usage and README.tunables has
1234 instructions on adding new tunables to the library.
1236 * A new version of condition variables functions have been implemented in
1237 the NPTL implementation of POSIX Threads to provide stronger ordering
1240 * A new version of pthread_rwlock functions have been implemented to use a more
1241 scalable algorithm primarily through not using a critical section anymore to
1244 Security related changes:
1246 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1247 have been created with makecontext could fail to terminate due to a
1248 missing .cantunwind annotation. This has been observed to lead to a hang
1249 (denial of service) in some Go applications compiled with gccgo. Reported
1250 by Andreas Schwab. (CVE-2016-6323)
1252 * The DNS stub resolver functions would crash due to a NULL pointer
1253 dereference when processing a query with a valid DNS question type which
1254 was used internally in the implementation. The stub resolver now uses a
1255 question type which is outside the range of valid question type values.
1258 The following bugs are resolved with this release:
1260 [4099] stdio: Overly agressive caching by stream i/o functions.
1261 [7065] build: Support building glibc with -fstack-protector or -fstack-
1263 [9842] localedata: en_CA: incorrect date format
1264 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1265 before it started waiting
1266 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1267 [15765] nptl: sem_open is wrongly a cancellation point
1268 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1269 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1271 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1272 library linked with pthread
1273 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1274 "FIXME: Ingo" issue)
1275 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1276 [17252] libc: getrandom and getentropy syscall
1277 [17730] malloc: thread-local storage is sometimes improperly free()'d
1278 after being __libc_memalign()'d
1279 [18241] stdio: failed fseek on memstream does not set errno and can fail
1281 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1283 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1284 PTHREAD_PRIO_INHERIT mutex on ARM
1285 [18784] network: res_query and related function crash for special record
1286 type queries (CVE-2015-5180)
1287 [19380] math: strtod does not raise "inexact"
1288 [19387] string: Integer overflow in memchr
1289 [19390] string: Integer overflow in strncat
1290 [19398] build: linknamespace tests fail with massively parallel build
1291 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1293 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1294 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1295 [19514] libc: [PATCH] Fix spelling errors in spelling
1296 "implement"/"implementation" in several places
1297 [19582] network: Deprecate RES_USE_INET6
1298 [19673] manual: clog10 docs appear to be erroneous
1299 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1300 causes a segmentation fault
1301 [19826] libc: invalid pointer returned from __tls_get_addr with static
1303 [20016] network: resolv: Remove hooks support from the API
1304 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1306 [20033] math: [x86_64] vectorized math function don't call the __finite
1308 [20116] nptl: use after free in pthread_create
1309 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1310 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1312 [20311] nptl: please install proc_service.h
1313 [20366] libc: Compilation errors in installed headers in strict-compliance
1315 [20370] malloc: malloc: Arena free list management is still racy
1316 (incorrect fix in bug 19243)
1317 [20386] libc: assert (X = 0) does not result in GCC warning
1318 [20432] malloc: malloc: Minimize interface required for interposition
1319 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1320 in _Unwind_Backtrace (CVE-2016-6323)
1321 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1322 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1324 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1325 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1326 [20477] network: resolv: incorrect double-checked locking related to
1328 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1330 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1332 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1333 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1334 cause transition penalty
1335 [20517] math: sparcv9 missing fdiml compat symbol
1336 [20524] manual: strverscmp is inconsistent
1337 [20525] libc: <sys/quota.h> should be based on kernel headers
1338 [20539] math: GCC 7 -static -lm fails to link at -O3
1339 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1341 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1343 [20591] network: Remove obsolete DNSSEC support
1344 [20592] network: DNS resource record type classification macros in
1345 <arpa/nameser.h> are incorrect
1346 [20593] network: Update DNS RR type definitions
1347 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1348 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1350 [20629] network: libresolv: Remove support for bitstring labels
1352 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1353 assertion: "maxidx >= 2"
1354 [20660] math: [arm] Use VSQRT
1355 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1356 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1357 [20689] libc: Test for FMA should also check for AVX.
1358 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1359 [20715] math: iszero macro breaks existing code
1360 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1361 [20728] libc: powerpc: Missing TOC stub in clone
1362 [20729] build: glibc-2.24 fails to build for i486 with -Os
1363 [20750] build: Build fails with default PIE enabled gcc-6.x
1364 [20768] math: [alpha] sqrt fegetenv namespace
1365 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1367 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1368 __FLT_EVAL_METHOD__ is 2
1369 [20790] network: rpcgen buffer overrun in get_prog_declaration
1370 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1371 [20829] libc: crypt snprintf namespace
1372 [20847] libc: tst-vfork3 failure
1373 [20855] math: Default bits/mathdef.h has inappropriate float_t
1374 [20859] math: [sh4] FP_ILOGB0 invalid
1375 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1376 U+20AC), not same as GBK
1377 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1378 [20916] math: pow handling of sNaN arguments
1379 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1380 [20919] math: Bad pow (qNaN, 0) result with -lieee
1381 [20940] math: hypot sNaN handling
1382 [20947] math: fmax, fmin sNaN handling
1383 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1384 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1386 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1387 [20973] nptl: robust mutexes: Lost wake-ups
1388 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1389 [20978] nis: strlen on null pointer in nss_nisplus
1390 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1392 [21014] string: i686 memchr overflows internal pointer check
1393 [21019] libc: [mips] n32 lseek incorrect on overflow
1394 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1395 [21026] libc: [MIPS] readahead syscall is broken on n64
1396 [21028] math: Fallback fesetexceptflag should always succeed
1397 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1398 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1400 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1401 [21061] librt: [microblaze] librt lost clock_* exports
1402 [21073] libc: tunables: insecure environment variables passed to
1403 subprocesses with AT_SECURE
1404 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1408 * The minimum Linux kernel version that this version of the GNU C Library
1409 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1410 kernel version 2.6.32 or later suffices (on architectures that already
1411 required kernel versions more recent than 3.2, those requirements remain
1412 unchanged). Linux 3.2 or later kernel headers are required on all
1415 * The pap_AN locale has been deleted. This has been deprecated for a long
1416 time. It has been replaced by pap_AW & pap_CW, both of which have long
1417 been included in previous releases.
1419 * The readdir_r and readdir64_r functions have been deprecated. It is
1420 recommended to use readdir and readdir64 instead.
1422 * The type “union wait” has been removed. It was deprecated in the early
1423 1990s and never part of POSIX. Application code should use the int type
1424 instead of “union wait”.
1426 * A new NSS action is added to facilitate large distributed system
1427 administration. The action, MERGE, allows remote user stores like LDAP
1428 to be merged into local user stores like /etc/groups in order to provide
1429 easy to use, updated, and managed sets of merged credentials. The new
1430 action can be used by configuring it in /etc/nsswitch.conf:
1431 group: files [SUCCESS=merge] nis
1432 Implemented by Stephen Gallagher (Red Hat).
1434 * The deprecated __malloc_initialize_hook variable has been removed from the
1437 * The long unused localedef --old-style option has been removed. It hasn't
1438 done anything in over 16 years. Scripts using this option can safely
1441 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1442 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1443 return the next representable value in the direction of positive infinity
1444 and the nextdown functions return the next representable value in the
1445 direction of negative infinity. These are currently enabled as GNU
1448 Security related changes:
1450 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1451 could result in a stack overflow when getnetbyname was called with an
1452 overly long name. (CVE-2016-3075)
1454 * Previously, getaddrinfo copied large amounts of address data to the stack,
1455 even after the fix for CVE-2013-4458 has been applied, potentially
1456 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1457 instead. Reported by Michael Petlan. (CVE-2016-3706)
1459 * The glob function suffered from a stack-based buffer overflow when it was
1460 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1461 Reported by Alexander Cherepanov. (CVE-2016-1234)
1463 * The Sun RPC UDP client could exhaust all available stack space when
1464 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1465 alloca plugin for GCC. (CVE-2016-4429)
1467 * The IPv6 name server management code in libresolv could result in a memory
1468 leak for each thread which is created, performs a failing naming lookup,
1469 and exits. Over time, this could result in a denial of service due to
1470 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1472 The following bugs are resolved with this release:
1474 [1170] localedata: ne_NP: update Nepali locale definition file
1475 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1477 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1478 [6796] math: fdim() does not set errno on overflow
1479 [10354] libc: posix_spawn should use vfork() in more cases than presently
1480 [11213] localedata: localedata: add copyright disclaimer to locale files
1481 [12143] localedata: chr_US: new Cherokee locale
1482 [12450] localedata: sgs_LT: new locale
1483 [12676] localedata: ln_CD: new locale
1484 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1486 [13304] math: fma, fmaf, fmal produce wrong results
1487 [14259] build: --localedir arg to configure is ignored
1488 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1489 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1490 [14934] localedata: es_CL: wrong first weekday chilean locale
1491 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1493 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1495 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1496 [15368] nptl: raise() is not async-signal-safe
1497 [15479] math: ceil, floor, round and trunc raise inexact exception
1498 [15578] localedata: kk_KZ: various updates
1499 [16003] localedata: pap_AN: punt old locale
1500 [16137] localedata: iw_IL: punt old locale
1501 [16190] localedata: eo: new esperanto locale
1502 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1503 [16742] malloc: race condition: pthread_atfork() called before first
1504 malloc() results in unexpected locking behaviour/deadlocks
1505 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1507 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1508 [17565] localedata: pt_PT: wrong (work-)week start
1509 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1510 [17950] build: Build fails with -msse
1511 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1512 [18433] libc: posix_spawn does not return correctly upon failure to
1514 [18453] localedata: charmaps/IBM875: incorrect codes
1515 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1517 [18896] localedata: he_IL: improvements for currency
1518 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1519 Romanian locale data
1520 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1522 [19056] libc: Deprecate readdir_r
1523 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1525 [19198] localedata: nl_NL: small improvements for Dutch locales
1526 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1527 nameservers (CVE-2016-5417)
1528 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1530 [19400] locale: Language missing in "iso-639.def", trivial fix in
1532 [19431] malloc: Deadlock between fflush, getdelim, and fork
1533 [19505] libc: Incorrect file descriptor validity checks in
1534 posix_spawn_file_actions_add{open,close,dup2}
1535 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1536 when using RTLD_NEXT
1537 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1538 `intl/{gettextP,loadinfo}.h'
1539 [19534] libc: execle, execlp may use malloc
1540 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1541 [19573] network: res_nclose and __res_maybe_init disagree about name
1542 server initialization, breaking Hesiod
1543 [19575] localedata: Status of GB18030 tables
1544 [19581] localedata: sr_* date_fmt string contains additional newline
1545 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1547 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1548 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1549 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1550 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1552 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1554 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1556 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1557 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1558 [19633] locale: strfmon_l applies global locale to number formatting
1559 [19642] network: Memory leak in getnameinfo
1560 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1561 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1563 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1564 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1566 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1567 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1569 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1571 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1573 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1574 pointers and lengths in error-case.
1575 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1576 report errors on UTF-16 surrogates values.
1577 [19755] nscd: nscd assertion failure in gc
1578 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1579 [19759] libc: mempcpy shouldn't be inlined
1580 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1581 [19765] libc: s390 needs an optimized mempcpy
1582 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1583 NAME_MAX limit assumption (CVE-2016-1234)
1584 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1585 [19787] network: Missing and incorrect truncation checks in getnameinfo
1586 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1588 [19791] network: Assertion failure in res_query.c with un-connectable name
1590 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1591 [19822] math: libm.so install clobbers old version
1592 [19825] network: resolv: send_vc can return uninitialized data in second
1593 response to getaddrinfo
1594 [19830] network: nss_dns: should check RDATA length against buffer length
1595 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1596 confronted with A/AAAA records of invalid size
1597 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1598 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1599 [19853] stdio: Printing IBM long double in decimal with high precision is
1601 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1602 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1603 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1605 [19865] network: Assertion failure or memory leak in
1606 _nss_dns_getcanonname_r
1607 [19868] network: nss_dns: netent code does not skip over non-PTR records
1608 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1610 [19881] string: Improve x86-64 memset
1611 [19907] string: Incorrect memcpy tests
1612 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1614 [19925] libc: termios.h XCASE namespace
1615 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1616 [19929] libc: limits.h NL_NMAX namespace
1617 [19931] stdio: Memory leak in vfprintf
1618 [19957] libc: clone(CLONE_VM) access invalid parent memory
1619 [19963] localedata: en_IL: New locale
1620 [19989] stdio: stdio.h cuserid namespace
1621 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1623 [19996] locale: langinfo.h nl_langinfo_l namespace
1624 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1625 [20010] network: getaddrinfo: Stack overflow in hostent translation
1627 [20012] stdio: libio: fmemopen append mode failure
1628 [20014] stdio: stdio.h namespace for pre-threads POSIX
1629 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1630 [20023] libc: fcntl.h timespec namespace
1631 [20024] math: [x86_64] vectorized sincos trashes the stack
1632 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1633 [20041] time: sys/time.h timespec namespace
1634 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1635 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1636 [20051] libc: ttyslot in wrong header under wrong conditions
1637 [20054] libc: gethostname not declared for XPG4
1638 [20055] libc: termios.h missing tcgetsid for XPG4
1639 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1641 [20073] libc: sys/stat.h fchmod namespace
1642 [20074] libc: stdlib.h rand_r namespace
1643 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1644 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1646 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1647 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1649 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1650 [20119] libc: Wrong mask for processors level type from CPUID
1651 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1652 AS not supporting AVX512
1653 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1655 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1656 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1658 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1659 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1660 [20178] libc: posix_spawn{p} should not call exit
1661 [20191] stdio: libio: vtables hardening
1662 [20195] string: FMA4 detection requires CPUID execution with register
1664 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1665 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1667 [20212] math: acos (sNaN) returns sNaN
1668 [20213] math: asin (sNaN) returns sNaN
1669 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1670 [20218] math: [i386] asinhl (sNaN) returns sNaN
1671 [20219] math: [i386] atanhl (sNaN) returns sNaN
1672 [20222] stdio: fopencookie: Mangle function pointers
1673 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1674 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1675 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1677 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1678 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1679 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1680 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1681 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1682 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1683 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1684 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1685 [20240] math: modf (sNaN) returns sNaN
1686 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1687 [20250] math: frexp (sNaN) returns sNaN
1688 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1689 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1691 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1692 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1693 subscript is below array bounds [-Werror=array-bounds]
1694 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1695 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1697 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1698 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1700 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1701 "invalid" exceptions
1702 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1703 [20316] localedata: id_ID: Februari instead of Pebruari
1704 [20327] string: POWER8 strcasecmp returns incorrect result
1705 [20347] math: Failure: Test: j0_downward (0xap+0)
1706 [20348] libc: FAIL: misc/tst-preadvwritev64
1707 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1708 [20350] libc: There is no test for p{read,write}64
1709 [20357] math: Incorrect cos result for 1.5174239687223976
1710 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1714 * Unicode 8.0.0 Support: Character encoding, character type info, and
1715 transliteration tables are all updated to Unicode 8.0.0, using new
1716 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1717 These updates cause user visible changes, such as the fixes for bugs
1718 89, 16061, and 18568.
1720 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1721 kernel-internal CPU set size. This means that requests that change the
1722 CPU affinity which failed before (for example, an all-ones CPU mask) will
1723 now succeed. Applications that need to determine the effective CPU
1724 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1725 setting it because the kernel can adjust it (and the previous size check
1726 would not detect this in the majority of cases).
1728 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1729 the following new symbols are used: fts64_children, fts64_close,
1730 fts64_open, fts64_read and fts64_set.
1732 * getaddrinfo now detects certain invalid responses on an internal netlink
1733 socket. If such responses are received, an affected process will
1734 terminate with an error message of "Unexpected error <number> on netlink
1735 descriptor <number>" or "Unexpected netlink response of size <number> on
1736 descriptor <number>". The most likely cause for these errors is a
1737 multi-threaded application which erroneously closes and reuses the netlink
1738 file descriptor while it is used by getaddrinfo.
1740 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1741 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1742 unnecessary serialization of memory allocation requests across threads.
1743 The defect is now corrected. Users should see a substantial increase in
1744 the concurent throughput of allocation requests for applications which
1745 trigger this bug. Affected applications typically create and
1746 destroy threads frequently. (Bug 19048 was reported and analyzed by
1749 * There is now a --disable-timezone-tools configure option for disabling the
1750 building and installing of the timezone related utilities (zic, zdump, and
1751 tzselect). This is useful for people who build the timezone data and code
1752 independent of the GNU C Library.
1754 * The obsolete header <regexp.h> has been removed. Programs that require
1755 this header must be updated to use <regex.h> instead.
1757 * The obsolete functions bdflush, create_module, get_kernel_syms,
1758 query_module and uselib are no longer available to newly linked binaries;
1759 the header <sys/kdaemon.h> has been removed. These functions and header
1760 were specific to systems using the Linux kernel and could not usefully be
1761 used with the GNU C Library on systems with version 2.6 or later of the
1764 * Optimized string, wcsmbs and memory functions for IBM z13.
1765 Implemented by Stefan Liebler.
1767 * Newly linked programs that define a variable called signgam will no longer
1768 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1769 require signgam to be set by those functions must ensure that they use the
1770 variable provided by the GNU C Library and declared in <math.h>, without
1771 defining their own copy.
1773 * The minimum GCC version that can be used to build this version of the GNU
1774 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1775 still be used to compile programs using the GNU C Library.
1777 Security related changes:
1779 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1780 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1782 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1783 the pointer guard feature. It is always enabled. Previously,
1784 LD_POINTER_GUARD could be used to disable security hardening in binaries
1785 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1788 * An integer overflow in hcreate and hcreate_r could lead to an
1789 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1791 * The catopen function no longer has unbounded stack usage. Reported by
1792 Max. (CVE-2015-8779)
1794 * The nan, nanf and nanl functions no longer have unbounded stack usage
1795 depending on the length of the string passed as an argument to the
1796 functions. Reported by Joseph Myers. (CVE-2014-9761)
1798 * A stack-based buffer overflow was found in libresolv when invoked from
1799 libnss_dns, allowing specially crafted DNS responses to seize control
1800 of execution flow in the DNS client. The buffer overflow occurs in
1801 the functions send_dg (send datagram) and send_vc (send TCP) for the
1802 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1803 family. The use of AF_UNSPEC triggers the low-level resolver code to
1804 send out two parallel queries for A and AAAA. A mismanagement of the
1805 buffers used for those queries could result in the response of a query
1806 writing beyond the alloca allocated buffer created by
1807 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1808 the overflow. Thanks to the Google Security Team and Red Hat for
1809 reporting the security impact of this issue, and Robert Holiday of
1810 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1812 The following bugs are resolved with this release:
1814 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1815 [887] math: Math library function "logb" and "nextafter" inconsistent
1816 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1817 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1818 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1819 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1821 [4404] localedata: German translation of "Alarm clock" is misleading
1822 [6799] math: nextafter() and nexttoward() doen't set errno on
1823 overflow/underflow errors
1824 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1826 [10432] nis: _nss_nis_setnetgrent assertion failure
1827 [11460] libc: fts has no LFS support
1828 [12926] network: getaddrinfo()/make_request() may spin forever
1829 [13065] nptl: Race condition in pthread barriers
1830 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1831 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1833 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1834 [14912] libc: Rename non-installed bits/*.h headers
1835 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1836 [15367] math: Let gcc use __builtin_isinf
1837 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1838 [15421] math: lgamma wrongly sets signgam for ISO C
1839 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1840 FE_INVALID with argument out of range
1841 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1843 [15786] dynamic-link: ifunc resolver functions can smash function
1845 [15918] math: Unnecessary check for equality in hypotf()
1846 [16061] localedata: Review / update transliteration data
1847 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1849 [16141] time: strptime %z offset restriction
1850 [16171] math: drem should be alias of remainder
1851 [16296] math: fegetround is pure?
1852 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1853 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1855 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1856 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1858 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1859 "Exception "Inexact" set" on ppc32
1860 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1861 [16517] math: Missing underflow exception from tanf/tan/tanl
1862 [16519] math: Missing underflow exception from sinhf
1863 [16520] math: Missing underflow exception from tanhf
1864 [16521] math: Missing underflow exception from exp2
1865 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1867 [16734] stdio: fopen calls mmap to allocate its buffer
1868 [16961] math: nan function incorrect handling of bad sequences
1869 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1870 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1871 [16985] locale: localedef: confusing error message when opening output
1873 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1874 [17197] locale: Redundant shift character in iconv conversion output at
1876 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1877 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1878 [17250] dynamic-link: static linking breaks nss loading
1879 (getaddrinfo/getpwnam/etc...)
1880 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1882 [17441] math: isnan() should use __builtin_isnan() in GCC
1883 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1884 (related to lock elision)
1885 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1886 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1887 [17887] time: strptime should be able to parse "+01:00" style timezones
1888 [17905] libc: catopen() Multiple unbounded stack allocations
1890 [18084] libc: backtrace (..., 0) dumps core on x86
1891 [18086] libc: nice() sets errno to 0 on success
1892 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1893 is too large (CVE-2015-8778)
1894 [18251] dynamic-link: SONAME missing when audit modules provides path
1895 [18265] libc: add attributes for wchar string and memory functions
1896 [18370] math: csqrt missing underflows
1897 [18421] libc: [hppa] read-only segment has dynamic relocations
1898 [18472] libc: Obsolete syscall wrappers should be compat symbols
1899 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1900 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1901 [18525] localedata: Remove locale timezone information
1902 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1903 [18568] localedata: Update locale data to Unicode 8.0
1904 [18589] locale: sort-test.sh fails at random
1905 [18595] math: ctan, ctanh missing underflows
1906 [18604] libc: assert macro-expands its argument
1907 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1908 contains a vector instruction exception.
1909 [18611] math: j1, jn missing errno setting on underflow
1910 [18618] localedata: sync Chechen locale definitions with other *_RU
1912 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1913 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1914 [18665] network: In send_dg, the recvfrom function is NOT always using the
1915 buffer size of a newly created buffer (CVE-2015-7547)
1916 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1918 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1920 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1921 [18699] math: tilegx cproj() for various complex infinities does not yield
1923 [18724] libc: Harden put*ent functions against data injection
1924 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1925 [18755] build: build errors with -DNDEBUG
1926 [18757] stdio: fmemopen fails to set errno on failure
1927 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1929 [18781] libc: openat64 lacks O_LARGEFILE
1930 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1931 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1932 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1933 [18790] math: [ldbl-128ibm] tanhl inaccurate
1934 [18795] libc: stpncpy fortification misses buffer lengths that are
1935 statically too large
1936 [18796] build: build fails for --disable-mathvec
1937 [18803] math: hypot missing underflows
1938 [18820] stdio: fmemopen may leak memory on failure
1939 [18823] math: csqrt spurious underflows
1940 [18824] math: fma spurious underflows
1941 [18825] math: pow missing underflows
1942 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1943 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1944 [18870] build: sem_open.c fails to compile with missing symbol
1946 [18872] stdio: Fix memory leak in printf_positional
1947 [18873] libc: posix_fallocate overflow check ineffective
1948 [18875] math: Excess precision leads incorrect libm
1949 [18877] libc: arm: mmap offset regression
1950 [18887] libc: memory corruption when using getmntent on blank lines
1951 [18918] localedata: hu_HU: change time to HH:MM:SS format
1952 [18921] libc: Regression: extraneous stat() and fstat() performed by
1954 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1955 binaries (CVE-2015-8777)
1956 [18951] math: tgamma missing underflows
1957 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1959 [18953] localedata: lt_LT: change currency symbol to the euro
1960 [18956] math: powf inaccuracy
1961 [18961] math: [i386] exp missing underflows
1962 [18966] math: [i386] exp10 missing underflows
1963 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1964 [18969] build: multiple string test failures due to missing locale
1966 [18970] libc: Reference of pthread_setcancelstate in libc.a
1967 [18977] math: float / long double Bessel functions not in XSI POSIX
1968 [18980] math: i386 libm functions return with excess range and precision
1969 [18981] math: i386 scalb*, ldexp return with excess range and precision
1970 [18982] stdio: va_list and vprintf
1971 [18985] time: Passing out of range data to strftime() causes a segfault
1973 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1974 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1975 [19012] locale: iconv_open leaks memory on error path
1976 [19016] math: clog, clog10 inaccuracy
1977 [19018] nptl: Mangle function pointers in tls_dtor_list
1978 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1979 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1980 [19048] malloc: malloc: arena free list can become cyclic, increasing
1982 [19049] math: [powerpc] erfc incorrect zero sign
1983 [19050] math: [powerpc] log* incorrect zero sign
1984 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1985 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1986 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1987 [19074] network: Data race in _res_hconf_reorder_addrs
1988 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1989 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1990 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1992 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1994 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1995 [19086] manual: posix_fallocate64 documented argument order is wrong.
1996 [19088] math: lround, llround missing exceptions close to overflow
1998 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1999 [19095] math: dbl-64 lrint incorrect for 64-bit long
2000 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2001 [19124] dynamic-link: ld.so failed to build with older assmebler
2002 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2003 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2004 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2005 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2006 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2007 pthread_setaffinity_np
2008 [19156] math: [ldbl-128] j0l spurious underflows
2009 [19164] nptl: tst-getcpu fails with many possible CPUs
2010 [19168] math: math/test-ildoubl and math/test-ldouble failure
2011 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2012 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2014 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2015 (FE_NOMASK_ENV) do not clear SSE exceptions
2016 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2017 _int_new_arena/reused_arena
2018 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2019 [19201] math: dbl-64 remainder incorrect sign of zero result
2020 [19205] math: bits/math-finite.h conditions do not match math.h and
2022 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2023 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2024 for C99-based standards
2025 [19212] libc: features.h not -Wundef clean
2026 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2028 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2029 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2030 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2032 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2033 "inexact" exceptions
2034 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2036 [19242] libc: strtol incorrect in Turkish locales
2037 [19243] malloc: reused_arena can pick an arena on the free list, leading
2038 to an assertion failure and reference count corruption
2039 [19253] time: tzset() ineffective when temporary TZ did not include DST
2041 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2042 [19270] math: [hppa] Shared libm missing __isnanl
2043 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2044 MAP_HUGETLB and MAP_STACK defines
2045 [19313] nptl: Wrong __cpu_mask for x32
2046 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2047 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2048 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2049 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2050 [19363] time: x32: times() return value wrongly truncates/sign extends
2052 [19367] dynamic-link: Improve branch prediction on Silvermont
2053 [19369] network: Default domain name not reset by res_ninit when "search"
2054 / "domain" entry is removed from resolv.conf
2055 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2057 [19385] localedata: bg_BG: time separator should be colon, not comma
2058 [19408] libc: linux personality syscall wrapper may erroneously return an
2059 error on 32-bit architectures
2060 [19415] libc: dladdr returns wrong names on hppa
2061 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2062 IBM905, IBM907, and IBM909
2063 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2064 [19443] build: build failures with -DDEBUG
2065 [19451] build: Make check fails on test-double-vlen2
2066 [19462] libc: Glibc failed to build with -Os
2067 [19465] math: Wrong code with -Os
2068 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2070 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2072 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2073 [19486] math: S390: Math tests fail with "Exception Inexact set".
2074 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2075 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2077 [19590] math: Fail to build shared objects that use libmvec.so functions.
2081 * The following bugs are resolved with this release:
2083 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2084 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2085 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2086 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2087 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2088 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2089 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2090 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2091 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2092 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2093 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2094 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2095 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2096 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2097 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2098 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2099 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2100 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2101 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2102 18657, 18676, 18694, 18696, 18887.
2104 * Cache information can be queried via sysconf() function on s390 e.g. with
2105 _SC_LEVEL1_ICACHE_SIZE as argument.
2107 * A buffer overflow in gethostbyname_r and related functions performing DNS
2108 requests has been fixed. If the NSS functions were called with a
2109 misaligned buffer, the buffer length change due to pointer alignment was
2110 not taken into account. This could result in application crashes or,
2111 potentially arbitrary code execution, using crafted, but syntactically
2112 valid DNS responses. (CVE-2015-1781)
2114 * The time zone file parser has been made more robust against crafted time
2115 zone files, avoiding heap buffer overflows related to the processing of
2116 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2117 large time zone data files. Overly long time zone specifiers in the TZ
2118 variable no longer result in stack overflows and crashes.
2120 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2121 for LD and GD on x86 and x86-64, has been implemented. You will need
2122 binutils-2.24 or later to enable this optimization.
2124 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2125 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2126 Hat). These updates cause user visible changes, such as the fix for bug
2129 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2130 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2131 condition in some applications.
2133 * Added vector math library named libmvec with the following vectorized x86_64
2134 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2136 The library can be disabled with --disable-mathvec. Use of the functions is
2137 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2138 Shared library libmvec.so is linked in as needed when using -lm (no need to
2139 specify -lmvec explicitly for not static builds).
2140 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2142 * A new fmemopen implementation has been added with the goal of POSIX
2143 compliance. The new implementation fixes the following long-standing
2144 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2145 old implementation is still present for use be by existing binaries.
2147 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2148 releases. It has been fixed to match 2.19 and older, but binaries built
2149 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2151 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2152 Contributed by Roland McGrath (Google).
2154 * The header <regexp.h> is deprecated, and will be removed in a future
2155 release. Use of this header will trigger a deprecation warning.
2156 Application developers should update their code to use <regex.h> instead.
2158 This header was formerly part of SUS, but was deprecated in 1994 and
2159 removed from the standard in 2001. Also, the glibc implementation
2160 leaks memory. See BZ#18681 for more details.
2164 * The following bugs are resolved with this release:
2166 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2167 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2168 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2169 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2170 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2171 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2172 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2173 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2174 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2175 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2178 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2179 too little memory for the to-be-scanned arguments and overflow the
2180 allocated buffer. The implementation now correctly computes the required
2181 buffer size when using malloc, and switches to malloc from alloca as
2184 * A new semaphore algorithm has been implemented in generic C code for all
2185 machines. Previous custom assembly implementations of semaphore were
2186 difficult to reason about or ensure that they were safe. The new version
2187 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2188 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2189 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2191 * Port to Altera Nios II has been contributed by Mentor Graphics.
2193 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2194 implementations for powerpc64/powerpc64le.
2195 Implemented by Adhemerval Zanella (IBM).
2197 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2198 and powerpc64le. This may improve lock scaling of existing programs on
2199 HTM capable systems. The lock elision code is only enabled with
2200 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2201 powerpc will issue a transaction abort on every syscall to avoid side
2202 effects being visible outside transactions.
2204 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2205 AArch64. Contributed by ARM Ltd.
2207 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2209 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2210 under certain input conditions resulting in the execution of a shell for
2211 command substitution when the application did not request it. The
2212 implementation now checks WRDE_NOCMD immediately before executing the
2213 shell and returns the error WRDE_CMDSUB as expected.
2215 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2216 processing format strings with a large number of format specifiers.
2218 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2219 infinite loop if the DNS response contained a PTR record of an unexpected
2222 * The minimum GCC version that can be used to build this version of the GNU
2223 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2224 still be used to compile programs using the GNU C Library.
2226 * The GNU C Library is now built with -Werror by default. This can be
2227 disabled by configuring with --disable-werror.
2229 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2231 * The obsolete sigvec function has been removed. This was the original
2232 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2233 programs have been using instead for about 25 years. Of course, ABI
2234 compatibility for old binaries using sigvec remains intact.
2236 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2237 with newer versions of bison.
2239 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2240 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2241 registers overlay two consecutive single-precision registers. MIPS32R2
2242 introduced a new FPU mode (FR=1) where double-precision registers extend the
2243 corresponding single-precision registers which is incompatible with the
2244 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2245 require the use of FR=1 making a transition necessary. New o32 ABI
2246 extensions enable users to migrate over time from the original o32 ABI
2247 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2248 tracks the ABI of any loaded object and verifies that new objects are
2249 compatible. Mode transitions will also be requested as required and
2250 unsupportable objects will be rejected. The ABI checks include both soft and
2251 hard float ABIs for o32, n32 and n64.
2253 GCC 5 with GNU binutils 2.25 onwards:
2254 It is strongly recommended that all o32 system libraries are built using the
2255 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2256 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2257 compiler using --with-fp-32=xx to set this by default.
2261 * The following bugs are resolved with this release:
2263 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2264 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2265 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2266 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2267 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2268 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2269 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2270 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2271 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2272 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2273 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2274 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2275 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2276 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2277 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2279 * Reverted change of ABI data structures for s390 and s390x:
2280 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2281 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2282 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2283 are preserved pointing straight to the same implementation as the old ones.
2284 Given that, new callers will simply provide a too-big buffer to these
2285 functions. Any applications/libraries out there that embed jmp_buf or
2286 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2287 against 2.19 headers will have to rebuilt again. This is necessary in any
2288 case to revert the breakage in their ABI caused by the glibc change.
2290 * Support for file description locks is added to systems running the
2291 Linux kernel. The standard file locking interfaces are extended to
2292 operate on file descriptions, not file descriptors, via the use of
2293 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2294 are associated with an open file instead of a process.
2296 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2298 * The minimum Linux kernel version that this version of the GNU C Library
2299 can be used with is 2.6.32.
2301 * Running the testsuite no longer terminates as soon as a test fails.
2302 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2303 with PASS or FAIL lines for individual tests. A summary of the results is
2304 printed, including a list of failing lists, and "make check" exits with
2305 error status if there were any unexpected failures. "make check
2306 stop-on-test-failure=y" may be used to keep the old behavior.
2308 * The am33 port, which had not worked for several years, has been removed
2311 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2312 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2313 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2314 conflicted with POSIX (support for which was removed in 2.19), the
2315 interfaces those macros enabled remain available when compiling with
2316 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2317 test macros defined.
2319 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2321 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2322 This may improve lock scaling of existing programs on TX capable systems.
2323 The lock elision code is only built with --enable-lock-elision=yes and
2324 then requires a GCC version supporting the TX builtins. With lock elision
2325 default mutexes are elided via __builtin_tbegin, if the cpu supports
2326 transactions. By default lock elision is not enabled and the elision code
2329 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2330 copy the path argument. This allowed programs to cause posix_spawn to
2331 deference a dangling pointer, or use an unexpected pathname argument if
2332 the string was modified after the posix_spawn_file_actions_addopen
2335 * All supported architectures now use the main glibc sysdeps directory
2336 instead of some being in a separate "ports" directory (which was
2337 distributed separately before glibc 2.17).
2339 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2340 On configurations that support it (all Linux configurations), it's now
2341 used regardless of the --enable-add-ons switch to configure. It is no
2342 longer possible to build such configurations without pthreads support.
2344 * Locale names, including those obtained from environment variables (LANG
2345 and the LC_* variables), are more tightly checked for proper syntax.
2346 setlocale will now fail (with EINVAL) for locale names that are overly
2347 long, contain slashes without starting with a slash, or contain ".." path
2348 components. (CVE-2014-0475) Previously, some valid locale names were
2349 silently replaced with the "C" locale when running in AT_SECURE mode
2350 (e.g., in a SUID program). This is no longer necessary because of the
2353 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2354 with application code using Intel MPX instructions. (With all previous
2355 versions, the MPX register state could be clobbered when making calls
2356 into or out of a shared library.) Note that while the new dynamic
2357 linker is compatible with all known x86 hardware whether or not it
2358 supports Intel MPX, some x86 instruction-set emulators might fail to
2359 handle the new instruction encodings. This is known to affect Valgrind
2360 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2361 release), and might affect other tools that do instruction emulation.
2363 * Support for loadable gconv transliteration modules has been removed.
2364 The support for transliteration modules has been non-functional for
2365 over a decade, and the removal is prompted by security defects. The
2366 normal gconv conversion modules are still supported. Transliteration
2367 with //TRANSLIT is still possible, and the //IGNORE specifier
2368 continues to be supported. (CVE-2014-5119)
2370 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2371 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2372 resulting a denial-of-service security vulnerability in applications which
2373 use functions related to iconv. (CVE-2014-6040)
2377 * The following bugs are resolved with this release:
2379 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2380 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2381 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2382 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2383 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2384 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2385 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2386 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2387 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2388 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2389 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2390 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2391 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2392 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2393 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2394 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2395 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2396 16431, 16453, 16474, 16506, 16510, 16529
2398 * Slovenian translations for glibc messages have been contributed by the
2399 Translation Project's Slovenian team of translators.
2401 * The public headers no longer use __unused nor __block. This change is to
2402 support compiling programs that are derived from BSD sources and use
2403 __unused internally, and to support compiling with Clang's -fblock
2404 extension which uses __block.
2406 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2407 large collation sequences to optimize multiple passes. This cache
2408 computation may overflow for large collation sequences and may cause a
2409 stack or buffer overflow. This is now fixed to use a slower algorithm
2410 which does not use a cache if there is an integer overflow.
2412 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2413 rules for large collation sequences to optimize multiple passes and falls
2414 back to alloca if malloc fails, resulting in a possible stack overflow.
2415 The implementation now falls back to an uncached collation sequence lookup
2418 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2419 initialized for static applications resulting in the security feature
2420 being disabled. The pointer guard is now correctly initialized to a
2421 random value for static applications. Existing static applications need
2422 to be recompiled to take advantage of the fix (bug 15754).
2424 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2425 to the d_name member of struct dirent, or omit the terminating NUL
2426 character. (Bugzilla #14699).
2428 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2429 aligned_alloc functions could allocate too few bytes or corrupt the
2430 heap when passed very large allocation size values (Bugzilla #15855,
2433 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2434 for AF_INET6 has been fixed (Bugzilla #16072).
2436 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2437 pap_CW, quz_PE, the_NP.
2439 * Substantially revised locales: gd_GB, ht_HT
2441 * The LC_ADDRESS field was updated to support country_car for almost all
2444 * ISO 1427 definitions were updated.
2446 * ISO 3166 definitions were updated.
2448 * The localedef utility now supports --big-endian and --little-endian
2449 command-line options to generate locales for a different system from that
2450 for which the C library was built.
2452 * Binary locale files now only depend on the endianness of the system for
2453 which they are generated and not on other properties of that system. As a
2454 consequence, binary files generated with new localedef may be incompatible
2455 with old versions of the GNU C Library, and binary files generated with
2456 old localedef may be incompatible with this version of the GNU C Library,
2457 in the following circumstances:
2459 + Locale files may be incompatible on m68k systems.
2461 + Locale archive files (but not separate files for individual locales) may
2462 be incompatible on systems where plain "char" is signed.
2464 * The configure option --disable-versioning has been removed. Builds with
2465 --disable-versioning had not worked for several years.
2467 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2468 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2470 * SystemTap probes for malloc have been introduced.
2472 * SystemTap probes for slow multiple precision fallback paths of
2473 transcendental functions have been introduced.
2475 * Support for powerpc64le has been added.
2477 * The soft-float powerpc port now supports e500 processors.
2479 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2481 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2482 set of header declarations that are enabled by default, even when other
2483 feature test macros or compiler options such as -std=c99 would otherwise
2484 disable some of those declarations.
2486 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2487 conflict with POSIX. The libbsd-compat library (which was a dummy library
2488 that did nothing) has also been removed.
2490 * Preliminary documentation about Multi-Thread, Async-Signal and
2491 Async-Cancel Safety has been added.
2493 * Change of ABI data structures for s390 and s390x:
2494 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2495 allow for future hardware extensions. All library functions that accept or
2496 return these structures were versioned in order to provide backwards
2497 compatibility with existing code. However, not all cases can be handled
2498 with interface versioning. If an application embeds either structure into
2499 another structure and that structure is passed to another compilation unit
2500 which expects the newer larger-sized object then any access to the new
2501 fields will result in undefined behaviour. Similarly any access to fields
2502 that were moved by the enlarging of the structures will cause undefined
2503 behaviour. To fix the undefined behaviour all objects that are part of the
2504 final in-memory image must agree on the size of structures, and this may
2505 require recompilation.
2509 * The following bugs are resolved with this release:
2511 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2512 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2513 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2514 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2515 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2516 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2517 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2518 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2519 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2520 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2521 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2522 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2523 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2526 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2527 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2528 Distributions can re-enable building and using pt_chown via the new configure
2529 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2530 considerable security risks and should only be used if the distribution
2531 understands and accepts the risks.
2533 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2536 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2537 fixed (Bugzilla #15330).
2539 * Add support for calling C++11 thread_local object destructors on thread
2540 and program exit. This needs compiler support for offloading C++11
2541 destructor calls to glibc.
2543 * Improved worst case performance of libm functions with double inputs and
2546 * Support for priority inherited mutexes in pthread condition variables on
2547 non-x86 architectures.
2549 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2551 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2553 * Optimized string functions for ARM. Implemented by Will Newton and
2556 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2558 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2561 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2562 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2564 * Added a benchmark framework to track performance of functions in glibc.
2566 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2567 It is based on draft TS 18661 and currently enabled as a GNU extension.
2569 * On Linux, the clock function now uses the clock_gettime system call
2570 for improved precision, rather than old times system call.
2572 * Added support for version-3 tz format files. This is needed when using
2573 the tz database release 2013e or later, and affects a few unusual cases --
2574 currently only TZ='America/Godthab' for time stamps after 2037.
2576 * Added new API functions pthread_getattr_default_np and
2577 pthread_setattr_default_np to get and set the default pthread
2578 attributes of a process.
2580 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2581 This may improve lock scaling of existing programs on TSX capable systems.
2582 When the --enable-lock-elision=yes parameter is specified at configure
2583 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2586 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2587 availability). Implemented by Ryan S. Arnold.
2589 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2593 * The following bugs are resolved with this release:
2595 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2596 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2597 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2598 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2599 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2600 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2601 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2602 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2603 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2604 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2605 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2606 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2607 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2608 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2609 14879, 14889, 14893, 14898, 14914.
2611 * Optimization of memcpy for MIPS.
2613 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2614 EMFILE has been fixed (Bugzilla #14889).
2616 * The add-on ports collection is now distributed in the "ports" subdirectory
2617 of the main GNU C Library distribution, rather than separately.
2619 * Port to ARM AArch64 contributed by Linaro.
2621 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2622 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2624 Implemented by Andreas Krebbel.
2626 * The new function secure_getenv allows secure access to the environment,
2627 returning NULL if running in a SUID/SGID process. This function replaces
2628 the internal function __secure_getenv.
2630 * SystemTap static probes have been added into the dynamic linker.
2631 Implemented by Gary Benson.
2633 * Optimizations of string functions strstr, strcasestr and memmem.
2634 Implemented by Maxim Kuvyrkov.
2636 * The minimum Linux kernel version that this version of the GNU C Library
2637 can be used with is 2.6.16.
2639 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2640 powerpc POWER7. Implemented by Will Schmidt.
2642 * New configure option --disable-nscd builds the C library such that it
2643 never attempts to contact the Name Service Caching Daemon (nscd).
2644 New configure option --disable-build-nscd avoids building nscd itself;
2645 this is the default if --disable-nscd is used.
2647 * Improved support for cross-compilation, including cross-testing and
2648 bootstrap builds without a previously built glibc.
2650 * Several testsuite tests are now able to test multiple IFUNC variants of an
2651 interface, rather than just testing the one that would be chooen by
2654 * New configure options --with-bugurl and --with-pkgversion, for
2655 distributors to use to embed their bug-reporting and package version
2656 information in --help and --version output.
2658 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2659 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2660 allows creation of chroots without the procfs mounted on /proc.
2662 * The `crypt' function now fails if passed salt bytes that violate the
2663 specification for those values. On Linux, the `crypt' function will
2664 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2665 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2666 when the mode is enabled.
2668 * The `clock_*' suite of functions (declared in <time.h>) is now available
2669 directly in the main C library. Previously it was necessary to link with
2670 -lrt to use these functions. This change has the effect that a
2671 single-threaded program that uses a function such as `clock_gettime' (and
2672 is not linked with -lrt) will no longer implicitly load the pthreads
2673 library at runtime and so will not suffer the overheads associated with
2674 multi-thread support in other code such as the C++ runtime library.
2676 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2681 * The following bugs are resolved with this release:
2683 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2684 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2685 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2686 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2687 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2688 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2689 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2690 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2691 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2692 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2693 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2694 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2695 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2696 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2697 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2698 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2699 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2700 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2701 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2702 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2703 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2704 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2707 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2708 configuring glibc with:
2709 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2710 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2711 Implemented by H.J. Lu.
2715 + define static_assert
2717 + do not declare gets
2719 + declare at_quick_exit and quick_exit also for ISO C11
2721 + aligned_alloc. NB: The code is deliberately allows the size parameter
2722 to not be a multiple of the alignment. This is a moronic requirement
2723 in the standard but it is only a requirement on the caller, not the
2726 + timespec_get added
2728 + uchar.h support added
2730 + CMPLX, CMPLXF, CMPLXL added
2732 Implemented by Ulrich Drepper.
2734 * Support for the IA-64 has been moved to ports.
2736 * Remove support for anything but ELF binary format
2738 * Checking versions of poll, ppoll added.
2739 Implemented by Ulrich Drepper.
2741 * More generic and 64-bit performance optimizations to math functions.
2742 Implemented by Ulrich Drepper.
2744 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2745 headers and functions available at compile time as they were before
2746 version 2.14. This option will be removed at some time in the future
2747 after the TI-RPC library becomes fully sufficient for the needs of
2748 existing applications.
2750 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2751 Note that glibc is not expected to work with any Linux kernel version
2754 * New header <sys/auxv.h> and function getauxval allowing easy access to
2755 the AT_* key-value pairs passed from the Linux kernel. The header also
2756 defines the HWCAP_* bits associated with the AT_HWCAP key.
2758 * New locales: mag_IN
2760 * New configure option --enable-systemtap builds SystemTap static probes
2761 into libc for setjmp and longjmp and into libpthread for various operations.
2762 So far the setjmp/longjmp probes and some of the libpthread probes are
2763 provided only for i*86 and x86_64.
2764 Implemented by Roland McGrath and Rayson Ho.
2766 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2768 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2771 * More optimized functions for SPARC. Implemented by David S. Miller.
2773 * Improved support for cross-compilation, especially bootstrap builds
2774 without a previously built glibc.
2776 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2777 Chris Metcalf from Tilera.
2779 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2780 now supported for ARM processors.
2782 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2783 as the name of the dynamic linker, to distinguish it from the
2784 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2786 * Support for CRIS has been removed from ports.
2788 * A new class of installed header has been documented for low-level
2789 platform-specific functionality. PowerPC added the first instance with a
2790 function to provide time base register access. Contributed by Tulio
2791 Magno Quites Machado Filho.
2793 * ix86 configurations now install header files that are consistent with
2794 what x86-64 configurations install. These same header files can be used
2795 for -m32, -m64, or -mx32 builds.
2796 Contributed by H.J. Lu.
2798 * Math library bug fixes. A thorough audit of all open math library bugs was
2799 conducted by Joseph Myers. Significant progress was made on many math
2800 library bugs resulting in more accurate exceptions and function results.
2801 Many thanks to all those that contributed including Andreas Jaeger for his
2802 patch review and work on the x87 trigonometric instruction issues.
2804 * Timezone data is no longer installed. Timezone-related binaries and scripts
2805 will continue to be installed. Users should obtain their timezone data from
2806 their distribution provider or from the tzdata package at
2807 <ftp://munnari.oz.au/pub/>.
2811 * The following bugs are resolved with this release:
2813 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2814 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2815 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2816 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2817 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2818 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2819 13484, 13506, 13515, 13523, 13524, 13538, 13540
2821 * New program pldd to list loaded object of a process
2822 Implemented by Ulrich Drepper.
2824 * Add nss_db support back to glibc. No more dependency on Berkeley db
2825 and support for initgroups lookups.
2826 Implemented by Ulrich Drepper.
2828 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2829 Contributed by HJ Lu.
2831 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2832 Contributed by HJ Lu.
2834 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2835 on x86-32 and x86-64.
2836 Contributed by Liubov Dmitrieva.
2838 * Optimized strchr and strrchr for SSE on x86-32.
2839 Contributed by Liubov Dmitrieva.
2841 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2842 for x86-64 and x86-32.
2843 Contributed by Liubov Dmitrieva.
2845 * New interfaces: scandirat, scandirat64
2846 Implemented by Ulrich Drepper.
2848 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2849 Implemented by Ulrich Drepper.
2851 * nscd now also caches the netgroup database.
2852 Implemented by Ulrich Drepper.
2854 * Integrate libm with gcc's -ffinite-math-only option.
2855 Implemented by Ulrich Drepper.
2857 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2858 to math functions. Implemented by Ulrich Drepper.
2860 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2861 Implemented by Ulrich Drepper.
2863 * New Linux interfaces: process_vm_readv, process_vm_writev
2865 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2866 Implemented by Ulrich Drepper.
2868 * Optimized nearbyint and strcasecmp for PPC.
2869 Implemented by Adhemerval Zanella.
2871 * New locales: bho_IN, unm_US, es_CU, ta_LK
2875 * The following bugs are resolved with this release:
2877 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2878 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2879 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2880 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2881 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2882 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2883 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2884 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2885 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2886 12795, 12811, 12813, 12814, 12841
2888 * The RPC implementation in libc is obsoleted. Old programs keep working
2889 but new programs cannot be linked with the routines in libc anymore.
2890 Programs in need of RPC functionality must be linked against TI-RPC.
2891 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2893 Visible changes of this change include (obviously) the inability to link
2894 programs using RPC functions without referencing the TI-RPC library and the
2895 removal of the RPC headers from the glibc headers.
2896 Implemented by Ulrich Drepper.
2898 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2899 syncfs, setns, sendmmsg
2901 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2902 yue_HK, lij_IT, mhr_RU
2904 * New encodings: CP770, CP771, CP772, CP773, CP774
2906 * New program sotruss to trace calls through PLTs
2907 Implemented by Ulrich Drepper.
2909 * The malloc hook implementation is marked deprecated and will be removed
2910 from the default implementation in the next version. The design never
2911 worked ever since the introduction of threads. Even programs which do
2912 not create threads themselves can use multiple threads created internally.
2916 * The following bugs are resolved with this release:
2918 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2919 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2920 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2921 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2924 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2926 * POWER7 optimizations: memset, memcmp, strncmp
2928 * New optimized string functions for x86-64: strnlen (SSE2),
2929 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2930 Implemented by Ulrich Drepper.
2934 * The following bugs are resolved with this release:
2936 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2937 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2938 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2939 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2940 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2941 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2942 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2943 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2945 * New interfaces: pthread_getname_np, pthread_setname_np
2947 * New Linux interface: recvmmsg
2949 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2951 * The dynamic linker now recognizes supported ABI versions from the
2952 EI_ABIVERSION field in the ELF header.
2953 Implemented by Ulrich Drepper.
2955 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2956 will not be used to fill in password fields in the passwd.byname replies.
2957 Instead it is used to synthesize the shadow.byname table, should it be
2958 missing. This is a useful mode in some installations involving Solaris.
2959 Implemented by Ulrich Drepper.
2961 * New locales: kok_IN, sq_MK, cv_RU
2965 * The following bugs are resolved with this release:
2967 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2968 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2969 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2970 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2971 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2972 10789, 10817, 10823, 10840, 10847
2974 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2976 Implemented by Ulrich Drepper.
2978 * Checking version of longjmp added that fails if an uninitialized stack
2979 frame would be created. Implemented by Ulrich Drepper.
2981 * STT_GNU_IFUNC is now supported in static executables.
2982 Implemented by H.J. Lu.
2984 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2986 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2987 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2988 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2989 Contributed by H.J. Lu.
2991 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2992 Implemented by Ulrich Drepper.
2994 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2996 Contributed by H.J. Lu.
2998 * Support for fma instruction in AVX on x86-64.
2999 Implemented by H.J. Lu and Ulrich Drepper.
3001 * AVX support in x86-64 auditing support in ld.so.
3002 Implemented by H.J. Lu.
3004 * STB_GNU_UNIQUE support added.
3005 Implemented by Ulrich Drepper.
3007 * Implement second fallback mode for optimized DNS lookup for even more
3008 broken environments. If two requests from the same port are not
3009 handled correctly close the socket and open a new one before sending
3010 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3011 can be used to select this mode right away, instead of rediscovering the
3012 necessity is every process again.
3013 Implemented by Ulrich Drepper.
3015 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3016 Implemented by Adam Tkac.
3018 * Optimized iconv conversions for S390x.
3019 Implemented by Andreas Krebbel.
3021 * Using condvars with PI mutexes is now more efficient due to kernel
3022 support for requeueing to PI futexes. NPTL support added for x86-64.
3023 Implemented by Ulrich Drepper.
3025 * New locales: ps_AF, my_MM
3029 * The following bugs are resolved with this release:
3031 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3032 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3033 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3034 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3035 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3037 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3038 Implemented by Ulrich Drepper.
3040 * New Linux interfaces: accept4, fallocate, fallocate64.
3041 Implemented by Ulrich Drepper.
3043 * Correct declarations of string function when used in C++ code. This
3044 could lead to compile errors for invalid C++ code.
3046 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3049 * New POSIX 2008 interface: psiginfo
3050 Implemented by Ulrich Drepper.
3052 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3053 Implemented by Ulrich Drepper.
3055 * Support for selecting between multiple function definitions at runtime
3056 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3058 * The libcrypt library can now use the hash function implementations in
3059 NSS. Implemented by Ulrich Drepper.
3061 * The malloc implementation can be compiled to be less memory efficient
3062 but higher performing in multi-threaded programs.
3063 Implemented by Ulrich Drepper.
3065 * New locales: nan_TW@latin, ks_IN
3067 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3068 Implemented by Ulrich Drepper.
3070 * Extended printf hook support. It is possible to use user-defined types
3071 and extend existing format specifiers.
3072 Implemented by Ulrich Drepper.
3074 * Handling for group shadow files has been added.
3075 Implemented by Ulrich Drepper.
3077 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3078 servers (the case, e.g., for some people using the built-in DNS
3079 server in ADSL modems/routers). There is a once-per-process timeout
3080 in case of a broken server. To avoid it, users can run nscd or put
3081 'options single-request' in /etc/resolv.conf.
3082 Implemented by Ulrich Drepper.
3086 * The following bugs are resolved with this release:
3088 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3089 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3090 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3091 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3092 6968, 6974, 6980, 6995, 7008, 7009, 7029
3094 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3095 up at the same time. Implemented by Ulrich Drepper.
3097 * TLS descriptors for LD and GD on x86 and x86-64.
3098 Implemented by Alexandre Oliva.
3100 * getaddrinfo now handles DCCP and UDPlite.
3101 Implemented by Ulrich Drepper.
3103 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3104 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3105 Implemented by Ulrich Drepper.
3107 * New implementation of memmem, strstr, and strcasestr which is O(n).
3108 Implemented by Eric Blake.
3110 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3112 * Implement "e" option for popen to open file descriptor with the
3113 close-on-exec flag set. Implemented by Ulrich Drepper.
3115 * Implement "b" mode for fmemopen. In this mode writes writes don't
3116 implicitly add a NUL byte and seeks from the end of the buffer really
3117 use the buffer end, not the string length as the basis.
3118 Implemented by Ulrich Drepper.
3120 * Many functions, exported and internal, now atomically set the close-on-exec
3121 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3123 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3125 Implemented by Pravin Satpute.
3127 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3131 * The following bugs are resolved with this release:
3133 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3134 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3135 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3136 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3137 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3138 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3139 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3141 * New locales: bo_CN, bo_IN, shs_CA.
3143 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3145 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3146 Implemented by Pravin Satpute.
3148 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3150 * nscd caches now all timeouts for DNS entries
3151 Implemented by Ulrich Drepper.
3153 * nscd is more efficient and wakes up less often.
3154 Implemented by Ulrich Drepper.
3156 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3157 vdprintf, and obstack_vprintf.
3158 Implemented by Jakub Jelinek.
3160 * Faster memset for x86-64.
3161 Implemented by Harsha Jagasia and H.J. Lu.
3163 * Faster memcpy on x86.
3164 Implemented by Ulrich Drepper.
3166 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3167 Implemented by Ulrich Drepper.
3169 * Faster sqrt and sqrtf implemention for some PPC variants.
3170 Implemented by Steven Munroe.
3174 * The following bugs are resolved with this release:
3176 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3177 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3178 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3179 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3180 5071, 5103, 5104, 5112, 5113, 5184, 5186
3182 * More checking functions: fread, fread_unlocked, open*, mq_open.
3183 Implemented by Jakub Jelinek and Ulrich Drepper.
3185 * Extend fortification to C++. Implemented by Jakub Jelinek.
3187 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3188 by not recognizing 'a' as a modifier when those specs are requested.
3189 Implemented by Jakub Jelinek.
3191 * PPC optimizations to math and string functions.
3192 Implemented by Steven Munroe.
3194 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3195 options to be passed. Implemented by Ulrich Drepper.
3197 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3199 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3200 Implemented by Ulrich Drepper.
3202 * Handle private futexes in the NPTL implementation.
3203 Implemented by Jakub Jelinek and Ulrich Drepper.
3205 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3206 Implemented by Roland McGrath and Ulrich Drepper.
3208 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3210 * SHA-256 and SHA-512 based password encryption.
3211 Implemented by Ulrich Drepper.
3213 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3214 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3217 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3218 Implemented by Ulrich Drepper.
3222 * The following bugs are resolved with this release:
3224 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3225 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3226 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3227 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3228 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3229 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3230 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3233 * New Linux interfaces: epoll_pwait, sched_getcpu.
3235 * New generic interfaces: strerror_l.
3237 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3241 * The following bugs are resolved with this release:
3243 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3244 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3245 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3246 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3247 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3248 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3249 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3250 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3251 3225, 3251, 3252, 3253, 3273
3253 * For Linux, the sorting of addresses returned by getaddrinfo now also
3254 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3255 Implemented by Ulrich Drepper.
3257 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3258 Implemented by Ulrich Drepper.
3260 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3262 * New iconv module for MIK. Contributed by Alexander Shopov.
3264 * For sites with broken group and/or passwd database, the auto-propagate
3265 option of nscd can prevent creating ID lookup entries from the results
3266 of a name lookup and vice versa. This usually is no problem but some
3267 site might have problems with the default behavior.
3268 Implemented by Ulrich Drepper.
3270 * Iterating over entire database in NIS can be slow. With the
3271 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3272 to trade time for memory. The entire database will be read at once.
3273 Implemented by Ulrich Drepper.
3275 * The interfaces introduced in RFC 3542 have been implemented by
3278 * Support for the new ELF hash table format was added by Ulrich Drepper.
3280 * Support for priority inheritance mutexes added by Jakub Jelinek and
3283 * Support for priority protected mutexes added by Jakub Jelinek.
3285 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3286 or_IN, csb_PL, fy_NL, sr_ME.
3290 * More overflow detection functions.
3292 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3293 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3295 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3296 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3297 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3298 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3299 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3300 by Masahide Washizawa.
3302 * It is now possible to install audit libraries for the dynamic linker using
3303 LD_AUDIT. Implemented by Ulrich Drepper.
3305 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3306 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3307 For a libc and libpthread that works well on Linux 2.4 kernels, we
3308 recommend using the stable 2.3 branch.
3310 * The new function open_wmemstream acts like open_memstream,
3311 but uses a wchar_t wide character buffer.
3313 * The new function ppoll is an improved version of poll, similar to pselect.
3315 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3316 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3317 renameat, unlinkat, linkat, symlinkat, readlinkat.
3319 * New Linux kernel system calls: unshare,
3320 inotify_init, inotify_add_watch, inotify_rm_watch.
3322 * The euidaccess function is now also known by the alias eaccess,
3323 for compatibility with some other systems.
3325 * Timezone data updated to 2006b version.
3329 * The following bugs are resolved with this release:
3331 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3332 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3333 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3334 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3335 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3336 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3338 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3340 * As of this release, GCC 4 can be used to compile the C Library.
3342 * Timezone data updated to 2005m version.
3346 * The following bugs are resolved with this release:
3348 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3349 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3350 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3351 777, 787, 821, 822, 823, 825
3353 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3357 * Support for RFC 3678. Real implementations exist only for Linux so far.
3358 Implemented by Ulrich Drepper.
3360 * nscd can now cache entries persistently. Expiring entries are reloaded.
3361 For speedups the cache can be shared in memory with client processes.
3362 Implemented by Ulrich Drepper.
3364 * nscd can now perform SELinux checks.
3365 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3367 * getaddrinfo queries are now cached. Canonical name lookup is performed
3369 Implemented by Ulrich Drepper.
3371 * The nothrow function attribute is used when headers are used by gcc when
3372 compiling C code. This can avoid significant amounts of exception
3375 * The malloc functions perform more error checking and are stricter when
3376 it comes to reacting on errors. The default action is to terminate
3377 the process after showing an error message. Implemented by Ulrich Drepper.
3379 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3380 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3382 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3383 namespaces using the new function dlmopen(). This feature is of course,
3384 like most other dynamic loading functionality, not available in statically
3385 linked applications. Implemented by Ulrich Drepper.
3387 * Low-overhead boundary checking variants of string and some stdio functions
3388 were added. These are to be used in conjunction with a gcc patch by
3389 Jakub Jelinek which adds calls to these functions if possible.
3390 Implemented by Jakub Jelinek and Ulrich Drepper.
3392 * Old code for several operating systems and machine architectures that
3393 have not been in working condition in a long time have been removed from
3394 the main source tree maintained by the GNU C Library's maintainers.
3395 These files are now reside in the separate `ports' source module
3396 that is usable as an add-on when building the library.
3400 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3401 interrogate the dynamic linker, compatible with the Solaris interface.
3403 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3404 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3406 * getifaddrs now uses the netlink interface on Linux to get its information.
3407 Implemented by Thorsten Kukuk.
3409 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3410 Implemented by Ulrich Drepper.
3412 * support for non-executable stacks on x86 has been added. Changes mostly
3415 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3418 * getaddrinfo now performs destination address selection according to
3423 * Thread-safe interfaces for many functions that access locale data
3424 were added in version 2.3, but these features were omitted from NEWS.
3425 Many functions have variants with an `_l' suffix that take a `locale_t'
3426 object as a parameter rather than consulting the current locale.
3427 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3428 create and maintain `locale_t' objects. Additionally, the new function
3429 `uselocale' sets "the current locale" (as used by functions not so
3430 parameterized) set for an individual thread. These features were added
3431 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3433 * The functions getresuid, getresgid, setresuid, and setresgid, which
3434 have long been available on Linux, are now declared in <unistd.h>
3435 and are now also available on the Hurd.
3437 * ELF thread-local storage support (TLS) now works on x86-64.
3439 * The new dynamic string token $LIB is expanded in shared library names.
3440 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3442 * Aldy Hernandez contributed complete software floating point support for
3443 PowerPC machines with no FPU.
3445 * fexecve is implemented on Linux.
3447 * The `btowc' function should work at least twice as fast due to
3448 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3450 * With approriate thread add-ons cancelable functions are now implemented
3451 in libc.so as well. No need to call the function in libpthread. This
3452 change allowed to finally disable the incorrect and expensive handling
3453 of weak definition in ld.so.
3455 * Yet more PLT entries in libc.so have been removed. We finally arrived
3456 at the bare minimum. Startup times improved appropriately.
3458 * Support for the new Linux/x86 system call interface was added. The
3459 AT_SYSINFO auxiliary vector entry is recognized and handled.
3463 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3466 * iconv (the program and the interface) now accepts empty names (excluding
3467 options like //TRANSLIT) to mean "use charset of current locale".
3469 * localedef can now transliterate characters in strings which are not in
3470 the provided charmap. The information from the input locale is used.
3472 * Prelinking support was added for ELF targets. This requires additional
3473 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3475 * Read-only stdio streams now use mmap to speed up operation by eliminating
3476 copying and buffer underflows. To use add 'm' to the mode string of
3477 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3479 * The malloc functions were completely rewritten by Wolfram Gloger based
3480 on Doug Lea's malloc-2.7.0.c.
3482 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3483 implementation of regex.
3485 * Bruno Haible upgraded the iconv and locale implementation to support
3488 * Contents of the LC_* and LANG environment variables in the CEN style are
3489 not recognized anymore. It never was used. Change by Ulrich Drepper.
3491 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3492 thread-local storage (TLS) ABI on some platforms.
3493 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3495 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3496 EUC-JISX0213, and TSCII.
3498 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3499 BSD-compatible interface for getting all network interface addresses.
3500 Implementation for IPv4 by Roland McGrath.
3502 * Loading of locale data is faster due to the introduction of a locale
3503 archive. Implemented by Roland McGrath and Ulrich Drepper.
3505 * Startup times are significantly reduced by not using exported functions
3506 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3509 * Steven Munroe contributed a port to PowerPC64/Linux.
3513 * The Hurd now uses the GNU libio implementation of stdio rather than the
3514 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3516 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3517 with the same behavior as the Linux system call of the same name.
3521 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3522 128-bit long double format.
3524 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3525 IBM1161, and IBM1162 charsets.
3527 * Andreas Jaeger contributed a port to x86-64/Linux.
3529 * Peter Bruin contributed a port to PowerPC/Hurd.
3531 * libc functions using I/O streams now can handle wide-oriented streams
3534 * optimizations in the dynamic linker. Binaries created by recent binutils
3535 versions start up quicker due to reduced time spent on relocations.
3537 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3541 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3542 asinh, atanh, j0 for the 128-bit long double format.
3544 * Bruno Haible updated all the code handling Unicode in some form to
3545 support Unicode 3.1.
3547 * Speed of regex for single-byte locales is back to previous levels.
3548 Patch by Isamu Hasegawa.
3550 * Alpha, SPARC, and IA-64 now also using floating stacks.
3552 * Startup time of internationalized applications greatly improved through
3553 iconv cache. Use iconvconfig to generate the cache file.
3554 Contributed by Ulrich Drepper.
3556 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3557 severe performance problems. Patch by David Mosberger.
3559 * The Hurd port got a lot more functionality like AIO, various stdio
3560 extensions, etc. Mainly done by Roland McGrath.
3562 * mtrace can now lookup symbols in shared libraries.
3566 * Intel's IA-64 math library is largely integrated. It provides fast and
3567 accurate implementatations for most basic and standard math functions
3568 in float, double, and long double format.
3570 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3571 and asin for the 96-bit long double format and asin, log, tan for the
3572 128-bit long double format.
3574 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3575 The basic double functions exist today. Contributed by Abraham Ziv
3576 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3577 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3579 * An asynchronous name lookup library was added. The interface is designed
3580 after POSIX AIO. The proposal was circulated beforehand to get comments.
3581 No negative ones came in. Implemented by Ulrich Drepper.
3583 * Port to S390/64bit contributed by Martin Schwidefsky
3584 <schwidefsky@de.ibm.com>.
3586 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3587 of functions for Linux/IA-64.
3589 * The RPC code is now thread safe. Threads can now use the same service
3590 of different services at the same time. Patch by Eric Norum
3591 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3593 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3594 family of functions for Linux/S390.
3596 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3597 of functions for Linux/x86.
3599 * Port to Linux/CRIS contributed by Axis Communications.
3603 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3604 we can now check for namespace violations and missing declarations. The
3605 result is that almost all headers are now Unix-compliant (as defined in
3606 the upcoming XPG6). The negative side is that some programs might need
3607 corrections, too, if they depend on the incorrect form of the headers in
3608 previous versions which defined too many symbols and included too many
3611 * regex now handles multibyte character sets correctly.
3612 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3614 * iconv (the program) does now conform to the upcoming XPG6 and handles
3615 charmaps. Instead of the charset names the path of charmaps can be
3616 provided and the conversion happens based on this data.
3617 Contributed by Ulrich Drepper.
3619 * The locale program now provides detailed information about the installed
3624 only lists the names of the supported locales
3628 provides details such as country, language, and codeset name.
3629 Contributed by Ulrich Drepper.
3633 * The gencat program now parses the input file according to the charset
3634 selected by the LC_CTYPE category. This is important for stateful
3635 character sets. To make generating catalogs easier there is a way
3636 to overwrite the charset selected by the locale: before the first
3637 message or $ quote line the catalog can contain a line like
3639 $ codeset=ISO-8859-2
3641 to select the charset (ISO-8859-2 in this case).
3643 Implemented by Shinya Hanataka and Ulrich Drepper.
3645 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3646 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3647 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3649 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3650 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3652 * The LANGUAGE environment variable is now ignored unless the locale is
3653 changed from the default "C" locale.
3655 * The usual bug fixes.
3659 * Greg McGary added runtime support for bounds checking using gcc's
3660 new -fbounded-pointers option. ix86 target is complete. PowerPC
3663 * Thorsten Kukuk added secure mode support to nscd.
3665 * The Berkeley db libraries have been removed.
3667 Related, the nss_db module is now in a separate package since it
3668 obviously requires a database library being available.
3670 * Wide character I/O streams implemented by Ulrich Drepper.
3672 * Functions from the extended socket API added by Ulrich Drepper.
3674 * Functions feenableexcept and fedisableexcept to control the
3675 behaviour of individual exceptions have been added by Andreas Jaeger.
3677 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3679 * The resolver code has been updated from bind 8.2.3-T5B which supports
3680 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3683 This change could in some situations effect backward compatibility. Since
3684 now `_res' is a thread-local instead of a global variable, modifying it
3685 in one thread does not have any effect in other threads.
3687 The resolver library was also extended to allow IPv6 as the transport
3688 protocol for the requests. This work was done by Stig Venaas.
3690 * Compatibility code for K&R C compilers has been removed from the
3691 header files. A ISO C compiler is needed to use the library
3692 (conforming to either C89 or C99 standard).
3694 * Complete rewrite of the localedef program to support multibyte character
3695 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3696 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3697 Implemented by Ulrich Drepper.
3699 Bruno Haible significantly improved the generation and use of the data
3700 structures for the wide character tables.
3702 * Plural handling in gettext implemented by Ulrich Drepper.
3704 * The utmp daemon has been removed.
3706 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3708 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3711 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3713 * POSIX barriers implemented by Kaz Kylheku.
3715 * POSIX spawn function family implemented by Ulrich Drepper.
3717 * POSIX spinlocks are now available.
3719 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3721 * the configure option --enable-kernel=X.Y.Z allows to strip out
3722 compatibility for kernel versions before X.Y.Z. This is currently only
3723 implemented for Linux.
3725 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3726 field sin6_scope_id. This means that all programs using IPv6 should be
3727 recompiled. Don't expect binary compatibility with previous glibc
3730 * various conversion modules for IBM character sets contributed by
3733 * IA-64 port by Jes Sorensen and HJ Lu.
3747 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3749 * New xdr functions are added; some rpc functions are now 64bit clean.
3751 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3753 * Fixed known incompatibilities with glibc 2.0.
3755 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3757 * Optimized a number of functions (especially the ELF dynamic loader).
3759 * Update timezone data files.
3761 * lots of charmaps corrections
3763 * some new locale definitions and charmaps
3768 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3769 conform to POSIX member type requirements. Symbol versions have been
3770 adjusted as needed within the library, and for direct use by applications,
3771 but there is potential for problems if third-party libraries use
3772 struct timeval as part of their interface. This does not present
3773 a problem for X and other "essential" system libraries.
3775 * An additional locale model to support C++ Standard Library locale
3776 model and probably more was implemented by Ulrich Drepper.
3778 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3781 * Miles Bader provided the `argp' function family to support hierachical
3782 command line argument parsing, layered on top of getopt.
3784 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3786 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3789 * scanf recognizes the %a and %A format for scanning floating point numbers.
3791 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3792 information and interfaces for the available integer types.
3794 * about 130 new math functions were added to implement the ISO C9x math
3797 * the new header <complex.h> contains definitions of the complex math
3798 functions from ISO C 9X.
3800 * the new header <tgmath.h> defines generic macros to use complex or
3801 real valued functions.
3803 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3805 * Andreas Jaeger provided a test suite for the math library.
3807 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3809 * Ulrich Drepper added character set conversion functions (iconv).
3811 * Optimized string functions have been added.
3813 * The localedata addon is now part of glibc.
3815 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3817 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3818 daemon for NSS (nscd).
3820 Missing a better place here are some numbers on improvements. Under
3821 Linux 2.1.125 un-tar-ing the kernel sources takes
3825 using local files 12.19s 6.88s 22.91s
3827 using NIS 13.92s 8.91s 26.34s
3829 using NIS & nscd 10.37s 7.34s 25.30s
3831 using NIS+ 27.57s 30.37s 640.46s
3833 using NIS+ & nscd 10.25s 7.83s 26.51s
3835 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3837 Keep in mind that non-namelookup related operations dominate above times.
3838 It was just a common complain that using NIS+ unpacking the kernel is
3841 [1] The old nscd implementation is not available anymore since it was
3842 distributed with glibc up to version 2.0.98 and thus is now replaced.
3844 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3846 * Mark Kettenis provided a Hesiod NSS module.
3848 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3849 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3851 * rcmd can now handle netgroups (Dick Streefland).
3853 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3856 * Support for the IPv6 protocol has been added to the socket API, as per the
3857 latest draft standards.
3859 * Support for Linux 2.2 has been added.
3861 * Interface changes relative to the latest 2.0.x release:
3862 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3863 addseverity NEW: Unix98
3864 alphasort64 NEW: LFS
3865 argp_err_exit_status NEW: argp, GNU ext
3866 argp_error NEW: argp, GNU ext
3867 argp_failure NEW: argp, GNU ext
3868 argp_help NEW: argp, GNU ext
3869 argp_parse NEW: argp, GNU ext
3870 argp_program_bug_address NEW: argp, GNU ext
3871 argp_program_version NEW: argp, GNU ext
3872 argp_program_version_hook NEW: argp, GNU ext
3873 argp_state_help NEW: argp, GNU ext
3874 argp_usage NEW: argp, GNU ext
3875 authdes_create NEW: Secure RPC
3876 authdes_getucred NEW: Secure RPC
3877 authdes_pk_create NEW: Secure RPC
3878 backtrace NEW: GNU ext.
3879 backtrace_symbols NEW: GNU ext.
3880 backtrace_symbols_fd NEW: GNU ext.
3882 cacosf NEW: ISO C 9x
3883 cacosh NEW: ISO C 9x
3884 cacoshf NEW: ISO C 9x
3885 cacoshl NEW: ISO C 9x
3886 cacosl NEW: ISO C 9x
3893 casinf NEW: ISO C 9x
3894 casinh NEW: ISO C 9x
3895 casinhf NEW: ISO C 9x
3896 casinhl NEW: ISO C 9x
3897 casinl NEW: ISO C 9x
3899 catanf NEW: ISO C 9x
3900 catanh NEW: ISO C 9x
3901 catanhf NEW: ISO C 9x
3902 catanhl NEW: ISO C 9x
3903 catanl NEW: ISO C 9x
3904 cbc_crypt NEW: Secure RPC
3908 ccoshf NEW: ISO C 9x
3909 ccoshl NEW: ISO C 9x
3915 cimagf NEW: ISO C 9x
3916 cimagl NEW: ISO C 9x
3917 clearerr_locked REMOVED
3918 clntunix_create NEW: sunrpc ext
3920 clog10 NEW: ISO C 9x
3921 clog10f NEW: ISO C 9x
3922 clog10l NEW: ISO C 9x
3932 cprojf NEW: ISO C 9x
3933 cprojl NEW: ISO C 9x
3935 crealf NEW: ISO C 9x
3936 creall NEW: ISO C 9x
3941 csinhf NEW: ISO C 9x
3942 csinhl NEW: ISO C 9x
3945 csqrtf NEW: ISO C 9x
3946 csqrtl NEW: ISO C 9x
3950 ctanhf NEW: ISO C 9x
3951 ctanhl NEW: ISO C 9x
3953 des_setparity NEW: Secure RPC
3954 ecb_crypt NEW: Secure RPC
3955 endutxent NEW: Unix98
3957 exp10f NEW: ISO C 9x
3958 exp10l NEW: ISO C 9x
3962 fattach NEW: STREAMS
3963 fdetach NEW: STREAMS
3967 feclearexcept NEW: ISO C 9x
3968 fegetenv NEW: ISO C 9x
3969 fegetexceptflag NEW: ISO C 9x
3970 fegetround NEW: ISO C 9x
3971 feholdexcept NEW: ISO C 9x
3973 feraiseexcept NEW: ISO C 9x
3974 ferror_locked REMOVED
3975 fesetenv NEW: ISO C 9x
3976 fesetexceptflag NEW: ISO C 9x
3977 fesetround NEW: ISO C 9x
3978 fetestexcept NEW: ISO C 9x
3979 feupdateenv NEW: ISO C 9x
3980 fflush_locked REMOVED
3984 fgets_unlocked NEW: GNU ext.
3985 fileno_locked REMOVED
3997 fputc_locked REMOVED
3998 fputs_unlocked NEW: GNU ext.
3999 fread_unlocked NEW: GNU ext.
4004 fstatvfs NEW: Unix98
4008 ftruncate64 NEW: LFS
4010 fwrite_unlocked NEW: GNU ext.
4011 gai_strerror NEW: IPv6
4015 getchar_locked REMOVED
4017 getdate_err NEW: Unix98
4018 getdate_r NEW: GNU ext.
4020 getnameinfo NEW: IPv6
4021 getnetname NEW: Secure RPC
4022 getpmsg NEW: STREAMS
4023 getpt NEW: Unix98 PTY
4024 getrlimit64 NEW: LFS
4025 getutxent NEW: Unix98
4026 getutxid NEW: Unix98
4027 getutxline NEW: Unix98
4028 glob64 NEW: GNU ext.
4029 globfree64 NEW: GNU ext.
4030 gnu_get_libc_release NEW: GNU ext.
4031 gnu_get_libc_version NEW: GNU ext.
4032 grantpt NEW: Unix98 PTY
4033 host2netname NEW: Secure RPC
4035 iconv_close NEW: iconv
4036 iconv_open NEW: iconv
4037 if_freenameindex NEW: IPv6
4038 if_indextoname NEW: IPv6
4039 if_nameindex NEW: IPv6
4040 if_nametoindex NEW: IPv6
4041 in6addr_any NEW: IPv6
4042 in6addr_loopback NEW: IPv6
4043 inet6_isipv4mapped NEW: IPv6
4044 isastream NEW: STREAMS
4045 iswblank NEW: GNU ext.
4046 key_decryptsession NEW: Secure RPC
4047 key_decryptsession_pk NEW: Secure RPC
4048 key_encryptsession NEW: Secure RPC
4049 key_encryptsession_pk NEW: Secure RPC
4050 key_gendes NEW: Secure RPC
4051 key_get_conv NEW: Secure RPC
4052 key_secretkey_is_set NEW: Secure RPC
4053 key_setnet NEW: Secure RPC
4054 key_setsecret NEW: Secure RPC
4055 llrint NEW: ISO C 9x
4056 llrintf NEW: ISO C 9x
4057 llrintl NEW: ISO C 9x
4058 llround NEW: ISO C 9x
4059 llroundf NEW: ISO C 9x
4060 llroundl NEW: ISO C 9x
4065 lrintf NEW: ISO C 9x
4066 lrintl NEW: ISO C 9x
4067 lround NEW: ISO C 9x
4068 lroundf NEW: ISO C 9x
4069 lroundl NEW: ISO C 9x
4071 makecontext NEW: Unix98
4072 mempcpy NEW: GNU ext.
4075 modify_ldt NEW: kernel
4079 nearbyint NEW: ISO C 9x
4080 nearbyintf NEW: ISO C 9x
4081 nearbyintl NEW: ISO C 9x
4082 netname2host NEW: Secure RPC
4083 netname2user NEW: Secure RPC
4084 nexttoward NEW: ISO C 9x
4085 nexttowardf NEW: ISO C 9x
4086 nexttowardl NEW: ISO C 9x
4090 passwd2des NEW: Secure RPC
4092 pow10f NEW: GNU ext.
4093 pow10l NEW: GNU ext.
4096 printf_size NEW: GNU ext.
4097 printf_size_info NEW: GNU ext.
4098 profil_counter REMOVED
4099 pthread_mutexattr_getkind_np REPLACED
4100 pthread_mutexattr_setkind_np REPLACED
4101 ptsname NEW: Unix98 PTY
4102 ptsname_r NEW: Unix98 PTY
4104 putchar_locked REMOVED
4105 putgrent NEW: GNU ext.
4107 putpmsg NEW: STREAMS
4108 pututxline NEW: Unix98
4112 readdir64_r NEW: LFS
4113 remquo NEW: ISO C 9x
4114 remquof NEW: ISO C 9x
4115 remquol NEW: ISO C 9x
4117 roundf NEW: ISO C 9x
4118 roundl NEW: ISO C 9x
4120 scalbln NEW: ISO C 9x
4121 scalblnf NEW: ISO C 9x
4122 scalblnl NEW: ISO C 9x
4124 sendfile NEW: kernel
4125 setcontext NEW: Unix98
4126 setrlimit64 NEW: LFS
4127 setutxent NEW: Unix98
4129 sigignore NEW: Unix98
4130 sigqueue NEW: POSIX.1b
4131 sigrelse NEW: Unix98
4132 sigset NEW: POSIX.1b
4133 sigtimedwait NEW: POSIX.1b
4134 sigwaitinfo NEW: POSIX.1b
4135 sincos NEW: GNU ext.
4136 sincosf NEW: GNU ext.
4137 sincosl NEW: GNU ext.
4141 strcasestr NEW: GNU ext.
4142 strtoimax NEW: ISO C 9x
4143 strtoumax NEW: ISO C 9x
4144 strverscmp NEW: GNU ext.
4145 svcauthdes_stats NEW: Secure RPC
4146 svcunix_create NEW: sunrpc etx
4147 svcunixfd_create NEW: sunrpc ext
4148 swapcontext NEW: Unix98
4149 tcgetsid NEW: Unix98 PTY
4150 tdestroy NEW: GNU ext.
4151 tgamma NEW: ISO C 9x
4152 tgammaf NEW: ISO C 9x
4153 tgammal NEW: ISO C 9x
4157 truncf NEW: ISO C 9x
4158 truncl NEW: ISO C 9x
4160 unlockpt NEW: Unix98 PTY
4161 updwtmpx NEW: Unix98
4162 user2netname NEW: Secure RPC
4163 utmpxname NEW: Unix98
4164 versionsort NEW: GNU ext.
4165 versionsort64 NEW: GNU ext.
4167 wcscasecmp NEW: GNU ext.
4168 wcsncasecmp NEW: GNU ext.
4169 wcsnlen NEW: GNU ext.
4170 wcstoimax NEW: ISO C 9x
4171 wcstoll NEW: ISO C 9x
4172 wcstoull NEW: ISO C 9x
4173 wcstoumax NEW: ISO C 9x
4175 wordexp NEW: POSIX.2
4176 wordfree NEW: POSIX.2
4177 write_profiling REMOVED
4178 xdecrypt NEW: Secure RPC
4179 xdr_authdes_cred NEW: Secure RPC
4180 xdr_authdes_verf NEW: Secure RPC
4181 xdr_cryptkeyarg NEW: Secure RPC
4182 xdr_cryptkeyarg2 NEW: Secure RPC
4183 xdr_cryptkeyres NEW: Secure RPC
4184 xdr_getcredres NEW: Secure RPC
4185 xdr_key_netstarg NEW: Secure RPC
4186 xdr_key_netstres NEW: Secure RPC
4187 xdr_keybuf NEW: Secure RPC
4188 xdr_keystatus NEW: Secure RPC
4189 xdr_netnamestr NEW: Secure RPC
4190 xdr_sizeof NEW: Secure RPC
4191 xdr_unixcred NEW: sunrpc ext
4192 xencrypt NEW: Secure RPC
4193 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4204 * inet_ntoa is thread-safe
4206 * updwtmp is moved from libutil to libc
4208 * rewrite of cbrt function
4210 * update of timezone data
4224 * add atoll function
4226 * fix complex problems in Berkeley DB code
4228 * fix math functions
4232 * fixed lots of header problems (especially Linux/GNU specific)
4234 * dynamic loader preserves all registers
4236 * Roland McGrath provided support for handling of auxiliary objects in
4237 the ELF dynamic loader.
4239 * support for parallel builds is improved
4243 * GNU extensions are no longer declared by default. To enable them you
4244 must define the macro `_GNU_SOURCE' in your program or compile with
4247 * The library has changed from using GNU ld symbol aliases to using weak
4248 symbols where available. The ELF object file format supports weak
4249 symbols; GNU ld also supports weak symbols in the a.out format. (There
4250 is also now support for other GNU ld extensions in ELF. Use the
4251 `--with-elf' option to configure to indicate you have ELF, and
4252 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4253 of many files which contained only symbol aliases, reducing the size of
4254 the source and the compiled library; many other files were renamed to
4255 less cryptic names previously occupied by the symbol alias files.
4256 There is a new header file <elf.h> for programs which operate on
4257 files in the ELF format.
4259 * Converted to Autoconf version 2, so `configure' has more options.
4260 Run `configure --help' to see the details.
4262 * The library can now be configured to build profiling, highly-optimized
4263 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4264 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4265 `configure' enable building these extra libraries. The shared library is
4266 built by default when using both ELF and GNU ld. When shared libraries
4267 are enabled, the new library `-ldl' is available for arbitrary run-time
4268 loading of shared objects; its interface is defined in <dlfcn.h>. The
4269 new header file <link.h> gives access to the internals of the run-time
4270 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4271 application of same name on other systems and it provides information
4272 about dynamically linked binaries.
4274 * The C library now provides the run-time support code for profiling
4275 executables compiled with `-pg'. Programs can control the profiling code
4276 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4277 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4278 the support for this file format was contributed by David Mosberger-Tang.
4280 * The math code has been replaced with a math library based on fdlibm from
4281 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4282 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4283 functions. The math functions now reside in a separate library, so
4284 programs using them will need to use `-lm' their linking commands.
4286 * John C. Bowman contributed optimized ix87 assembler inline functions.
4288 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4289 mechanism similar to that found in Solaris 2. This is now used for the
4290 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4291 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4292 file controls what services are used for each individual database. This
4293 works by loading shared libraries with names specified in `nsswitch.conf',
4294 so service modules can be changed or added at any time without even
4295 relinking any program. Currently there are the file, db, and NIS based
4296 NSS services available.
4298 * The new functions `strtoq' and `strtouq' parse integer values from
4299 strings, like `strtol' and `strtoul', but they return `long long int' and
4300 `unsigned long long int' values, respectively (64-bit quantities).
4302 * The new functions `strtof' and `strtold' parse floating-point values from
4303 strings, like `strtod', but they return `float' and `long double' values,
4304 respectively (on some machines `double' and `long double' are the same).
4306 * Ulrich Drepper has contributed new implementations of the floating-point
4307 printing and reading code used in the `printf' family of functions and
4308 `strtod', `strtof', and `strtold'. These new functions are perfectly
4309 accurate, and much faster than the old ones.
4311 * The implementation of the POSIX locale model was completely rewritten by
4312 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4313 to compile the POSIX locale definition.
4315 * The former dummy implementations of the strcoll and strxfrm function are
4316 now replaced by fully functional code contributed by Ulrich Drepper. The
4317 collation information comes from the POSIX locale definitions.
4319 * The new header <langinfo.h> defines an interface for accessing
4320 various locale-dependent data (using the locale chosen with `setlocale').
4322 * Ulrich Drepper has contributed a new suite of functions for operation on
4323 wide-character and multibyte-character strings, in <wchar.h>;
4324 and classification and case conversion of wide characters, in <wctype.h>.
4325 These new functions are conforming to the ISO C, Amendement 1 specification.
4327 * There is now a second implementation of the standard I/O library available.
4328 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4329 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4331 * You can now use positional parameter specifications in format strings
4332 for the `printf' and `scanf' families of functions. For example,
4333 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4334 ``Number 6, Mr Jones''. This is mainly useful when providing different
4335 format strings for different languages, whose grammars may dictate
4336 different orderings of the values being printed. To support this
4337 feature, the interface for `register_printf_handler' has changed; see
4338 the header file <printf.h> for details.
4340 * The `printf' and `scanf' families of functions now understand a new
4341 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4342 to group numbers as indicated by the locale; for `scanf' and friends, this
4343 says to accept as valid only a number with all the proper grouping
4344 separators in the right places. In the default "C" locale, numbers are
4345 not grouped; but locales for specific countries will define the usual
4346 conventions (i.e. separate thousands with `,' in the US locale).
4348 * The pgrp functions have been regularized, slightly incompatibly but much
4349 less confusingly. The core functions are now `getpgid' and `setpgid',
4350 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4351 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4352 provided for compatibility. There is no longer an incompatible `getpgrp'
4353 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4355 * The new header file <fts.h> and suite of functions simplify programs that
4356 operate on directory trees. This code comes from 4.4 BSD.
4358 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4359 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4360 NSS scheme used in glibc.
4362 * There is a new malloc debugging hook `__memalign_hook'.
4364 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4365 `unsigned int' in <sys/types.h>. These are for compatibility only and
4366 their use is discouraged.
4368 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4369 done differently, so that it works even without GNU ld.
4371 * New function `euidaccess' checks allowed access to a file like `access',
4372 but using the effective IDs instead of the real IDs.
4374 * The time zone data files have been updated for the latest and greatest
4375 local time conventions of the countries of the world.
4377 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4380 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4381 for formatting floating-point numbers. They are provided only for
4382 compatibility; new programs should use `sprintf' instead. There are
4383 also equivalent function for the `long double' floating-point type and
4384 all functions also exist in a reentrant form.
4386 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4387 functions for maintaining the login-record files (primarily of use to
4388 system programs such as `login'), and convenient functions for
4389 allocating and initializing a pseudo-terminal (pty) device.
4391 * Ulrich Drepper has contributed new support for System V style
4392 shared memory and IPC on systems that support it.
4394 * Ulrich Drepper has contributed several miscellaneous new functions found
4395 in System V: The `hsearch' family of functions provide an effective
4396 implementation of hash tables; `a64l' and `l64a' provide a very simple
4397 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4400 * Ulrich Drepper has contributed new reentrant counterparts for the
4401 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4403 * Ulrich Drepper has contributed new, highly-optimized versions of several
4404 string functions for the i486/Pentium family of processors.
4406 * Ulrich Drepper has updated the Linux-specific code, based largely
4407 on work done in Hongjiu Lu's version of GNU libc for Linux.
4408 The GNU library now supports Linux versions 2.0.10 and later,
4409 using the ELF object file format (i[3456]86-*-linux).
4411 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4413 * David Mosberger-Tang and Richard Henderson have ported the C library
4414 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4415 dynamic linking support for ELF/Alpha.
4417 * Richard Henderson contributed several Alpha optimized assembler function
4418 for arithmetic and string handling.
4420 * Ulrich Drepper has contributed a new set of message catalog functions to
4421 support multiple languages using the <libintl.h> interface, for use with
4422 his new package GNU gettext. Translation volunteers have contributed
4423 catalogs of the library's messages in Spanish, German, and Korean.
4425 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4426 program and the `catgets' function for reading the catalog files it
4427 creates. (The <libintl.h> interface is preferred; we include the
4428 <nl_types.h> interface using `catgets' only for source compatibility with
4429 programs already written to use it.)
4431 * New header file <values.h> gives SVID-compatible names for <limits.h>
4434 * Various new macros, declarations, and small header files for compatibility
4437 * New function `group_member' is a convenient way to check if a process has
4438 a given effective group ID.
4440 * When using GCC 2.7 and later, the socket functions are now declared in a
4441 special way so that passing an argument of type `struct sockaddr_in *',
4442 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4443 `struct sockaddr *' type, does not generate a type-clash warning.
4445 * New function `error' declared in header file <error.h> is a convenient
4446 function for printing error messages and optionally exiting; this is the
4447 canonical function used in GNU programs. The new functions `err', `warn',
4448 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4449 doing the same thing.
4451 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4452 POSIX.2 `glob' function to do ~ and {...} expansion.
4454 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4455 `clearenv' which is used in POSIX.9 is also available.
4457 * New function `getsid' returns session ID number on systems that support it.
4459 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4460 files <db.h> and <mpool.h> provide a rich set of functions for several
4461 types of simple databases stored in memory and in files, and <ndbm.h> is
4462 an old `ndbm'-compatible interface using the `db' functions. Link with
4463 `-ldb' to get these functions.
4465 * New macro `strdupa' copies a string like `strdup', but uses local stack
4466 space from `alloca' instead of dynamic heap space from `malloc'.
4468 * New function `strnlen' is like `strlen' but searches only a given maximum
4469 number of characters for the null terminator. `stpncpy', `strndup' and
4470 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4473 * New function `statfs' in header <sys/statfs.h>.
4475 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4476 provide convenient functions for operating on blocks of null-terminated
4479 * A new suite of functions in <utmp.h> handle all the details of reading
4480 and writing the utmp file.
4482 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4485 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4486 completely POSIX compliant and also implemented the extended functionality
4487 to handle alternate digit representation and alternate era date formats.
4489 * Ulrich Drepper provided an implementation of the `strptime' function
4490 defined in XPG4.2 which transforms a string into a `struct tm' value.
4492 * Paul Eggert provided the tzselect shell script as part of the timezone
4493 code. The shell script makes it easy to select the correct timezone
4496 * The implementation of the malloc family of functions is completely replaced
4497 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4498 The implementation uses the mmap function (if available) and it is
4499 optimized for the use in multi threaded programs.
4501 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4502 functions. This new functionality is usable by specifying a special
4503 salt string and it is compatible with implementation on *BSD systems.
4505 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4506 `getsubopt' to handle second level command line options, `bsd_signal'
4507 to access BSD style `signal' functionality, the obsolete `regexp' style
4510 * the `lchown' function is available on system which support this
4513 * The implementation of the shadow password handling function was contributed
4516 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4518 * POSIX.1g support was added. The <sys/select.h> header is available,
4519 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4520 implementation of `getaddrinfo'.
4524 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4526 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4527 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4529 * The new function `daemon' from 4.4 BSD is useful for server programs that
4530 want to put themselves in the background.
4532 * Joel Sherrill has contributed support for several standalone boards that
4533 run without an operating system.
4535 * `printf', `scanf' and friends now accept a `q' type modifier for long
4536 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4538 * All of the code taken from BSD (notably most of the math and networking
4539 routines) has been updated from the BSD 4.4-Lite release.
4541 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4543 * The new functions `getdomainname' and `setdomainname' fetch or change the
4544 YP/NIS domain name. These are system calls which exist on systems which
4547 * The time zone data files have been updated for the latest international
4550 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4551 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4555 * The C library now includes support for Sun RPC, from Sun's free
4556 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4557 are included. (There is still no support for YP.)
4559 * Tom Quinn has contributed a port of the C library to SGI machines running
4560 Irix 4 (mips-sgi-irix4).
4562 * The new `lockf' function is a simplified interface to the locking
4563 facilities of `fcntl', included for compatibility.
4565 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4567 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4569 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4572 * The error message printed by `assert' for a failed assertion now includes
4573 the name of the program (if using GNU ld) and the name of the calling
4574 function (with versions of GCC that support this).
4576 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4578 * The library now includes the <sys/mman.h> header file and memory
4579 management functions `mmap', `munmap', `mprotect', `msync', and
4580 `madvise', on systems that support those facilities.
4582 * The interface for `mcheck' has changed slightly: the function called to
4583 abort the program when an allocation inconsistency is detected now takes
4584 an argument that indicates the type of failure. The new function
4585 `mprobe' lets you request a consistency check for a particular block at
4586 any time (checks are normally done only when you call `free' or `realloc'
4589 * It is now possible to easily cross-compile the C library, building on one
4590 system a library to run on another machine and/or operating system. All
4591 you need to do is set the variable `HOST_CC' in `configparms' to the
4592 native compiler for programs to run on the machine you are building on (a
4593 few generator programs are used on Unix systems); set `CC' to the
4596 * The new function `fexecve' (only implemented on the GNU system) executes
4597 a program file given a file descriptor already open on the file.
4601 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4602 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4604 * You can set the variable `libprefix' in `configparms' to specify a prefix
4605 to be prepended to installed library files; this makes it easy to install
4606 the GNU C library to be linked as `-lgnuc' or whatever.
4608 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4609 copies a limited number of characters from a string, and returns the
4610 address of the last character written.
4612 * You no longer need to check for whether the installed `stddef.h' is
4613 compatible with the GNU C library. configure now checks for you.
4615 * You can now define a per-stream `fileno' function to convert the
4616 stream's cookie into an integral file descriptor.
4618 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4619 allocates zero bytes of storage, and returns a unique pointer which
4620 you can pass to `realloc' or `free'. The behavior is undefined if
4621 you dereference this pointer.
4623 * The C library now runs on Sony NEWS m68k machines running either
4624 NewsOS 3 or NewsOS 4.
4626 * The new `syscall' function is a system-dependent primitive function
4627 for invoking system calls. It has the canonical behavior on Unix
4628 systems, including unreliable return values for some calls (such as
4629 `pipe', `fork' and `getppid').
4631 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4632 to `EAGAIN', which is the preferred name. On systems whose kernels
4633 use two distinct codes, the C library now translates EWOULDBLOCK to
4634 EAGAIN in every system call function.
4638 * The GNU C Library Reference Manual is now distributed with the library.
4639 `make dvi' will produce a DVI file of the printed manual.
4640 `make info' will produce Info files that you can read on line using C-h i
4641 in Emacs or the `info' program.
4642 Please send comments on the manual to bug-glibc-manual@gnu.org.
4644 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4646 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4648 * Jason Merrill has contributed a port to the Sequent Symmetry running
4649 Dynix version 3 (i386-sequent-dynix).
4651 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4652 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4654 * New function `memory_warnings' lets you arrange to get warnings when
4655 malloc is running out of memory to allocate, like Emacs gives you.
4657 * The C library now contains the relocating allocator used in Emacs 19 for
4658 its editing buffers. This allocator (ralloc) minimizes allocation
4659 overhead and fragmentation by moving allocated regions around whenever it
4660 needs to. You always refer to a ralloc'd region with a "handle" (a
4661 pointer to a pointer--an object of type `void **').
4663 * There is a new `printf' format: `%m' gives you the string corresponding
4664 to the error code in `errno'.
4666 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4667 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4668 pass, the `a' modifier says to fill in a `char **' you pass with a
4671 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4672 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4673 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4675 * `mkstemp' is a traditional Unix function to atomically create and open a
4676 uniquely-named temporary file.
4680 * The standard location for the file that says what the local timezone is
4681 has changed again. It is now `/usr/local/etc/localtime' (or more
4682 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4684 * The distribution no longer contains any files with names longer than 14
4687 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4688 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4690 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4694 * The configuration process has changed quite a bit. The `configure'
4695 script is now used just like the configuration scripts for other GNU
4696 packages. The `sysdeps' directory hierarchy is much rearranged.
4697 The file `INSTALL' explains the new scheme in detail.
4699 * The header files no longer need to be processed into ANSI C and
4700 traditional C versions. There is just one set of files to install, and
4701 it will work with ANSI or old C compilers (including `gcc -traditional').
4703 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4704 MIPS DECStation running Ultrix 4.
4706 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4707 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4708 made itself into a shared library.
4710 * Yet further improved support for the i386, running 4.3 BSD-like systems
4711 (such as Mach 3 with the Unix single-server), or System V.
4713 * New function `strncasecmp' to do case-insensitive string comparison
4714 with limited length.
4716 * New function `strsep' is a reentrant alternative to `strtok'.
4718 * New functions `scandir' and `alphasort' for searching directories.
4720 * New function `setenv' is a better interface to `putenv'.
4722 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4723 function for traversing a directory tree.
4725 * The GNU obstack package is now also part of the C library.
4726 The new function `open_obstack_stream' creates a stdio stream that
4727 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4728 formatted output directly to an obstack.
4730 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4731 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4733 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4735 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4736 things to your strings.
4738 * There are some new test programs: test-fseek, testmb, and testrand.
4740 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4741 These ports are not finished, but are a good starting place for really
4742 supporting those systems.
4744 * `/etc/localtime' is now the standard location for the file that says what
4745 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4746 This follows the general principle that `/etc' is the place for all local
4747 configuration files.
4749 * The C library header files now use `extern "C"' when used by the C++
4750 compiler, so the C library should now work with C++ code.
4752 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4753 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4756 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4757 function which uses the `merge sort' algorithm, and is said to be
4758 significantly faster than the old GNU `qsort' function. Merge sort is
4759 now the standard `qsort' function. The new algorithm can require a lot
4760 of temporary storage; so, the old sorting function is called when the
4761 required storage is not available.
4763 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4764 provides the Unix `crypt' function, plus some other entry points.
4766 * The code and header files taken from 4.4 BSD have been updated with the
4767 latest files released from Berkeley.
4769 ----------------------------------------------------------------------
4770 Copyright information:
4772 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4774 Permission is granted to anyone to make or distribute verbatim copies
4775 of this document as received, in any medium, provided that the
4776 copyright notice and this permission notice are preserved,
4777 thus giving the recipient permission to redistribute in turn.
4779 Permission is granted to distribute modified versions
4780 of this document, or of portions of it,
4781 under the above conditions, provided also that they
4782 carry prominent notices stating who last changed them.
4785 version-control: never