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 * Unicode 11.0.0 Support: Character encoding, character type info, and
13 transliteration tables are all updated to Unicode 11.0.0, using
14 generator scripts contributed by Mike FABIAN (Red Hat).
16 * <math.h> functions that round their results to a narrower type are added
17 from TS 18661-1:2014 and TS 18661-3:2015:
19 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
22 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
25 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
28 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
31 * Nominative and genitive month names are now supported for the following
32 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Scottish Gaelic,
33 Upper Sorbian, and Walloon. The following languages now support
34 abbreviated alternative month names: Catalan, Greek, and Kashubian.
36 * New locale added: Lower Sorbian (dsb_DE), including the support of the
37 nominative and genitive month names.
39 * Building and running on GNU/Hurd systems now works without out-of-tree
42 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
43 library if installed. libidn2 version 2.0.5 or later is recommended. If
44 libidn2 is not available, internationalized domain names are not encoded
45 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
46 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
47 with an encoding error.) Flags which used to change the IDN encoding and
48 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
49 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
50 deprecated. They no longer have any effect.
52 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
53 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
54 range of ELF gABI expressions including such constructs as
55 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
56 have been further restricted, and where in the past a dynamic string
57 token sequence may have been interpreted as a literal string it will
58 now cause a load failure. These load failures were always considered
59 unspecified behaviour from the perspective of the dynamic loader, and
60 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
61 results in a load failure now.
63 Deprecated and removed features, and other changes affecting compatibility:
65 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
66 installed. Software that was using either header should be updated to
67 use standard <stdio.h> interfaces instead.
69 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
70 This was never required by the C standard, and the macros just expanded
71 to call alternative names for the same functions. If you hoped getc and
72 putc would provide performance improvements over fgetc and fputc, instead
73 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
74 necessary, flockfile and funlockfile.
76 * All stdio functions now treat end-of-file as a sticky condition. If you
77 read from a file until EOF, and then the file is enlarged by another
78 process, you must call clearerr or another function with the same effect
79 (e.g. fseek, rewind) before you can read the additional data. This
80 corrects a longstanding C99 conformance bug. It is most likely to affect
81 programs that use stdio to read interactive input from a terminal.
84 * The macros 'major', 'minor', and 'makedev' are now only available from
85 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
86 headers that happen to include <sys/types.h>. These macros are rarely
87 used, not part of POSIX nor XSI, and their names frequently collide with
88 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
91 <sys/sysmacros.h> is a GNU extension. Portable programs that require
92 these macros should first include <sys/types.h>, and then include
93 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
95 * The tilegx*-*-linux-gnu configurations are no longer supported.
97 * The obsolete function ustat is no longer available to newly linked
98 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
99 function has been deprecated in favor of fstatfs and statfs.
101 * The obsolete function nfsservctl is no longer available to newly linked
102 binaries. This function was specific to systems using the Linux kernel
103 and could not usefully be used with the GNU C Library on systems with
104 version 3.1 or later of the Linux kernel.
106 * The obsolete function name llseek is no longer available to newly linked
107 binaries. This function was specific to systems using the Linux kernel
108 and was not declared in a header. Programs should use the lseek64 name
109 for this function instead.
111 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
112 getaddrinfo and getnameinfo functions have been deprecated. The behavior
113 previously selected by them is now always enabled.
115 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
116 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
117 restriction (rejecting '_' in host names, among other things) has been
118 removed, for increased compatibility with non-IDN name resolution.
120 * The fcntl function now have a Long File Support variant named fcntl64. It
121 is added to fix some Linux Open File Description (OFD) locks usage on non
122 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
123 fcntl and LFS support is handled transparently. Also for Linux, the OFD
124 locks act as a cancellation entrypoint.
126 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
127 ecb_crypt, and des_setparity are no longer available to newly linked
128 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
129 longer installed. These functions encrypted and decrypted data with the
130 DES block cipher, which is no longer considered secure. Software that
131 still uses these functions should switch to a modern cryptography library,
134 * Reflecting the removal of the encrypt and setkey functions above, the
135 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
136 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
139 * The obsolete function fcrypt is no longer available to newly linked
140 binaries. It was just another name for the standard function crypt,
141 and it has not appeared in any header file in many years.
143 * We have tentative plans to hand off maintenance of the passphrase-hashing
144 library, libcrypt, to a separate development project that will, we hope,
145 keep up better with new passphrase-hashing algorithms. We will continue
146 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
147 'crypt_r' should not need to change at all; however, distributions will
148 need to install <crypt.h> and libcrypt from a separate project.
150 In this release, if the configure option --disable-crypt is used, glibc
151 will not install <crypt.h> or libcrypt, making room for the separate
152 project's versions of these files. The plan is to make this the default
153 behavior in a future release.
155 Changes to build and runtime requirements:
157 [Add changes to build and runtime requirements here]
159 Security related changes:
161 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
162 been fixed by removing the glibc-internal IDNA implementation and using
163 the system-provided libidn2 library instead. Originally reported by Hanno
164 Böck and Christian Weisgerber.
166 CVE-2017-18269: An SSE2-based memmove implementation for the i386
167 architecture could corrupt memory. Reported by Max Horn.
169 CVE-2018-11236: Very long pathname arguments to realpath function could
170 result in an integer overflow and buffer overflow. Reported by Alexey
173 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
174 architecture could write beyond the target buffer, resulting in a buffer
175 overflow. Reported by Andreas Schwab.
177 The following bugs are resolved with this release:
179 [The release manager will add the list generated by
180 scripts/list-fixed-bugs.py just before the release.]
187 * The GNU C Library can now be compiled with support for building static
188 PIE executables (See --enable-static-pie in INSTALL). These static PIE
189 executables are like static executables but can be loaded at any address
190 and provide additional security hardening benefits at the cost of some
191 memory and performance. When the library is built with --enable-static-pie
192 the resulting libc.a is usable with GCC 8 and above to create static PIE
193 executables using the GCC option '-static-pie'. This feature is currently
194 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
195 aarch64 with binutils 2.30 or later.
197 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
198 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
201 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
203 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
205 * In order to support faster and safer process termination the malloc API
206 family of functions will no longer print a failure address and stack
207 backtrace after detecting heap corruption. The goal is to minimize the
208 amount of work done after corruption is detected and to avoid potential
209 security issues in continued process execution. Reducing shutdown time
210 leads to lower overall process restart latency, so there is benefit both
211 from a security and performance perspective.
213 * The abort function terminates the process immediately, without flushing
214 stdio streams. Previous glibc versions used to flush streams, resulting
215 in deadlocks and further data corruption. This change also affects
216 process aborts as the result of assertion failures.
218 * On platforms where long double has the IEEE binary128 format (aarch64,
219 alpha, mips64, riscv, s390 and sparc), the math library now implements
220 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
221 These are the same interfaces added in version 2.26 for some platforms where
222 this format is supported but is not the format of long double.
224 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
225 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
226 implements interfaces for that type, as defined by ISO/IEC TS
227 18661-3:2015. These are corresponding interfaces to those supported for
230 * The math library now implements interfaces for the _Float32, _Float64 and
231 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
232 corresponding interfaces to those supported for _Float128.
234 * glibc now implements the memfd_create and mlock2 functions on Linux.
236 * Support for memory protection keys was added. The <sys/mman.h> header now
237 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
240 * The copy_file_range function was added.
242 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
244 * The ldconfig utility now processes `include' directives using the C/POSIX
245 collation ordering. Previous glibc versions used locale-specific
246 ordering, the change might break systems that relied on that.
248 * Support for two grammatical forms of month names has been added.
249 In a call to strftime, the "%B" and "%b" format specifiers will now
250 produce the grammatical form required when the month is used as part
251 of a complete date. New "%OB" and "%Ob" specifiers produce the form
252 required when the month is named by itself. For instance, in Greek
253 and in many Slavic and Baltic languages, "%B" will produce the month
254 in genitive case, and "%OB" will produce the month in nominative case.
256 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
257 are all valid and will all accept any known form of month
258 name---standalone or complete, abbreviated or full. In a call to
259 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
260 the strings used by "%B" and "%b", respectively. New query
261 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
262 used by "%OB" and "%Ob", respectively.
264 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
265 define the strings for %OB and %Ob, respectively; these have the
266 same syntax as "mon" and "abmon". These arrays are optional; if they
267 are not provided then they have the same content as "mon" and "abmon",
270 These features are provided for locales which define "alt_mon" and/or
271 "ab_alt_mon" in their locale source data. This release includes such
272 alternative month name data for the following languages: Belarusian,
273 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
275 This feature is currently a GNU extension, but it is expected to
276 be added to the next revision of POSIX, and it is also already
277 available on some BSD-derived operating systems.
279 This feature will cause existing statically compiled applications
280 to fail to load locales and fall back to the builtin C/POSIX locales.
281 See notes below for other changes affecting compatibility.
283 * Support for the RISC-V ISA running on Linux has been added. This port
284 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
285 for the following ISA and ABI pairs:
291 Deprecated and removed features, and other changes affecting compatibility:
293 * Statically compiled applications attempting to load locales compiled for the
294 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
295 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
296 support for two grammatical forms of the month names, also extends the locale
297 data binary format. Static applications needing locale support must be
298 recompiled to match the runtime and data they are deployed with. In some
299 distributions there is an upgrade window where dynamically linked applications
300 may use a new library but the old locale data and also fall back to the
301 builtin C/POSIX locales; restarting the application process is sufficient to
304 * Support for statically linked applications which call dlopen is deprecated
305 and will be removed in a future version of glibc. Applications which call
306 dlopen need to be linked dynamically instead.
308 * Support for old programs which use internal stdio data structures and
309 functions is deprecated. This includes programs which use the C++ streams
310 provided by libstdc++ in GCC 2.95. Programs which use the internal
311 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
312 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
313 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
314 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
315 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
316 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
317 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
318 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
319 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
320 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
321 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
322 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
323 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
324 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
325 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
326 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
327 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
328 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
329 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
330 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
331 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
332 future version of glibc. Unlike other symbol removals, these old
333 applications will not be supported using compatibility symbols.
335 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
336 defined by <sys/ptrace.h>.
338 * libm no longer supports SVID error handling (calling a user-provided
339 matherr function on error) or the _LIB_VERSION variable to control error
340 handling. (SVID error handling and the _LIB_VERSION variable still work
341 for binaries linked against older versions of the GNU C Library.) The
342 libieee.a library is no longer provided. math.h no longer defines struct
343 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
344 TLOSS, PLOSS and HUGE.
346 * The libm functions pow10, pow10f and pow10l are no longer supported for
347 new programs. Programs should use the standard names exp10, exp10f and
348 exp10l for these functions instead.
350 * The mcontext_t type is no longer the same as struct sigcontext. On
351 platforms where it was previously the same, this changes the C++ name
352 mangling for interfaces involving this type.
354 * The add-ons mechanism for building additional packages at the same time as
355 glibc has been removed. The --enable-add-ons configure option is now
358 * The --without-fp configure option is now ignored. Whether hardware
359 floating-point instructions are used is now configured based on whether
360 the compiler used at configure time (without any options implied by a
361 --with-cpu= configure option) uses such instructions.
363 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
364 the specified string can be parsed as a domain name.
366 * In the malloc_info output, the <heap> element may contain another <aspace>
367 element, "subheaps", which contains the number of sub-heaps.
369 * The libresolv function p_secstodate is no longer supported for new
372 * The tilepro-*-linux-gnu configuration is no longer supported.
374 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
375 and will be removed in a future release. Software that is still using
376 either header should be updated to use standard <stdio.h> interfaces
379 libio.h was originally the header for a set of supported GNU extensions,
380 but they have not been maintained as such in many years, they are now
381 standing in the way of improvements to stdio, and we don't think there are
382 any remaining external users. _G_config.h was never intended for public
383 use, but predates the bits convention.
385 Changes to build and runtime requirements:
387 * bison version 2.7 or later is required to generate code in the 'intl'
390 Security related changes:
392 CVE-2009-5064: The ldd script would sometimes run the program under
393 examination directly, without preventing code execution through the
394 dynamic linker. (The glibc project disputes that this is a security
395 vulnerability; only trusted binaries must be examined using the ldd
398 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
399 suffered from a one-byte overflow during ~ operator processing (either
400 on the stack or the heap, depending on the length of the user name).
401 Reported by Tim Rühsen.
403 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
404 would sometimes fail to free memory allocated during ~ operator
405 processing, leading to a memory leak and, potentially, to a denial
408 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
409 without GLOB_NOESCAPE, could write past the end of a buffer while
410 unescaping user names. Reported by Tim Rühsen.
412 CVE-2017-17426: The malloc function, when called with an object size near
413 the value SIZE_MAX, would return a pointer to a buffer which is too small,
414 instead of NULL. This was a regression introduced with the new malloc
415 thread cache in glibc 2.26. Reported by Iain Buclaw.
417 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
418 to the allocation of too much memory. (This is not a security bug per se,
419 it is mentioned here only because of the CVE assignment.) Reported by
422 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
423 of the number of search path components. (This is not a security
424 vulnerability per se because no trust boundary is crossed if the fix for
425 CVE-2017-1000366 has been applied, but it is mentioned here only because
426 of the CVE assignment.) Reported by Qualys.
428 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
429 for AT_SECURE or SUID binaries could be used to load libraries from the
432 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
433 succeeds without returning an absolute path due to unexpected behaviour
434 of the Linux kernel getcwd syscall. Reported by halfdog.
436 CVE-2018-6485: The posix_memalign and memalign functions, when called with
437 an object size near the value of SIZE_MAX, would return a pointer to a
438 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
440 CVE-2018-6551: The malloc function, when called with an object size near
441 the value of SIZE_MAX, would return a pointer to a buffer which is too
442 small, instead of NULL.
444 The following bugs are resolved with this release:
446 [866] glob: glob should match dangling symlinks
447 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
448 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
450 [5997] math: Very slow execution of sinf function
451 [10580] localedata: hr_HR: updated locale
452 [10871] locale: 'mon' array should contain both nominative and genitive
454 [12349] localedata: eu_ES: incorrect thousands separator
455 [13605] localedata: shn_MM: new Shan locale
456 [13805] localedata: ru_RU: currency should use ',' as radix point
457 [13953] localedata: km_KH: locale update
458 [13994] localedata: mjw_IN: new locale
459 [14121] build: make writes .mo files in po directory
460 [14333] libc: Fix the race between atexit() and exit()
461 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
462 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
464 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
465 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
466 width Latin characters
467 [15332] localedata: es_CU: locale update
468 [15436] stdio: Don't close or flush stdio streams on abort
469 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
471 [16148] localedata: ca_ES: incorrect thousands separator
472 [16750] dynamic-link: ldd should not try to execute the binaries
474 [16777] localedata: pl_PL: incorrect thousands separator in locale
475 [16905] localedata: hanzi: new collation
476 [17563] localedata: cmn_TW: add hanzi collation
477 [17750] localedata: wrong collation order of diacritics in most locales
478 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
480 [17956] build: Build fails on missing definitions from header file
481 nss/nss.h when Mozilla NSS is used for cryptography
482 [18203] libc: realpath() does not handle unreachable paths correctly
483 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
484 [18812] localedata: kab_DZ: new Kabyle Algeria locale
485 [18822] libc: Internal functions are called via PLT
486 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
487 [19170] libc: __gmon_start__ defined in hppa in crtn.S
488 [19574] libc: glibc should support building static PIE binaries
489 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
491 [19971] glob: glob: Do not skip entries with zero d_ino values
492 [19982] localedata: fr.po: spelling mistake for error code EXDEV
493 [20008] localedata: km_KH: convert to translit_neutral
494 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
495 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
496 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
497 [20482] localedata: de_CH: abbreviated weekdays should be two letters
498 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
500 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
501 they're set, wrong errors returned by gaih_inet when lookup functions
503 [20756] localedata: [PATCH] Use Unicode wise thousands separator
504 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
506 [20952] localedata: yuw_PG: new locale
507 [21084] localedata: charmaps/IBM858: new codepage
508 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
509 [21242] libc: assert gives pedantic warning in old gcc versions
510 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
511 __regcall calling convention
512 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
513 [21326] libc: C99 functions are not declared for C++11 and later if
514 _GNU_SOURCE is not predefined by g++
515 [21457] libc: sys/ucontext.h namespace
516 [21530] libc: tmpfile() should be implemented using O_TMPFILE
517 [21660] math: GCC fails to compile a formula with tgmath.h
518 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
519 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
520 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
521 [21684] math: tgmath.h handling of complex integers
522 [21685] math: tgmath.h handling of bit-fields
523 [21686] math: tgmath.h handling of __int128
524 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
526 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
528 [21750] localedata: column width of characters incompatible with classical
530 [21754] malloc: malloc: Perform as little work as possible after heap
531 consistency check failures
532 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
533 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
534 [21791] string: Unused XXX_chk_XXX functions in libc.a
535 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
537 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
539 [21845] localedata: Added new Locale bho_NP
540 [21853] localedata: Fix abday Which looks same as day in zh_SG
541 [21854] localedata: Added New Locale en_SC
542 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
543 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
544 _dl_runtime_resolve_avx_slow
545 [21885] network: getaddrinfo: gethosts does not release resolver context
546 on memory allocation failure
547 [21899] libc: XPG4.2 sigaction namespace
548 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
550 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
552 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
553 errno value, causing getaddrinfo to fail
554 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
555 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
556 instead of EAI_NODATA
557 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
558 constant PTRACE_SEIZE_DEVEL
559 [21930] math: C-only gcc builtins used in <math.h> isinf
560 [21932] network: Unpaired __resolv_context_get in generic get*_r
562 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
563 [21944] libc: sigval namespace
564 [21951] localedata: Update hanzi collation by stroke
565 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
566 sysdeps/x86_64/fpu/e_expf.S
567 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
568 allocation in syscall loops)
569 [21959] localedata: Fix Country name for xh_ZA
570 [21960] localedata: Fix abmon for bem_ZM
571 [21966] math: AVX2 mathvec functions use FMA without checking
572 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
573 [21971] localedata: Added New Locale for mfe_MU
574 [21972] libc: assert macro requires operator== (int) for its argument type
575 [21973] math: [sparc] libm missing sqrtl compat symbol
576 [21974] libc: Remove __bb_init_func and __bb_exit_func
577 [21982] string: stratcliff.c: error: assuming signed overflow does not
579 [21986] stdio: __guess_grouping is called incorrectly
580 [21987] math: [sparc32] wrong bits/long-double.h installed
581 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
583 [22022] localedata: Missing country_name for mni_IN
584 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
586 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
588 [22026] locale: iconv_open: heap overflow on gconv_init failure
589 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
590 [22035] math: [m68k] bits/math-inline.h macro namespace
591 [22038] localedata: Fix abbreviated weeks and months for Somali
592 [22044] localedata: Remove redundant data for Limburgish Language
593 [22050] malloc: Linking with -lmcheck does not hook
594 __malloc_initialize_hook correctly
595 [22051] libc: zero terminator in the middle of glibc's .eh_frame
596 [22052] malloc: malloc failed to compile with GCC 7 and -O3
597 [22070] localedata: charmaps/UTF-8: wcwidth for
598 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
599 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
600 Jungseong and Jongseong) should be 0
601 [22078] nss: nss_files performance issue in multi mode
602 [22082] math: bits/math-finite.h exp10 condition
603 [22086] libc: pcprofiledump incorrect cross-endian condition
604 [22093] dynamic-link: ld.so no longer searches in .../x86_64
605 [22095] network: Name server address allocation memory leak in resolv.conf
607 [22096] network: __resolv_conf_attach can incorrectly free passed conf
609 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
610 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
612 [22111] malloc: malloc: per thread cache is not returned when thread exits
613 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
614 [22134] libc: [linux] implement fexecve with execveat
615 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
617 [22145] libc: ttyname() gives up too early in the face of namespaces
618 [22146] math: C++ build issue with float128 on x86_64
619 [22153] nptl: nptl: save error code before process termination
620 [22156] libc: [hppa,ia64,microblaze] Executable stack default
621 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
622 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
623 [22165] libc: [hppa] Text relocations in libc.so
624 [22180] libc: destructor registered via __cxa_atexit is called twice
625 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
626 symlinks") cause "make" segfaults
627 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
629 [22207] libc: FAIL: stdlib/test-atexit-race
630 [22225] math: nearbyint arithmetic moved before feholdexcept
631 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
632 [22235] math: iscanonical in C++ and float128
633 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
634 the svid compat wrapper
635 [22244] math: ynf and yn are wrong without the svid compat wrapper
636 [22273] libc: Improper assert in Linux posix_spawn implementation
637 [22284] libc: -pg -pie doesn't work
638 [22292] locale: localedef exits with error 4 when it should be error 1
639 [22294] locale: Allow "" for int_currency_symbol definition in locales.
640 [22295] locale: Don't warn on non-symbolic characters in locale sources in
642 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
643 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
645 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
646 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
647 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
648 [22322] libc: [mips64] wrong bits/long-double.h installed
649 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
650 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
652 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
653 [22343] malloc: Integer overflow in posix_memalign
654 [22347] libc: getrandom() returns the number of bytes that were copied to
655 the buffer even though the comments say "Return 0 on success and -1 on
657 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
658 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
660 [22370] dynamic-link: Incorrect note padding check
661 [22375] libc: malloc returns pointer from tcache_get when should return
662 NULL (CVE-2017-17426)
663 [22377] math: iseqsig, float128 and C++
664 [22382] localedata: Error in tpi_PG locale
665 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
666 inside the ASCII printable range
667 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
669 [22403] localedata: Slash needs escaping in some locales
670 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
671 [22409] network: res_hnok does not accept some host names used on the
673 [22412] network: res_dnok, res_hnok should perform syntax checks
674 [22413] network: ns_name_pton ignores syntactically invalid trailing
676 [22415] stdio: setvbuf can lead to invalid free/segfault
677 [22432] build: Non-deterministic build
678 [22439] malloc: malloc_info should compute summary statistics for all sub-
680 [22442] network: if_nametoindex could report index for the wrong
682 [22446] build: aliasing violation calling readlink in handle_request
683 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
684 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
685 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
686 __nscd_hash/__nss_hash
687 [22463] network: p_secstodate overflow handling
688 [22469] localedata: pl_PL LC_COLLATE does not use i18n
689 [22478] libc: sigwait can fail with EINTR
690 [22505] libc: ldconfig processes include directive in locale-specific
692 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
693 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
694 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
695 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
696 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
697 [22534] localedata: Collation rules for Serbian and Bosnian should be the
699 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
700 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
701 [22577] libc: missing newline after "cannot allocate TLS data structures
703 [22588] manual: manual/conf.texi: missing underscore in front of
705 [22593] math: nextafter and nexttoward are declared with const attribute
706 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
707 [22603] string: ia64 memchr overflows internal pointer check
708 [22605] libc: SH clone does not set the exit code correctly
709 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
711 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
712 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
713 modified in case of success
714 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
715 [22615] manual: manual: ambiguous wording about errno value in case of
717 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
719 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
720 binaries or if /proc is not mounted (CVE-2017-16997)
721 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
722 [22630] build: $(no-pie-ldflag) is no longer effective
723 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
724 [22635] nptl: pthread_self returns NULL before libpthread is loaded
725 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
726 [22637] nptl: guard size is subtracted from thread stack size instead of
728 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
729 [22657] localedata: hu_HU: Avoid double space in date
730 [22660] math: fmax, fmin sNaN handling on alpha
731 [22664] libc: New warning of GCC8
732 [22665] math: alpha: ceil and floor raise inexact exceptions
733 [22666] math: alpha: trunc raise inexact exceptions
734 [22667] libc: makecontext lacks stack alignment on i386
735 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
736 [22679] libc: getcwd(3) can succeed without returning an absolute path
738 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
740 [22687] math: [powerpc-nofpu] complex long double functions spurious
742 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
743 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
744 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
746 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
747 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
749 [22701] nis: Incomplete removal of libnsl
750 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
751 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
752 [22715] dynamic-link: FAIL: elf/tst-audit10
753 [22719] libc: Backtrace tests fail on hppa
754 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
755 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
756 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
757 before the first call to crypt_r () in crypt/badsalttest.c
764 * A per-thread cache has been added to malloc. Access to the cache requires
765 no locks and therefore significantly accelerates the fast path to allocate
766 and free small amounts of memory. Refilling an empty cache requires locking
767 the underlying arena. Performance measurements show significant gains in a
768 wide variety of user workloads. Workloads were captured using a special
769 instrumented malloc and analyzed with a malloc simulator. Contributed by
770 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
772 * Unicode 10.0.0 Support: Character encoding, character type info, and
773 transliteration tables are all updated to Unicode 10.0.0, using
774 generator scripts contributed by Mike FABIAN (Red Hat).
775 These updates cause user visible changes, especially the changes in
776 wcwidth for many emoji characters cause problems when emoji sequences
777 are rendered with pango, see for example:
778 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
780 * Collation of Hungarian has been overhauled and is now consistent with "The
781 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
784 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
786 - The GNU C Library will now detect when /etc/resolv.conf has been
787 modified and reload the changed configuration. The new resolver option
788 “no-reload” (RES_NORELOAD) disables this behavior.
790 - The GNU C Library now supports an arbitrary number of search domains
791 (configured using the “search” directive in /etc/resolv.conf);
792 previously, there was a hard limit of six domains. For backward
793 compatibility, applications that directly modify the ‘_res’ global
794 object are still limited to six search domains.
796 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
797 Library will now randomly pick a name server from the configuration as a
798 starting point. (Previously, the second name server was always used.)
800 * The tunables feature is now enabled by default. This allows users to tweak
801 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
803 * New function reallocarray, which resizes an allocated block (like realloc)
804 to the product of two sizes, with a guaranteed clean failure upon integer
805 overflow in the multiplication. Originally from OpenBSD, contributed by
806 Dennis Wölfing and Rüdiger Sonderfeld.
808 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
809 These are extended versions of preadv and pwritev, respectively, taking an
810 additional flags argument. The set of supported flags depends on the
811 running kernel; full support currently requires kernel 4.7 or later.
813 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
814 create a new session ID for the spawned process. This feature is
815 scheduled to be added to the next major revision of POSIX; for the time
816 being, it is available under _GNU_SOURCE.
818 * errno.h is now safe to use from C-preprocessed assembly language on all
819 supported operating systems. In this context, it will only define the
820 Exxxx constants, as preprocessor macros expanding to integer literals.
822 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
823 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
824 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
825 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
827 To compile programs that use this feature, the compiler must support
828 128-bit floating point with the type name _Float128 (as defined by TS
829 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
830 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
831 must be defined to make the new interfaces visible.
833 The new functions and macros correspond to those present for other
834 floating-point types (except for a few obsolescent interfaces not
835 supported for the new type), with F128 or f128 suffixes; for example,
836 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
837 printf or scanf formats for the new type; the strfromf128 and strtof128
838 interfaces should be used instead.
840 Deprecated and removed features, and other changes affecting compatibility:
842 * The synchronization that pthread_spin_unlock performs has been changed to
843 now be equivalent to a C11 atomic store with release memory order to the
844 spin lock's memory location. Previously, several (but not all)
845 architectures used stronger synchronization (e.g., containing what is
846 often called a full barrier). This change can improve performance, but
847 may affect odd fringe uses of spin locks that depend on the previous
848 behavior (e.g., using spin locks as atomic variables to try to implement
849 Dekker's mutual exclusion algorithm).
851 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
854 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
855 will only be built and installed when the GNU C Library is configured with
856 --enable-obsolete-rpc. This allows alternative RPC implementations, such
857 as TIRPC or rpcsvc-proto, to be used.
859 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
860 libnss_compat, are deprecated, and will not be built or installed by
863 The NIS(+) support library, libnsl, is also deprecated. By default, a
864 compatibility shared library will be built and installed, but not headers
865 or development libraries. Only a few NIS-related programs require this
866 library. (In particular, the GNU C Library has never required programs
867 that use 'gethostbyname' to be linked with libnsl.)
869 Replacement implementations based on TIRPC, which additionally support
870 IPv6, are available from <https://github.com/thkukuk/>. The configure
871 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
872 name service modules, to be built and installed.
874 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
875 support is enabled, the configured recursive resolver must support EDNS.
876 (Responding to EDNS-enabled queries with responses which are not
877 EDNS-enabled is fine, but FORMERR responses are not.)
879 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
880 servers have not supported this opcode for a long time.
882 * The _res_opcodes variable has been removed from libresolv. It had been
883 exported by accident.
885 * <string.h> no longer includes inline versions of any string functions,
886 as this kind of optimization is better done by the compiler. The macros
887 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
889 * The nonstandard header <xlocale.h> has been removed. Most programs should
890 use <locale.h> instead. If you have a specific need for the definition of
891 locale_t with no other declarations, please contact
892 libc-alpha@sourceware.org and explain.
894 * The obsolete header <sys/ultrasound.h> has been removed.
896 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
898 * The obsolete function cfree has been removed. Applications should use
901 * The stack_t type no longer has the name struct sigaltstack. This changes
902 the C++ name mangling for interfaces involving this type.
904 * The ucontext_t type no longer has the name struct ucontext. This changes
905 the C++ name mangling for interfaces involving this type.
907 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
908 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
909 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
910 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
911 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
912 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
913 fpregset_t type no longer has the name struct fpu. This changes the C++
914 name mangling for interfaces involving those types.
916 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
917 synced with the kernel:
919 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
920 are not supported on this architecture and have been removed.
922 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
923 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
924 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
926 Programs that assume the GET/SETREGS ptrace requests are universally
927 available will now fail to build, instead of malfunctioning at runtime.
929 Changes to build and runtime requirements:
931 * Linux kernel 3.2 or later is required at runtime, on all architectures
932 supported by that kernel. (This is a change from version 2.25 only for
935 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
937 * On most architectures, GCC 4.9 or later is required to build the GNU C
938 Library. On powerpc64le, GCC 6.2 or later is required.
940 Older GCC versions and non-GNU compilers are still supported when
941 compiling programs that use the GNU C Library. (We do not know exactly
942 how old, and some GNU extensions to C may be _de facto_ required. If you
943 are interested in helping us make this statement less vague, please
944 contact libc-alpha@sourceware.org.)
946 Security related changes:
948 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
949 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
951 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
952 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
954 * Avoid printing a backtrace from the __stack_chk_fail function since it is
955 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
958 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
959 fixed (CVE-2017-12133).
961 The following bugs are resolved with this release:
963 [984] network: Respond to changed resolv.conf in gethostbyname
964 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
965 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
967 [12189] libc: __stack_chk_fail should not attempt a backtrace
969 [14096] time: Race condition on timezone/tst-timezone.out
970 [14172] localedata: az_IR: new locale
971 [14995] build: glibc fails to build if gold is the default linker, even if
973 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
974 [16637] network: inet_pton function is accepting IPv6 with bad format
975 [16640] string: string/strtok.c: undefined behaviour inconsistent between
976 x86 and other generic code
977 [16875] localedata: ko_KR: fix lang_name
978 [17225] localedata: ar_SY: localized month names for May and June are
980 [17297] localedata: da_DK: wrong date_fmt string
981 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
982 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
983 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
984 [19066] localedata: ar_SA abbreviated day and month names are in English
985 [19569] network: resolv: Support an arbitrary number of search domains
986 [19570] network: Implement random DNS server selection in the stub
988 [19838] locale: localedef fails on PA-RISC
989 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
990 order of 0D36 and 0D37
991 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
993 [20098] libc: FAIL: debug/backtrace-tst on hppa
994 [20257] network: sunrpc: clntudp_call does not enforce timeout when
996 [20275] localedata: locale day/abday/mon/abmon should not have trailing
998 [20313] localedata: Update locale data to Unicode 9.0
999 [20424] manual: Document how to provide a malloc replacement
1000 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1002 [20686] locale: Add el_GR@euro to SUPPORTED.
1003 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1004 failures consistently
1005 [21015] dynamic-link: Document and fix --enable-bind-now
1006 [21016] nptl: pthread_cond support is broken on hppa
1007 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1008 frame-pointer on i386
1009 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1011 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1012 [21088] libc: Build fails with --enable-static-nss
1013 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1014 generic c code is used
1015 [21109] libc: Tunables broken on big-endian
1016 [21112] math: powf has large ulp errors with base close to 1 and exponent
1018 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1020 [21120] malloc: glibc malloc is incompatible with GCC 7
1021 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1023 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1024 (-0.0) when linking with -lieee
1025 [21171] math: log10, log2 and lgamma return incorrect results
1026 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1027 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1028 [21207] localedata: ce_RU: update weekdays from CLDR
1029 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1030 [21217] localedata: Update months from CLDR-31
1031 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1032 [21243] libc: support_delete_temp_file should issue warning for failed
1034 [21244] libc: support resolv_test_start() socket fd close should be
1036 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1037 new posix_spawn implementation
1038 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1039 leads to lower CPU frequency
1040 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1041 [21261] libc: [sparc64] bits/setjmp.h namespace
1042 [21267] network: [mips] bits/socket.h IOC* namespace
1043 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1044 [21270] libc: mmap64 silently truncates large offset values
1045 [21275] libc: posix_spawn always crashes on ia64 now
1046 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1047 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1048 [21289] libc: Incorrect declaration for 32-bit platforms with
1049 _FILE_OFFSET_BITS=64 causes build error
1050 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1051 not support gethostbyname4_r
1052 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1054 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1056 [21340] libc: Support POSIX_SPAWN_SETSID
1057 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1058 [21359] network: ns_name_pack needs additional byte in destination buffer
1059 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1060 against fragmentation attacks (CVE-2017-12132)
1061 [21369] network: resolv: Remove EDNS fallback
1062 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1064 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1065 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1066 [21393] stdio: Missing dup3 error check in freopen, freopen64
1067 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1068 [21399] localedata: Bad description for U00EC in
1069 localedata/charmaps/CP1254
1070 [21411] malloc: realloc documentation error
1071 [21426] network: sys/socket.h uio.h namespace
1072 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1073 [21445] libc: signal.h bsd_signal namespace
1074 [21455] network: Network headers stdint.h namespace
1075 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1077 [21475] network: resolv: Overlong search path is truncated mid-label
1078 [21511] libc: sigstack namespace
1079 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1080 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1081 bits/syscall.h: No such file or directory
1082 [21517] libc: struct sigaltstack namespace
1083 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1084 [21533] localedata: Update locale data to Unicode 10.0
1086 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1087 junk at end of line, first unrecognized character is `@'
1088 [21538] libc: SIG_HOLD missing for XPG4
1089 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1090 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1091 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1092 [21543] libc: sigevent namespace
1093 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1094 size for general purpose registers in mcontext_t structure
1095 [21550] libc: sigwait namespace
1096 [21552] libc: XPG4 bsd_signal namespace
1097 [21554] libc: sigpause namespace
1098 [21560] libc: sys/wait.h signal.h namespace
1099 [21561] libc: waitid namespace
1100 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1101 '/usr/include/stdlib.h': Permission denied
1102 [21575] libc: sys/wait.h missing struct rusage definition
1103 [21584] libc: sigaltstack etc namespace
1104 [21597] libc: siginterrupt namespace
1105 [21607] math: hppa: FAIL: math/test-tgmath
1106 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1108 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1109 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1110 stack and heap (CVE-2017-1000366)
1111 [21625] libc: wait3 namespace
1112 [21654] nss: Incorrect pointer alignment in NSS group merge result
1114 [21657] network: Parse interface zone id for node-local multicast
1115 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1117 [21666] libc: .symver is used on common symbol
1118 [21668] network: resolv: res_init cross-thread broadcast introduces race
1120 [21687] math: tgmath.h totalorder, totalordermag return type
1121 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1123 [21696] libc: Incorrect assumption of of __cpu_mask in
1124 posix/sched_cpucount.c
1125 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1126 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1128 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1129 leading to relocation crash
1130 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1131 `resolv_conf_matches (resp, conf)' failed.
1132 [21710] localedata: Added Samoan language locale for Samoa
1133 [21711] localedata: Pashto yesstr/nostr locale are missing
1134 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1135 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1136 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1137 [21723] localedata: yesstr/nostr missing for Chinese language locale
1138 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1140 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1142 [21728] localedata: New Locale for Tongan language
1143 [21729] localedata: incorrect LC_NAME fields for hi_IN
1144 [21733] localedata: yesstr and nostr are missing for zh_HK
1145 [21734] localedata: Missing yesstr and nostr are for kw_GB
1146 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1147 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1148 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1149 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1151 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1153 [21749] localedata: Wrong abbreviated day name (“abday”) for
1155 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1156 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1157 [21759] localedata: missing yesstr and nostr for Tigrinya
1158 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1159 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1160 [21767] localedata: Missing Bislama locales
1161 [21768] localedata: Missing yesstr and nostr for aa_ET
1162 [21770] localedata: Missing Field in li_NL
1163 [21778] nptl: Robust mutex may deadlock
1164 [21779] libc: MicroBlaze segfaults when loading libpthread
1165 [21783] localedata: Fix int_select international_call_prefixes
1166 [21784] localedata: Inconsistency in country_isbn
1167 [21788] localedata: Missing Country Postal Abbreviations
1168 [21794] localedata: Added-country_isbn-for-Italy
1169 [21795] localedata: Add/Fix country_isbn for France
1170 [21796] localedata: Added country_isbn for Republic of Korea
1171 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1172 [21799] localedata: Added int_select international_call_prefixes
1173 [21801] localedata: Added int_select international_call_prefixes
1174 [21804] nptl: Double semicolon in thread-shared-types.h
1175 [21807] localedata: LC_ADDRESS fix for pap_CW
1176 [21808] localedata: Fix LC_ADDRESS for pap_AW
1177 [21821] localedata: Added country_name in mai_IN
1178 [21822] localedata: Fix LC_TIME for mai_IN
1179 [21823] localedata: missing yesstr, nostr for sa_IN
1180 [21825] localedata: Fix name_mrs for mag_IN
1181 [21828] localedata: 2.26 changelog should mention user visible changes
1183 [21835] localedata: Added Maithili language locale for Nepal
1184 [21838] localedata: Removed redundant data for the_NP
1185 [21839] localedata: Fix LC_MONETARY for ta_LK
1186 [21844] localedata: Fix Latin characters and Months Sequence.
1187 [21848] localedata: Fix mai_NP Title Name
1192 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1193 24731-2:2010, is supported to enable declarations of functions from that
1194 TR. Note that not all functions from that TR are supported by the GNU C
1197 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1198 18661-1:2014, is supported to enable declarations of functions and macros
1199 from that TS. Note that not all features from that TS are supported by
1202 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1203 18661-4:2015, is supported to enable declarations of functions and macros
1204 from that TS. Note that most features from that TS are not supported by
1207 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1208 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1209 Since the GNU C Library defaults to a much newer revision of POSIX, this
1210 will only affect programs that specifically request an old conformance
1211 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1212 see a change in the visible declarations, but a program compiled with
1213 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1216 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1217 defined by all multithreaded code, but glibc has not required this for
1220 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1221 means that in a future release, the macros “major”, “minor”, and “makedev”
1222 will only be available from <sys/sysmacros.h>.
1224 These macros are not part of POSIX nor XSI, and their names frequently
1225 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1226 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1227 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1230 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1231 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1232 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1234 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1235 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1236 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1237 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1238 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1239 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1240 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1241 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1242 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1243 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1244 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1247 * New <math.h> features are added from TS 18661-1:2014:
1249 - Signaling NaN macros: SNANF, SNAN, SNANL.
1251 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1252 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1253 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1255 - llogb functions: the llogb, llogbf and llogbl functions, and the
1256 FP_LLOGB0 and FP_LLOGBNAN macros.
1258 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1261 - Comparison macros: iseqsig.
1263 - Classification macros: iscanonical, issubnormal, iszero.
1265 - Total order functions: totalorder, totalorderf, totalorderl,
1266 totalordermag, totalordermagf, totalordermagl.
1268 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1270 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1271 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1273 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1274 are added to libc. They convert a floating-point number into string.
1276 * Most of glibc can now be built with the stack smashing protector enabled.
1277 It is recommended to build glibc with --enable-stack-protector=strong.
1278 Implemented by Nick Alcock (Oracle).
1280 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1281 intended to be used instead of memset() to erase sensitive data after use;
1282 the compiler will not optimize out calls to explicit_bzero even if they
1283 are "unnecessary" (in the sense that no _correct_ program can observe the
1284 effects of the memory clear).
1286 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1287 to float instead of double. This does not affect the ABI of any libraries
1288 that are part of the GNU C Library, but may affect the ABI of other
1289 libraries that use this type in their interfaces.
1291 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1292 float_t and double_t types are now defined to long double instead of float
1293 and double. These options are not the default, and this does not affect
1294 the ABI of any libraries that are part of the GNU C Library, but it may
1295 affect the ABI of other libraries that use this type in their interfaces,
1296 if they are compiled or used with those options.
1298 * The getentropy and getrandom functions, and the <sys/random.h> header file
1301 * The buffer size for byte-oriented stdio streams is now limited to 8192
1302 bytes by default. Previously, on Linux, the default buffer size on most
1303 file systems was 4096 bytes (and thus remains unchanged), except on
1304 network file systems, where the buffer size was unpredictable and could be
1305 as large as several megabytes.
1307 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1308 for the Linux quota interface which predates kernel version 2.4.22 has
1311 * The malloc_get_state and malloc_set_state functions have been removed.
1312 Already-existing binaries that dynamically link to these functions will
1313 get a hidden implementation in which malloc_get_state is a stub. As far
1314 as we know, these functions are used only by GNU Emacs and this change
1315 will not adversely affect already-built Emacs executables. Any undumped
1316 Emacs executables, which normally exist only during an Emacs build, should
1317 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1319 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1320 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1321 “no-ip6-dotint” had already been the default, and support for the
1322 “ip6-dotint” option was removed from the Internet in 2006.
1324 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1325 flag from <resolv.h> have been removed. The option relied on a
1326 backwards-incompatible DNS extension which was never deployed on the
1329 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1330 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1331 They were already unimplemented.
1333 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1334 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1335 removed again from the IETF name lookup interface specification in RFC
1336 2553. Applications should use getaddrinfo instead.
1338 * DNSSEC-related declarations and definitions have been removed from the
1339 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1340 decode the data part of DNSSEC record types. Previous versions of glibc
1341 only implemented minimal support for the previous version of DNSSEC, which
1342 is incompatible with the currently deployed version.
1344 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1345 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1346 <arpa/nameser.h> header file because the distinction between RR types and
1347 meta-RR types is not officially standardized, subject to revision, and
1348 thus not suitable for encoding in a macro.
1350 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1351 and rhook members of the res_state type in <resolv.h> have been removed.
1352 The glibc stub resolver did not support these hooks, but the header file
1353 did not reflect that.
1355 * For multi-arch support it is recommended to use a GCC which has
1356 been built with support for GNU indirect functions. This ensures
1357 that correct debugging information is generated for functions
1358 selected by IFUNC resolvers. This support can either be enabled by
1359 configuring GCC with '--enable-gnu-indirect-function', or by
1360 enabling it by default by setting 'default_gnu_indirect_function'
1361 variable for a particular architecture in the GCC source file
1364 * GDB pretty printers have been added for mutex and condition variable
1365 structures in POSIX Threads. When installed and loaded in gdb these pretty
1366 printers show various pthread variables in human-readable form when read
1367 using the 'print' or 'display' commands in gdb.
1369 * Tunables feature added to allow tweaking of the runtime for an application
1370 program. This feature can be enabled with the '--enable-tunables' configure
1371 flag. The GNU C Library manual has details on usage and README.tunables has
1372 instructions on adding new tunables to the library.
1374 * A new version of condition variables functions have been implemented in
1375 the NPTL implementation of POSIX Threads to provide stronger ordering
1378 * A new version of pthread_rwlock functions have been implemented to use a more
1379 scalable algorithm primarily through not using a critical section anymore to
1382 Security related changes:
1384 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1385 have been created with makecontext could fail to terminate due to a
1386 missing .cantunwind annotation. This has been observed to lead to a hang
1387 (denial of service) in some Go applications compiled with gccgo. Reported
1388 by Andreas Schwab. (CVE-2016-6323)
1390 * The DNS stub resolver functions would crash due to a NULL pointer
1391 dereference when processing a query with a valid DNS question type which
1392 was used internally in the implementation. The stub resolver now uses a
1393 question type which is outside the range of valid question type values.
1396 The following bugs are resolved with this release:
1398 [4099] stdio: Overly agressive caching by stream i/o functions.
1399 [7065] build: Support building glibc with -fstack-protector or -fstack-
1401 [9842] localedata: en_CA: incorrect date format
1402 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1403 before it started waiting
1404 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1405 [15765] nptl: sem_open is wrongly a cancellation point
1406 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1407 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1409 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1410 library linked with pthread
1411 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1412 "FIXME: Ingo" issue)
1413 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1414 [17252] libc: getrandom and getentropy syscall
1415 [17730] malloc: thread-local storage is sometimes improperly free()'d
1416 after being __libc_memalign()'d
1417 [18241] stdio: failed fseek on memstream does not set errno and can fail
1419 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1421 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1422 PTHREAD_PRIO_INHERIT mutex on ARM
1423 [18784] network: res_query and related function crash for special record
1424 type queries (CVE-2015-5180)
1425 [19380] math: strtod does not raise "inexact"
1426 [19387] string: Integer overflow in memchr
1427 [19390] string: Integer overflow in strncat
1428 [19398] build: linknamespace tests fail with massively parallel build
1429 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1431 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1432 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1433 [19514] libc: [PATCH] Fix spelling errors in spelling
1434 "implement"/"implementation" in several places
1435 [19582] network: Deprecate RES_USE_INET6
1436 [19673] manual: clog10 docs appear to be erroneous
1437 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1438 causes a segmentation fault
1439 [19826] libc: invalid pointer returned from __tls_get_addr with static
1441 [20016] network: resolv: Remove hooks support from the API
1442 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1444 [20033] math: [x86_64] vectorized math function don't call the __finite
1446 [20116] nptl: use after free in pthread_create
1447 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1448 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1450 [20311] nptl: please install proc_service.h
1451 [20366] libc: Compilation errors in installed headers in strict-compliance
1453 [20370] malloc: malloc: Arena free list management is still racy
1454 (incorrect fix in bug 19243)
1455 [20386] libc: assert (X = 0) does not result in GCC warning
1456 [20432] malloc: malloc: Minimize interface required for interposition
1457 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1458 in _Unwind_Backtrace (CVE-2016-6323)
1459 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1460 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1462 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1463 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1464 [20477] network: resolv: incorrect double-checked locking related to
1466 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1468 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1470 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1471 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1472 cause transition penalty
1473 [20517] math: sparcv9 missing fdiml compat symbol
1474 [20524] manual: strverscmp is inconsistent
1475 [20525] libc: <sys/quota.h> should be based on kernel headers
1476 [20539] math: GCC 7 -static -lm fails to link at -O3
1477 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1479 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1481 [20591] network: Remove obsolete DNSSEC support
1482 [20592] network: DNS resource record type classification macros in
1483 <arpa/nameser.h> are incorrect
1484 [20593] network: Update DNS RR type definitions
1485 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1486 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1488 [20629] network: libresolv: Remove support for bitstring labels
1490 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1491 assertion: "maxidx >= 2"
1492 [20660] math: [arm] Use VSQRT
1493 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1494 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1495 [20689] libc: Test for FMA should also check for AVX.
1496 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1497 [20715] math: iszero macro breaks existing code
1498 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1499 [20728] libc: powerpc: Missing TOC stub in clone
1500 [20729] build: glibc-2.24 fails to build for i486 with -Os
1501 [20750] build: Build fails with default PIE enabled gcc-6.x
1502 [20768] math: [alpha] sqrt fegetenv namespace
1503 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1505 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1506 __FLT_EVAL_METHOD__ is 2
1507 [20790] network: rpcgen buffer overrun in get_prog_declaration
1508 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1509 [20829] libc: crypt snprintf namespace
1510 [20847] libc: tst-vfork3 failure
1511 [20855] math: Default bits/mathdef.h has inappropriate float_t
1512 [20859] math: [sh4] FP_ILOGB0 invalid
1513 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1514 U+20AC), not same as GBK
1515 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1516 [20916] math: pow handling of sNaN arguments
1517 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1518 [20919] math: Bad pow (qNaN, 0) result with -lieee
1519 [20940] math: hypot sNaN handling
1520 [20947] math: fmax, fmin sNaN handling
1521 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1522 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1524 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1525 [20973] nptl: robust mutexes: Lost wake-ups
1526 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1527 [20978] nis: strlen on null pointer in nss_nisplus
1528 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1530 [21014] string: i686 memchr overflows internal pointer check
1531 [21019] libc: [mips] n32 lseek incorrect on overflow
1532 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1533 [21026] libc: [MIPS] readahead syscall is broken on n64
1534 [21028] math: Fallback fesetexceptflag should always succeed
1535 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1536 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1538 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1539 [21061] librt: [microblaze] librt lost clock_* exports
1540 [21073] libc: tunables: insecure environment variables passed to
1541 subprocesses with AT_SECURE
1542 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1546 * The minimum Linux kernel version that this version of the GNU C Library
1547 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1548 kernel version 2.6.32 or later suffices (on architectures that already
1549 required kernel versions more recent than 3.2, those requirements remain
1550 unchanged). Linux 3.2 or later kernel headers are required on all
1553 * The pap_AN locale has been deleted. This has been deprecated for a long
1554 time. It has been replaced by pap_AW & pap_CW, both of which have long
1555 been included in previous releases.
1557 * The readdir_r and readdir64_r functions have been deprecated. It is
1558 recommended to use readdir and readdir64 instead.
1560 * The type “union wait” has been removed. It was deprecated in the early
1561 1990s and never part of POSIX. Application code should use the int type
1562 instead of “union wait”.
1564 * A new NSS action is added to facilitate large distributed system
1565 administration. The action, MERGE, allows remote user stores like LDAP
1566 to be merged into local user stores like /etc/groups in order to provide
1567 easy to use, updated, and managed sets of merged credentials. The new
1568 action can be used by configuring it in /etc/nsswitch.conf:
1569 group: files [SUCCESS=merge] nis
1570 Implemented by Stephen Gallagher (Red Hat).
1572 * The deprecated __malloc_initialize_hook variable has been removed from the
1575 * The long unused localedef --old-style option has been removed. It hasn't
1576 done anything in over 16 years. Scripts using this option can safely
1579 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1580 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1581 return the next representable value in the direction of positive infinity
1582 and the nextdown functions return the next representable value in the
1583 direction of negative infinity. These are currently enabled as GNU
1586 Security related changes:
1588 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1589 could result in a stack overflow when getnetbyname was called with an
1590 overly long name. (CVE-2016-3075)
1592 * Previously, getaddrinfo copied large amounts of address data to the stack,
1593 even after the fix for CVE-2013-4458 has been applied, potentially
1594 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1595 instead. Reported by Michael Petlan. (CVE-2016-3706)
1597 * The glob function suffered from a stack-based buffer overflow when it was
1598 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1599 Reported by Alexander Cherepanov. (CVE-2016-1234)
1601 * The Sun RPC UDP client could exhaust all available stack space when
1602 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1603 alloca plugin for GCC. (CVE-2016-4429)
1605 * The IPv6 name server management code in libresolv could result in a memory
1606 leak for each thread which is created, performs a failing naming lookup,
1607 and exits. Over time, this could result in a denial of service due to
1608 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1610 The following bugs are resolved with this release:
1612 [1170] localedata: ne_NP: update Nepali locale definition file
1613 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1615 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1616 [6796] math: fdim() does not set errno on overflow
1617 [10354] libc: posix_spawn should use vfork() in more cases than presently
1618 [11213] localedata: localedata: add copyright disclaimer to locale files
1619 [12143] localedata: chr_US: new Cherokee locale
1620 [12450] localedata: sgs_LT: new locale
1621 [12676] localedata: ln_CD: new locale
1622 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1624 [13304] math: fma, fmaf, fmal produce wrong results
1625 [14259] build: --localedir arg to configure is ignored
1626 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1627 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1628 [14934] localedata: es_CL: wrong first weekday chilean locale
1629 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1631 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1633 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1634 [15368] nptl: raise() is not async-signal-safe
1635 [15479] math: ceil, floor, round and trunc raise inexact exception
1636 [15578] localedata: kk_KZ: various updates
1637 [16003] localedata: pap_AN: punt old locale
1638 [16137] localedata: iw_IL: punt old locale
1639 [16190] localedata: eo: new esperanto locale
1640 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1641 [16742] malloc: race condition: pthread_atfork() called before first
1642 malloc() results in unexpected locking behaviour/deadlocks
1643 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1645 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1646 [17565] localedata: pt_PT: wrong (work-)week start
1647 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1648 [17950] build: Build fails with -msse
1649 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1650 [18433] libc: posix_spawn does not return correctly upon failure to
1652 [18453] localedata: charmaps/IBM875: incorrect codes
1653 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1655 [18896] localedata: he_IL: improvements for currency
1656 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1657 Romanian locale data
1658 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1660 [19056] libc: Deprecate readdir_r
1661 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1663 [19198] localedata: nl_NL: small improvements for Dutch locales
1664 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1665 nameservers (CVE-2016-5417)
1666 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1668 [19400] locale: Language missing in "iso-639.def", trivial fix in
1670 [19431] malloc: Deadlock between fflush, getdelim, and fork
1671 [19505] libc: Incorrect file descriptor validity checks in
1672 posix_spawn_file_actions_add{open,close,dup2}
1673 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1674 when using RTLD_NEXT
1675 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1676 `intl/{gettextP,loadinfo}.h'
1677 [19534] libc: execle, execlp may use malloc
1678 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1679 [19573] network: res_nclose and __res_maybe_init disagree about name
1680 server initialization, breaking Hesiod
1681 [19575] localedata: Status of GB18030 tables
1682 [19581] localedata: sr_* date_fmt string contains additional newline
1683 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1685 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1686 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1687 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1688 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1690 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1692 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1694 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1695 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1696 [19633] locale: strfmon_l applies global locale to number formatting
1697 [19642] network: Memory leak in getnameinfo
1698 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1699 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1701 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1702 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1704 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1705 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1707 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1709 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1711 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1712 pointers and lengths in error-case.
1713 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1714 report errors on UTF-16 surrogates values.
1715 [19755] nscd: nscd assertion failure in gc
1716 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1717 [19759] libc: mempcpy shouldn't be inlined
1718 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1719 [19765] libc: s390 needs an optimized mempcpy
1720 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1721 NAME_MAX limit assumption (CVE-2016-1234)
1722 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1723 [19787] network: Missing and incorrect truncation checks in getnameinfo
1724 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1726 [19791] network: Assertion failure in res_query.c with un-connectable name
1728 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1729 [19822] math: libm.so install clobbers old version
1730 [19825] network: resolv: send_vc can return uninitialized data in second
1731 response to getaddrinfo
1732 [19830] network: nss_dns: should check RDATA length against buffer length
1733 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1734 confronted with A/AAAA records of invalid size
1735 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1736 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1737 [19853] stdio: Printing IBM long double in decimal with high precision is
1739 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1740 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1741 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1743 [19865] network: Assertion failure or memory leak in
1744 _nss_dns_getcanonname_r
1745 [19868] network: nss_dns: netent code does not skip over non-PTR records
1746 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1748 [19881] string: Improve x86-64 memset
1749 [19907] string: Incorrect memcpy tests
1750 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1752 [19925] libc: termios.h XCASE namespace
1753 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1754 [19929] libc: limits.h NL_NMAX namespace
1755 [19931] stdio: Memory leak in vfprintf
1756 [19957] libc: clone(CLONE_VM) access invalid parent memory
1757 [19963] localedata: en_IL: New locale
1758 [19989] stdio: stdio.h cuserid namespace
1759 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1761 [19996] locale: langinfo.h nl_langinfo_l namespace
1762 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1763 [20010] network: getaddrinfo: Stack overflow in hostent translation
1765 [20012] stdio: libio: fmemopen append mode failure
1766 [20014] stdio: stdio.h namespace for pre-threads POSIX
1767 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1768 [20023] libc: fcntl.h timespec namespace
1769 [20024] math: [x86_64] vectorized sincos trashes the stack
1770 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1771 [20041] time: sys/time.h timespec namespace
1772 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1773 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1774 [20051] libc: ttyslot in wrong header under wrong conditions
1775 [20054] libc: gethostname not declared for XPG4
1776 [20055] libc: termios.h missing tcgetsid for XPG4
1777 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1779 [20073] libc: sys/stat.h fchmod namespace
1780 [20074] libc: stdlib.h rand_r namespace
1781 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1782 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1784 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1785 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1787 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1788 [20119] libc: Wrong mask for processors level type from CPUID
1789 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1790 AS not supporting AVX512
1791 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1793 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1794 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1796 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1797 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1798 [20178] libc: posix_spawn{p} should not call exit
1799 [20191] stdio: libio: vtables hardening
1800 [20195] string: FMA4 detection requires CPUID execution with register
1802 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1803 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1805 [20212] math: acos (sNaN) returns sNaN
1806 [20213] math: asin (sNaN) returns sNaN
1807 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1808 [20218] math: [i386] asinhl (sNaN) returns sNaN
1809 [20219] math: [i386] atanhl (sNaN) returns sNaN
1810 [20222] stdio: fopencookie: Mangle function pointers
1811 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1812 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1813 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1815 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1816 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1817 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1818 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1819 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1820 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1821 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1822 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1823 [20240] math: modf (sNaN) returns sNaN
1824 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1825 [20250] math: frexp (sNaN) returns sNaN
1826 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1827 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1829 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1830 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1831 subscript is below array bounds [-Werror=array-bounds]
1832 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1833 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1835 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1836 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1838 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1839 "invalid" exceptions
1840 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1841 [20316] localedata: id_ID: Februari instead of Pebruari
1842 [20327] string: POWER8 strcasecmp returns incorrect result
1843 [20347] math: Failure: Test: j0_downward (0xap+0)
1844 [20348] libc: FAIL: misc/tst-preadvwritev64
1845 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1846 [20350] libc: There is no test for p{read,write}64
1847 [20357] math: Incorrect cos result for 1.5174239687223976
1848 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1852 * Unicode 8.0.0 Support: Character encoding, character type info, and
1853 transliteration tables are all updated to Unicode 8.0.0, using new
1854 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1855 These updates cause user visible changes, such as the fixes for bugs
1856 89, 16061, and 18568.
1858 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1859 kernel-internal CPU set size. This means that requests that change the
1860 CPU affinity which failed before (for example, an all-ones CPU mask) will
1861 now succeed. Applications that need to determine the effective CPU
1862 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1863 setting it because the kernel can adjust it (and the previous size check
1864 would not detect this in the majority of cases).
1866 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1867 the following new symbols are used: fts64_children, fts64_close,
1868 fts64_open, fts64_read and fts64_set.
1870 * getaddrinfo now detects certain invalid responses on an internal netlink
1871 socket. If such responses are received, an affected process will
1872 terminate with an error message of "Unexpected error <number> on netlink
1873 descriptor <number>" or "Unexpected netlink response of size <number> on
1874 descriptor <number>". The most likely cause for these errors is a
1875 multi-threaded application which erroneously closes and reuses the netlink
1876 file descriptor while it is used by getaddrinfo.
1878 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1879 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1880 unnecessary serialization of memory allocation requests across threads.
1881 The defect is now corrected. Users should see a substantial increase in
1882 the concurent throughput of allocation requests for applications which
1883 trigger this bug. Affected applications typically create and
1884 destroy threads frequently. (Bug 19048 was reported and analyzed by
1887 * There is now a --disable-timezone-tools configure option for disabling the
1888 building and installing of the timezone related utilities (zic, zdump, and
1889 tzselect). This is useful for people who build the timezone data and code
1890 independent of the GNU C Library.
1892 * The obsolete header <regexp.h> has been removed. Programs that require
1893 this header must be updated to use <regex.h> instead.
1895 * The obsolete functions bdflush, create_module, get_kernel_syms,
1896 query_module and uselib are no longer available to newly linked binaries;
1897 the header <sys/kdaemon.h> has been removed. These functions and header
1898 were specific to systems using the Linux kernel and could not usefully be
1899 used with the GNU C Library on systems with version 2.6 or later of the
1902 * Optimized string, wcsmbs and memory functions for IBM z13.
1903 Implemented by Stefan Liebler.
1905 * Newly linked programs that define a variable called signgam will no longer
1906 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1907 require signgam to be set by those functions must ensure that they use the
1908 variable provided by the GNU C Library and declared in <math.h>, without
1909 defining their own copy.
1911 * The minimum GCC version that can be used to build this version of the GNU
1912 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1913 still be used to compile programs using the GNU C Library.
1915 Security related changes:
1917 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1918 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1920 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1921 the pointer guard feature. It is always enabled. Previously,
1922 LD_POINTER_GUARD could be used to disable security hardening in binaries
1923 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1926 * An integer overflow in hcreate and hcreate_r could lead to an
1927 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1929 * The catopen function no longer has unbounded stack usage. Reported by
1930 Max. (CVE-2015-8779)
1932 * The nan, nanf and nanl functions no longer have unbounded stack usage
1933 depending on the length of the string passed as an argument to the
1934 functions. Reported by Joseph Myers. (CVE-2014-9761)
1936 * A stack-based buffer overflow was found in libresolv when invoked from
1937 libnss_dns, allowing specially crafted DNS responses to seize control
1938 of execution flow in the DNS client. The buffer overflow occurs in
1939 the functions send_dg (send datagram) and send_vc (send TCP) for the
1940 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1941 family. The use of AF_UNSPEC triggers the low-level resolver code to
1942 send out two parallel queries for A and AAAA. A mismanagement of the
1943 buffers used for those queries could result in the response of a query
1944 writing beyond the alloca allocated buffer created by
1945 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1946 the overflow. Thanks to the Google Security Team and Red Hat for
1947 reporting the security impact of this issue, and Robert Holiday of
1948 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1950 The following bugs are resolved with this release:
1952 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1953 [887] math: Math library function "logb" and "nextafter" inconsistent
1954 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1955 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1956 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1957 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1959 [4404] localedata: German translation of "Alarm clock" is misleading
1960 [6799] math: nextafter() and nexttoward() doen't set errno on
1961 overflow/underflow errors
1962 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1964 [10432] nis: _nss_nis_setnetgrent assertion failure
1965 [11460] libc: fts has no LFS support
1966 [12926] network: getaddrinfo()/make_request() may spin forever
1967 [13065] nptl: Race condition in pthread barriers
1968 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1969 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1971 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1972 [14912] libc: Rename non-installed bits/*.h headers
1973 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1974 [15367] math: Let gcc use __builtin_isinf
1975 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1976 [15421] math: lgamma wrongly sets signgam for ISO C
1977 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1978 FE_INVALID with argument out of range
1979 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1981 [15786] dynamic-link: ifunc resolver functions can smash function
1983 [15918] math: Unnecessary check for equality in hypotf()
1984 [16061] localedata: Review / update transliteration data
1985 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1987 [16141] time: strptime %z offset restriction
1988 [16171] math: drem should be alias of remainder
1989 [16296] math: fegetround is pure?
1990 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1991 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1993 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1994 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1996 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1997 "Exception "Inexact" set" on ppc32
1998 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1999 [16517] math: Missing underflow exception from tanf/tan/tanl
2000 [16519] math: Missing underflow exception from sinhf
2001 [16520] math: Missing underflow exception from tanhf
2002 [16521] math: Missing underflow exception from exp2
2003 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2005 [16734] stdio: fopen calls mmap to allocate its buffer
2006 [16961] math: nan function incorrect handling of bad sequences
2007 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2008 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2009 [16985] locale: localedef: confusing error message when opening output
2011 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2012 [17197] locale: Redundant shift character in iconv conversion output at
2014 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2015 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2016 [17250] dynamic-link: static linking breaks nss loading
2017 (getaddrinfo/getpwnam/etc...)
2018 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2020 [17441] math: isnan() should use __builtin_isnan() in GCC
2021 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2022 (related to lock elision)
2023 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2024 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2025 [17887] time: strptime should be able to parse "+01:00" style timezones
2026 [17905] libc: catopen() Multiple unbounded stack allocations
2028 [18084] libc: backtrace (..., 0) dumps core on x86
2029 [18086] libc: nice() sets errno to 0 on success
2030 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2031 is too large (CVE-2015-8778)
2032 [18251] dynamic-link: SONAME missing when audit modules provides path
2033 [18265] libc: add attributes for wchar string and memory functions
2034 [18370] math: csqrt missing underflows
2035 [18421] libc: [hppa] read-only segment has dynamic relocations
2036 [18472] libc: Obsolete syscall wrappers should be compat symbols
2037 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2038 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2039 [18525] localedata: Remove locale timezone information
2040 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2041 [18568] localedata: Update locale data to Unicode 8.0
2042 [18589] locale: sort-test.sh fails at random
2043 [18595] math: ctan, ctanh missing underflows
2044 [18604] libc: assert macro-expands its argument
2045 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2046 contains a vector instruction exception.
2047 [18611] math: j1, jn missing errno setting on underflow
2048 [18618] localedata: sync Chechen locale definitions with other *_RU
2050 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2051 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2052 [18665] network: In send_dg, the recvfrom function is NOT always using the
2053 buffer size of a newly created buffer (CVE-2015-7547)
2054 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2056 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2058 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2059 [18699] math: tilegx cproj() for various complex infinities does not yield
2061 [18724] libc: Harden put*ent functions against data injection
2062 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2063 [18755] build: build errors with -DNDEBUG
2064 [18757] stdio: fmemopen fails to set errno on failure
2065 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2067 [18781] libc: openat64 lacks O_LARGEFILE
2068 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2069 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2070 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2071 [18790] math: [ldbl-128ibm] tanhl inaccurate
2072 [18795] libc: stpncpy fortification misses buffer lengths that are
2073 statically too large
2074 [18796] build: build fails for --disable-mathvec
2075 [18803] math: hypot missing underflows
2076 [18820] stdio: fmemopen may leak memory on failure
2077 [18823] math: csqrt spurious underflows
2078 [18824] math: fma spurious underflows
2079 [18825] math: pow missing underflows
2080 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2081 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2082 [18870] build: sem_open.c fails to compile with missing symbol
2084 [18872] stdio: Fix memory leak in printf_positional
2085 [18873] libc: posix_fallocate overflow check ineffective
2086 [18875] math: Excess precision leads incorrect libm
2087 [18877] libc: arm: mmap offset regression
2088 [18887] libc: memory corruption when using getmntent on blank lines
2089 [18918] localedata: hu_HU: change time to HH:MM:SS format
2090 [18921] libc: Regression: extraneous stat() and fstat() performed by
2092 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2093 binaries (CVE-2015-8777)
2094 [18951] math: tgamma missing underflows
2095 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2097 [18953] localedata: lt_LT: change currency symbol to the euro
2098 [18956] math: powf inaccuracy
2099 [18961] math: [i386] exp missing underflows
2100 [18966] math: [i386] exp10 missing underflows
2101 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2102 [18969] build: multiple string test failures due to missing locale
2104 [18970] libc: Reference of pthread_setcancelstate in libc.a
2105 [18977] math: float / long double Bessel functions not in XSI POSIX
2106 [18980] math: i386 libm functions return with excess range and precision
2107 [18981] math: i386 scalb*, ldexp return with excess range and precision
2108 [18982] stdio: va_list and vprintf
2109 [18985] time: Passing out of range data to strftime() causes a segfault
2111 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2112 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2113 [19012] locale: iconv_open leaks memory on error path
2114 [19016] math: clog, clog10 inaccuracy
2115 [19018] nptl: Mangle function pointers in tls_dtor_list
2116 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2117 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2118 [19048] malloc: malloc: arena free list can become cyclic, increasing
2120 [19049] math: [powerpc] erfc incorrect zero sign
2121 [19050] math: [powerpc] log* incorrect zero sign
2122 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2123 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2124 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2125 [19074] network: Data race in _res_hconf_reorder_addrs
2126 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2127 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2128 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2130 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2132 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2133 [19086] manual: posix_fallocate64 documented argument order is wrong.
2134 [19088] math: lround, llround missing exceptions close to overflow
2136 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2137 [19095] math: dbl-64 lrint incorrect for 64-bit long
2138 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2139 [19124] dynamic-link: ld.so failed to build with older assmebler
2140 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2141 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2142 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2143 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2144 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2145 pthread_setaffinity_np
2146 [19156] math: [ldbl-128] j0l spurious underflows
2147 [19164] nptl: tst-getcpu fails with many possible CPUs
2148 [19168] math: math/test-ildoubl and math/test-ldouble failure
2149 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2150 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2152 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2153 (FE_NOMASK_ENV) do not clear SSE exceptions
2154 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2155 _int_new_arena/reused_arena
2156 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2157 [19201] math: dbl-64 remainder incorrect sign of zero result
2158 [19205] math: bits/math-finite.h conditions do not match math.h and
2160 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2161 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2162 for C99-based standards
2163 [19212] libc: features.h not -Wundef clean
2164 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2166 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2167 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2168 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2170 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2171 "inexact" exceptions
2172 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2174 [19242] libc: strtol incorrect in Turkish locales
2175 [19243] malloc: reused_arena can pick an arena on the free list, leading
2176 to an assertion failure and reference count corruption
2177 [19253] time: tzset() ineffective when temporary TZ did not include DST
2179 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2180 [19270] math: [hppa] Shared libm missing __isnanl
2181 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2182 MAP_HUGETLB and MAP_STACK defines
2183 [19313] nptl: Wrong __cpu_mask for x32
2184 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2185 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2186 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2187 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2188 [19363] time: x32: times() return value wrongly truncates/sign extends
2190 [19367] dynamic-link: Improve branch prediction on Silvermont
2191 [19369] network: Default domain name not reset by res_ninit when "search"
2192 / "domain" entry is removed from resolv.conf
2193 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2195 [19385] localedata: bg_BG: time separator should be colon, not comma
2196 [19408] libc: linux personality syscall wrapper may erroneously return an
2197 error on 32-bit architectures
2198 [19415] libc: dladdr returns wrong names on hppa
2199 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2200 IBM905, IBM907, and IBM909
2201 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2202 [19443] build: build failures with -DDEBUG
2203 [19451] build: Make check fails on test-double-vlen2
2204 [19462] libc: Glibc failed to build with -Os
2205 [19465] math: Wrong code with -Os
2206 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2208 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2210 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2211 [19486] math: S390: Math tests fail with "Exception Inexact set".
2212 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2213 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2215 [19590] math: Fail to build shared objects that use libmvec.so functions.
2219 * The following bugs are resolved with this release:
2221 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2222 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2223 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2224 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2225 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2226 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2227 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2228 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2229 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2230 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2231 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2232 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2233 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2234 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2235 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2236 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2237 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2238 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2239 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2240 18657, 18676, 18694, 18696, 18887.
2242 * Cache information can be queried via sysconf() function on s390 e.g. with
2243 _SC_LEVEL1_ICACHE_SIZE as argument.
2245 * A buffer overflow in gethostbyname_r and related functions performing DNS
2246 requests has been fixed. If the NSS functions were called with a
2247 misaligned buffer, the buffer length change due to pointer alignment was
2248 not taken into account. This could result in application crashes or,
2249 potentially arbitrary code execution, using crafted, but syntactically
2250 valid DNS responses. (CVE-2015-1781)
2252 * The time zone file parser has been made more robust against crafted time
2253 zone files, avoiding heap buffer overflows related to the processing of
2254 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2255 large time zone data files. Overly long time zone specifiers in the TZ
2256 variable no longer result in stack overflows and crashes.
2258 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2259 for LD and GD on x86 and x86-64, has been implemented. You will need
2260 binutils-2.24 or later to enable this optimization.
2262 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2263 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2264 Hat). These updates cause user visible changes, such as the fix for bug
2267 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2268 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2269 condition in some applications.
2271 * Added vector math library named libmvec with the following vectorized x86_64
2272 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2274 The library can be disabled with --disable-mathvec. Use of the functions is
2275 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2276 Shared library libmvec.so is linked in as needed when using -lm (no need to
2277 specify -lmvec explicitly for not static builds).
2278 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2280 * A new fmemopen implementation has been added with the goal of POSIX
2281 compliance. The new implementation fixes the following long-standing
2282 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2283 old implementation is still present for use be by existing binaries.
2285 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2286 releases. It has been fixed to match 2.19 and older, but binaries built
2287 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2289 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2290 Contributed by Roland McGrath (Google).
2292 * The header <regexp.h> is deprecated, and will be removed in a future
2293 release. Use of this header will trigger a deprecation warning.
2294 Application developers should update their code to use <regex.h> instead.
2296 This header was formerly part of SUS, but was deprecated in 1994 and
2297 removed from the standard in 2001. Also, the glibc implementation
2298 leaks memory. See BZ#18681 for more details.
2302 * The following bugs are resolved with this release:
2304 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2305 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2306 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2307 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2308 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2309 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2310 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2311 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2312 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2313 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2316 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2317 too little memory for the to-be-scanned arguments and overflow the
2318 allocated buffer. The implementation now correctly computes the required
2319 buffer size when using malloc, and switches to malloc from alloca as
2322 * A new semaphore algorithm has been implemented in generic C code for all
2323 machines. Previous custom assembly implementations of semaphore were
2324 difficult to reason about or ensure that they were safe. The new version
2325 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2326 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2327 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2329 * Port to Altera Nios II has been contributed by Mentor Graphics.
2331 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2332 implementations for powerpc64/powerpc64le.
2333 Implemented by Adhemerval Zanella (IBM).
2335 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2336 and powerpc64le. This may improve lock scaling of existing programs on
2337 HTM capable systems. The lock elision code is only enabled with
2338 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2339 powerpc will issue a transaction abort on every syscall to avoid side
2340 effects being visible outside transactions.
2342 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2343 AArch64. Contributed by ARM Ltd.
2345 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2347 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2348 under certain input conditions resulting in the execution of a shell for
2349 command substitution when the application did not request it. The
2350 implementation now checks WRDE_NOCMD immediately before executing the
2351 shell and returns the error WRDE_CMDSUB as expected.
2353 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2354 processing format strings with a large number of format specifiers.
2356 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2357 infinite loop if the DNS response contained a PTR record of an unexpected
2360 * The minimum GCC version that can be used to build this version of the GNU
2361 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2362 still be used to compile programs using the GNU C Library.
2364 * The GNU C Library is now built with -Werror by default. This can be
2365 disabled by configuring with --disable-werror.
2367 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2369 * The obsolete sigvec function has been removed. This was the original
2370 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2371 programs have been using instead for about 25 years. Of course, ABI
2372 compatibility for old binaries using sigvec remains intact.
2374 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2375 with newer versions of bison.
2377 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2378 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2379 registers overlay two consecutive single-precision registers. MIPS32R2
2380 introduced a new FPU mode (FR=1) where double-precision registers extend the
2381 corresponding single-precision registers which is incompatible with the
2382 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2383 require the use of FR=1 making a transition necessary. New o32 ABI
2384 extensions enable users to migrate over time from the original o32 ABI
2385 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2386 tracks the ABI of any loaded object and verifies that new objects are
2387 compatible. Mode transitions will also be requested as required and
2388 unsupportable objects will be rejected. The ABI checks include both soft and
2389 hard float ABIs for o32, n32 and n64.
2391 GCC 5 with GNU binutils 2.25 onwards:
2392 It is strongly recommended that all o32 system libraries are built using the
2393 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2394 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2395 compiler using --with-fp-32=xx to set this by default.
2399 * The following bugs are resolved with this release:
2401 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2402 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2403 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2404 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2405 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2406 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2407 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2408 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2409 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2410 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2411 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2412 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2413 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2414 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2415 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2417 * Reverted change of ABI data structures for s390 and s390x:
2418 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2419 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2420 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2421 are preserved pointing straight to the same implementation as the old ones.
2422 Given that, new callers will simply provide a too-big buffer to these
2423 functions. Any applications/libraries out there that embed jmp_buf or
2424 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2425 against 2.19 headers will have to rebuilt again. This is necessary in any
2426 case to revert the breakage in their ABI caused by the glibc change.
2428 * Support for file description locks is added to systems running the
2429 Linux kernel. The standard file locking interfaces are extended to
2430 operate on file descriptions, not file descriptors, via the use of
2431 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2432 are associated with an open file instead of a process.
2434 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2436 * The minimum Linux kernel version that this version of the GNU C Library
2437 can be used with is 2.6.32.
2439 * Running the testsuite no longer terminates as soon as a test fails.
2440 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2441 with PASS or FAIL lines for individual tests. A summary of the results is
2442 printed, including a list of failing lists, and "make check" exits with
2443 error status if there were any unexpected failures. "make check
2444 stop-on-test-failure=y" may be used to keep the old behavior.
2446 * The am33 port, which had not worked for several years, has been removed
2449 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2450 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2451 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2452 conflicted with POSIX (support for which was removed in 2.19), the
2453 interfaces those macros enabled remain available when compiling with
2454 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2455 test macros defined.
2457 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2459 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2460 This may improve lock scaling of existing programs on TX capable systems.
2461 The lock elision code is only built with --enable-lock-elision=yes and
2462 then requires a GCC version supporting the TX builtins. With lock elision
2463 default mutexes are elided via __builtin_tbegin, if the cpu supports
2464 transactions. By default lock elision is not enabled and the elision code
2467 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2468 copy the path argument. This allowed programs to cause posix_spawn to
2469 deference a dangling pointer, or use an unexpected pathname argument if
2470 the string was modified after the posix_spawn_file_actions_addopen
2473 * All supported architectures now use the main glibc sysdeps directory
2474 instead of some being in a separate "ports" directory (which was
2475 distributed separately before glibc 2.17).
2477 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2478 On configurations that support it (all Linux configurations), it's now
2479 used regardless of the --enable-add-ons switch to configure. It is no
2480 longer possible to build such configurations without pthreads support.
2482 * Locale names, including those obtained from environment variables (LANG
2483 and the LC_* variables), are more tightly checked for proper syntax.
2484 setlocale will now fail (with EINVAL) for locale names that are overly
2485 long, contain slashes without starting with a slash, or contain ".." path
2486 components. (CVE-2014-0475) Previously, some valid locale names were
2487 silently replaced with the "C" locale when running in AT_SECURE mode
2488 (e.g., in a SUID program). This is no longer necessary because of the
2491 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2492 with application code using Intel MPX instructions. (With all previous
2493 versions, the MPX register state could be clobbered when making calls
2494 into or out of a shared library.) Note that while the new dynamic
2495 linker is compatible with all known x86 hardware whether or not it
2496 supports Intel MPX, some x86 instruction-set emulators might fail to
2497 handle the new instruction encodings. This is known to affect Valgrind
2498 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2499 release), and might affect other tools that do instruction emulation.
2501 * Support for loadable gconv transliteration modules has been removed.
2502 The support for transliteration modules has been non-functional for
2503 over a decade, and the removal is prompted by security defects. The
2504 normal gconv conversion modules are still supported. Transliteration
2505 with //TRANSLIT is still possible, and the //IGNORE specifier
2506 continues to be supported. (CVE-2014-5119)
2508 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2509 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2510 resulting a denial-of-service security vulnerability in applications which
2511 use functions related to iconv. (CVE-2014-6040)
2515 * The following bugs are resolved with this release:
2517 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2518 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2519 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2520 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2521 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2522 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2523 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2524 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2525 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2526 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2527 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2528 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2529 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2530 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2531 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2532 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2533 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2534 16431, 16453, 16474, 16506, 16510, 16529
2536 * Slovenian translations for glibc messages have been contributed by the
2537 Translation Project's Slovenian team of translators.
2539 * The public headers no longer use __unused nor __block. This change is to
2540 support compiling programs that are derived from BSD sources and use
2541 __unused internally, and to support compiling with Clang's -fblock
2542 extension which uses __block.
2544 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2545 large collation sequences to optimize multiple passes. This cache
2546 computation may overflow for large collation sequences and may cause a
2547 stack or buffer overflow. This is now fixed to use a slower algorithm
2548 which does not use a cache if there is an integer overflow.
2550 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2551 rules for large collation sequences to optimize multiple passes and falls
2552 back to alloca if malloc fails, resulting in a possible stack overflow.
2553 The implementation now falls back to an uncached collation sequence lookup
2556 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2557 initialized for static applications resulting in the security feature
2558 being disabled. The pointer guard is now correctly initialized to a
2559 random value for static applications. Existing static applications need
2560 to be recompiled to take advantage of the fix (bug 15754).
2562 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2563 to the d_name member of struct dirent, or omit the terminating NUL
2564 character. (Bugzilla #14699).
2566 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2567 aligned_alloc functions could allocate too few bytes or corrupt the
2568 heap when passed very large allocation size values (Bugzilla #15855,
2571 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2572 for AF_INET6 has been fixed (Bugzilla #16072).
2574 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2575 pap_CW, quz_PE, the_NP.
2577 * Substantially revised locales: gd_GB, ht_HT
2579 * The LC_ADDRESS field was updated to support country_car for almost all
2582 * ISO 1427 definitions were updated.
2584 * ISO 3166 definitions were updated.
2586 * The localedef utility now supports --big-endian and --little-endian
2587 command-line options to generate locales for a different system from that
2588 for which the C library was built.
2590 * Binary locale files now only depend on the endianness of the system for
2591 which they are generated and not on other properties of that system. As a
2592 consequence, binary files generated with new localedef may be incompatible
2593 with old versions of the GNU C Library, and binary files generated with
2594 old localedef may be incompatible with this version of the GNU C Library,
2595 in the following circumstances:
2597 + Locale files may be incompatible on m68k systems.
2599 + Locale archive files (but not separate files for individual locales) may
2600 be incompatible on systems where plain "char" is signed.
2602 * The configure option --disable-versioning has been removed. Builds with
2603 --disable-versioning had not worked for several years.
2605 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2606 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2608 * SystemTap probes for malloc have been introduced.
2610 * SystemTap probes for slow multiple precision fallback paths of
2611 transcendental functions have been introduced.
2613 * Support for powerpc64le has been added.
2615 * The soft-float powerpc port now supports e500 processors.
2617 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2619 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2620 set of header declarations that are enabled by default, even when other
2621 feature test macros or compiler options such as -std=c99 would otherwise
2622 disable some of those declarations.
2624 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2625 conflict with POSIX. The libbsd-compat library (which was a dummy library
2626 that did nothing) has also been removed.
2628 * Preliminary documentation about Multi-Thread, Async-Signal and
2629 Async-Cancel Safety has been added.
2631 * Change of ABI data structures for s390 and s390x:
2632 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2633 allow for future hardware extensions. All library functions that accept or
2634 return these structures were versioned in order to provide backwards
2635 compatibility with existing code. However, not all cases can be handled
2636 with interface versioning. If an application embeds either structure into
2637 another structure and that structure is passed to another compilation unit
2638 which expects the newer larger-sized object then any access to the new
2639 fields will result in undefined behaviour. Similarly any access to fields
2640 that were moved by the enlarging of the structures will cause undefined
2641 behaviour. To fix the undefined behaviour all objects that are part of the
2642 final in-memory image must agree on the size of structures, and this may
2643 require recompilation.
2647 * The following bugs are resolved with this release:
2649 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2650 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2651 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2652 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2653 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2654 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2655 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2656 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2657 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2658 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2659 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2660 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2661 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2664 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2665 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2666 Distributions can re-enable building and using pt_chown via the new configure
2667 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2668 considerable security risks and should only be used if the distribution
2669 understands and accepts the risks.
2671 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2674 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2675 fixed (Bugzilla #15330).
2677 * Add support for calling C++11 thread_local object destructors on thread
2678 and program exit. This needs compiler support for offloading C++11
2679 destructor calls to glibc.
2681 * Improved worst case performance of libm functions with double inputs and
2684 * Support for priority inherited mutexes in pthread condition variables on
2685 non-x86 architectures.
2687 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2689 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2691 * Optimized string functions for ARM. Implemented by Will Newton and
2694 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2696 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2699 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2700 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2702 * Added a benchmark framework to track performance of functions in glibc.
2704 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2705 It is based on draft TS 18661 and currently enabled as a GNU extension.
2707 * On Linux, the clock function now uses the clock_gettime system call
2708 for improved precision, rather than old times system call.
2710 * Added support for version-3 tz format files. This is needed when using
2711 the tz database release 2013e or later, and affects a few unusual cases --
2712 currently only TZ='America/Godthab' for time stamps after 2037.
2714 * Added new API functions pthread_getattr_default_np and
2715 pthread_setattr_default_np to get and set the default pthread
2716 attributes of a process.
2718 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2719 This may improve lock scaling of existing programs on TSX capable systems.
2720 When the --enable-lock-elision=yes parameter is specified at configure
2721 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2724 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2725 availability). Implemented by Ryan S. Arnold.
2727 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2731 * The following bugs are resolved with this release:
2733 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2734 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2735 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2736 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2737 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2738 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2739 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2740 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2741 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2742 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2743 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2744 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2745 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2746 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2747 14879, 14889, 14893, 14898, 14914.
2749 * Optimization of memcpy for MIPS.
2751 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2752 EMFILE has been fixed (Bugzilla #14889).
2754 * The add-on ports collection is now distributed in the "ports" subdirectory
2755 of the main GNU C Library distribution, rather than separately.
2757 * Port to ARM AArch64 contributed by Linaro.
2759 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2760 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2762 Implemented by Andreas Krebbel.
2764 * The new function secure_getenv allows secure access to the environment,
2765 returning NULL if running in a SUID/SGID process. This function replaces
2766 the internal function __secure_getenv.
2768 * SystemTap static probes have been added into the dynamic linker.
2769 Implemented by Gary Benson.
2771 * Optimizations of string functions strstr, strcasestr and memmem.
2772 Implemented by Maxim Kuvyrkov.
2774 * The minimum Linux kernel version that this version of the GNU C Library
2775 can be used with is 2.6.16.
2777 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2778 powerpc POWER7. Implemented by Will Schmidt.
2780 * New configure option --disable-nscd builds the C library such that it
2781 never attempts to contact the Name Service Caching Daemon (nscd).
2782 New configure option --disable-build-nscd avoids building nscd itself;
2783 this is the default if --disable-nscd is used.
2785 * Improved support for cross-compilation, including cross-testing and
2786 bootstrap builds without a previously built glibc.
2788 * Several testsuite tests are now able to test multiple IFUNC variants of an
2789 interface, rather than just testing the one that would be chooen by
2792 * New configure options --with-bugurl and --with-pkgversion, for
2793 distributors to use to embed their bug-reporting and package version
2794 information in --help and --version output.
2796 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2797 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2798 allows creation of chroots without the procfs mounted on /proc.
2800 * The `crypt' function now fails if passed salt bytes that violate the
2801 specification for those values. On Linux, the `crypt' function will
2802 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2803 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2804 when the mode is enabled.
2806 * The `clock_*' suite of functions (declared in <time.h>) is now available
2807 directly in the main C library. Previously it was necessary to link with
2808 -lrt to use these functions. This change has the effect that a
2809 single-threaded program that uses a function such as `clock_gettime' (and
2810 is not linked with -lrt) will no longer implicitly load the pthreads
2811 library at runtime and so will not suffer the overheads associated with
2812 multi-thread support in other code such as the C++ runtime library.
2814 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2819 * The following bugs are resolved with this release:
2821 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2822 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2823 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2824 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2825 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2826 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2827 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2828 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2829 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2830 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2831 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2832 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2833 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2834 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2835 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2836 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2837 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2838 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2839 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2840 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2841 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2842 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2845 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2846 configuring glibc with:
2847 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2848 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2849 Implemented by H.J. Lu.
2853 + define static_assert
2855 + do not declare gets
2857 + declare at_quick_exit and quick_exit also for ISO C11
2859 + aligned_alloc. NB: The code is deliberately allows the size parameter
2860 to not be a multiple of the alignment. This is a moronic requirement
2861 in the standard but it is only a requirement on the caller, not the
2864 + timespec_get added
2866 + uchar.h support added
2868 + CMPLX, CMPLXF, CMPLXL added
2870 Implemented by Ulrich Drepper.
2872 * Support for the IA-64 has been moved to ports.
2874 * Remove support for anything but ELF binary format
2876 * Checking versions of poll, ppoll added.
2877 Implemented by Ulrich Drepper.
2879 * More generic and 64-bit performance optimizations to math functions.
2880 Implemented by Ulrich Drepper.
2882 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2883 headers and functions available at compile time as they were before
2884 version 2.14. This option will be removed at some time in the future
2885 after the TI-RPC library becomes fully sufficient for the needs of
2886 existing applications.
2888 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2889 Note that glibc is not expected to work with any Linux kernel version
2892 * New header <sys/auxv.h> and function getauxval allowing easy access to
2893 the AT_* key-value pairs passed from the Linux kernel. The header also
2894 defines the HWCAP_* bits associated with the AT_HWCAP key.
2896 * New locales: mag_IN
2898 * New configure option --enable-systemtap builds SystemTap static probes
2899 into libc for setjmp and longjmp and into libpthread for various operations.
2900 So far the setjmp/longjmp probes and some of the libpthread probes are
2901 provided only for i*86 and x86_64.
2902 Implemented by Roland McGrath and Rayson Ho.
2904 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2906 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2909 * More optimized functions for SPARC. Implemented by David S. Miller.
2911 * Improved support for cross-compilation, especially bootstrap builds
2912 without a previously built glibc.
2914 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2915 Chris Metcalf from Tilera.
2917 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2918 now supported for ARM processors.
2920 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2921 as the name of the dynamic linker, to distinguish it from the
2922 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2924 * Support for CRIS has been removed from ports.
2926 * A new class of installed header has been documented for low-level
2927 platform-specific functionality. PowerPC added the first instance with a
2928 function to provide time base register access. Contributed by Tulio
2929 Magno Quites Machado Filho.
2931 * ix86 configurations now install header files that are consistent with
2932 what x86-64 configurations install. These same header files can be used
2933 for -m32, -m64, or -mx32 builds.
2934 Contributed by H.J. Lu.
2936 * Math library bug fixes. A thorough audit of all open math library bugs was
2937 conducted by Joseph Myers. Significant progress was made on many math
2938 library bugs resulting in more accurate exceptions and function results.
2939 Many thanks to all those that contributed including Andreas Jaeger for his
2940 patch review and work on the x87 trigonometric instruction issues.
2942 * Timezone data is no longer installed. Timezone-related binaries and scripts
2943 will continue to be installed. Users should obtain their timezone data from
2944 their distribution provider or from the tzdata package at
2945 <ftp://munnari.oz.au/pub/>.
2949 * The following bugs are resolved with this release:
2951 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2952 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2953 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2954 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2955 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2956 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2957 13484, 13506, 13515, 13523, 13524, 13538, 13540
2959 * New program pldd to list loaded object of a process
2960 Implemented by Ulrich Drepper.
2962 * Add nss_db support back to glibc. No more dependency on Berkeley db
2963 and support for initgroups lookups.
2964 Implemented by Ulrich Drepper.
2966 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2967 Contributed by HJ Lu.
2969 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2970 Contributed by HJ Lu.
2972 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2973 on x86-32 and x86-64.
2974 Contributed by Liubov Dmitrieva.
2976 * Optimized strchr and strrchr for SSE on x86-32.
2977 Contributed by Liubov Dmitrieva.
2979 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2980 for x86-64 and x86-32.
2981 Contributed by Liubov Dmitrieva.
2983 * New interfaces: scandirat, scandirat64
2984 Implemented by Ulrich Drepper.
2986 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2987 Implemented by Ulrich Drepper.
2989 * nscd now also caches the netgroup database.
2990 Implemented by Ulrich Drepper.
2992 * Integrate libm with gcc's -ffinite-math-only option.
2993 Implemented by Ulrich Drepper.
2995 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2996 to math functions. Implemented by Ulrich Drepper.
2998 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2999 Implemented by Ulrich Drepper.
3001 * New Linux interfaces: process_vm_readv, process_vm_writev
3003 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3004 Implemented by Ulrich Drepper.
3006 * Optimized nearbyint and strcasecmp for PPC.
3007 Implemented by Adhemerval Zanella.
3009 * New locales: bho_IN, unm_US, es_CU, ta_LK
3013 * The following bugs are resolved with this release:
3015 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3016 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3017 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3018 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3019 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3020 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3021 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3022 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3023 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3024 12795, 12811, 12813, 12814, 12841
3026 * The RPC implementation in libc is obsoleted. Old programs keep working
3027 but new programs cannot be linked with the routines in libc anymore.
3028 Programs in need of RPC functionality must be linked against TI-RPC.
3029 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3031 Visible changes of this change include (obviously) the inability to link
3032 programs using RPC functions without referencing the TI-RPC library and the
3033 removal of the RPC headers from the glibc headers.
3034 Implemented by Ulrich Drepper.
3036 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3037 syncfs, setns, sendmmsg
3039 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3040 yue_HK, lij_IT, mhr_RU
3042 * New encodings: CP770, CP771, CP772, CP773, CP774
3044 * New program sotruss to trace calls through PLTs
3045 Implemented by Ulrich Drepper.
3047 * The malloc hook implementation is marked deprecated and will be removed
3048 from the default implementation in the next version. The design never
3049 worked ever since the introduction of threads. Even programs which do
3050 not create threads themselves can use multiple threads created internally.
3054 * The following bugs are resolved with this release:
3056 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3057 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3058 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3059 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3062 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3064 * POWER7 optimizations: memset, memcmp, strncmp
3066 * New optimized string functions for x86-64: strnlen (SSE2),
3067 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3068 Implemented by Ulrich Drepper.
3072 * The following bugs are resolved with this release:
3074 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3075 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3076 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3077 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3078 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3079 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3080 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3081 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3083 * New interfaces: pthread_getname_np, pthread_setname_np
3085 * New Linux interface: recvmmsg
3087 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3089 * The dynamic linker now recognizes supported ABI versions from the
3090 EI_ABIVERSION field in the ELF header.
3091 Implemented by Ulrich Drepper.
3093 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3094 will not be used to fill in password fields in the passwd.byname replies.
3095 Instead it is used to synthesize the shadow.byname table, should it be
3096 missing. This is a useful mode in some installations involving Solaris.
3097 Implemented by Ulrich Drepper.
3099 * New locales: kok_IN, sq_MK, cv_RU
3103 * The following bugs are resolved with this release:
3105 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3106 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3107 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3108 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3109 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3110 10789, 10817, 10823, 10840, 10847
3112 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3114 Implemented by Ulrich Drepper.
3116 * Checking version of longjmp added that fails if an uninitialized stack
3117 frame would be created. Implemented by Ulrich Drepper.
3119 * STT_GNU_IFUNC is now supported in static executables.
3120 Implemented by H.J. Lu.
3122 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3124 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3125 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3126 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3127 Contributed by H.J. Lu.
3129 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3130 Implemented by Ulrich Drepper.
3132 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3134 Contributed by H.J. Lu.
3136 * Support for fma instruction in AVX on x86-64.
3137 Implemented by H.J. Lu and Ulrich Drepper.
3139 * AVX support in x86-64 auditing support in ld.so.
3140 Implemented by H.J. Lu.
3142 * STB_GNU_UNIQUE support added.
3143 Implemented by Ulrich Drepper.
3145 * Implement second fallback mode for optimized DNS lookup for even more
3146 broken environments. If two requests from the same port are not
3147 handled correctly close the socket and open a new one before sending
3148 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3149 can be used to select this mode right away, instead of rediscovering the
3150 necessity is every process again.
3151 Implemented by Ulrich Drepper.
3153 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3154 Implemented by Adam Tkac.
3156 * Optimized iconv conversions for S390x.
3157 Implemented by Andreas Krebbel.
3159 * Using condvars with PI mutexes is now more efficient due to kernel
3160 support for requeueing to PI futexes. NPTL support added for x86-64.
3161 Implemented by Ulrich Drepper.
3163 * New locales: ps_AF, my_MM
3167 * The following bugs are resolved with this release:
3169 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3170 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3171 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3172 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3173 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3175 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3176 Implemented by Ulrich Drepper.
3178 * New Linux interfaces: accept4, fallocate, fallocate64.
3179 Implemented by Ulrich Drepper.
3181 * Correct declarations of string function when used in C++ code. This
3182 could lead to compile errors for invalid C++ code.
3184 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3187 * New POSIX 2008 interface: psiginfo
3188 Implemented by Ulrich Drepper.
3190 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3191 Implemented by Ulrich Drepper.
3193 * Support for selecting between multiple function definitions at runtime
3194 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3196 * The libcrypt library can now use the hash function implementations in
3197 NSS. Implemented by Ulrich Drepper.
3199 * The malloc implementation can be compiled to be less memory efficient
3200 but higher performing in multi-threaded programs.
3201 Implemented by Ulrich Drepper.
3203 * New locales: nan_TW@latin, ks_IN
3205 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3206 Implemented by Ulrich Drepper.
3208 * Extended printf hook support. It is possible to use user-defined types
3209 and extend existing format specifiers.
3210 Implemented by Ulrich Drepper.
3212 * Handling for group shadow files has been added.
3213 Implemented by Ulrich Drepper.
3215 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3216 servers (the case, e.g., for some people using the built-in DNS
3217 server in ADSL modems/routers). There is a once-per-process timeout
3218 in case of a broken server. To avoid it, users can run nscd or put
3219 'options single-request' in /etc/resolv.conf.
3220 Implemented by Ulrich Drepper.
3224 * The following bugs are resolved with this release:
3226 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3227 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3228 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3229 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3230 6968, 6974, 6980, 6995, 7008, 7009, 7029
3232 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3233 up at the same time. Implemented by Ulrich Drepper.
3235 * TLS descriptors for LD and GD on x86 and x86-64.
3236 Implemented by Alexandre Oliva.
3238 * getaddrinfo now handles DCCP and UDPlite.
3239 Implemented by Ulrich Drepper.
3241 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3242 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3243 Implemented by Ulrich Drepper.
3245 * New implementation of memmem, strstr, and strcasestr which is O(n).
3246 Implemented by Eric Blake.
3248 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3250 * Implement "e" option for popen to open file descriptor with the
3251 close-on-exec flag set. Implemented by Ulrich Drepper.
3253 * Implement "b" mode for fmemopen. In this mode writes writes don't
3254 implicitly add a NUL byte and seeks from the end of the buffer really
3255 use the buffer end, not the string length as the basis.
3256 Implemented by Ulrich Drepper.
3258 * Many functions, exported and internal, now atomically set the close-on-exec
3259 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3261 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3263 Implemented by Pravin Satpute.
3265 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3269 * The following bugs are resolved with this release:
3271 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3272 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3273 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3274 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3275 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3276 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3277 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3279 * New locales: bo_CN, bo_IN, shs_CA.
3281 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3283 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3284 Implemented by Pravin Satpute.
3286 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3288 * nscd caches now all timeouts for DNS entries
3289 Implemented by Ulrich Drepper.
3291 * nscd is more efficient and wakes up less often.
3292 Implemented by Ulrich Drepper.
3294 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3295 vdprintf, and obstack_vprintf.
3296 Implemented by Jakub Jelinek.
3298 * Faster memset for x86-64.
3299 Implemented by Harsha Jagasia and H.J. Lu.
3301 * Faster memcpy on x86.
3302 Implemented by Ulrich Drepper.
3304 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3305 Implemented by Ulrich Drepper.
3307 * Faster sqrt and sqrtf implemention for some PPC variants.
3308 Implemented by Steven Munroe.
3312 * The following bugs are resolved with this release:
3314 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3315 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3316 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3317 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3318 5071, 5103, 5104, 5112, 5113, 5184, 5186
3320 * More checking functions: fread, fread_unlocked, open*, mq_open.
3321 Implemented by Jakub Jelinek and Ulrich Drepper.
3323 * Extend fortification to C++. Implemented by Jakub Jelinek.
3325 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3326 by not recognizing 'a' as a modifier when those specs are requested.
3327 Implemented by Jakub Jelinek.
3329 * PPC optimizations to math and string functions.
3330 Implemented by Steven Munroe.
3332 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3333 options to be passed. Implemented by Ulrich Drepper.
3335 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3337 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3338 Implemented by Ulrich Drepper.
3340 * Handle private futexes in the NPTL implementation.
3341 Implemented by Jakub Jelinek and Ulrich Drepper.
3343 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3344 Implemented by Roland McGrath and Ulrich Drepper.
3346 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3348 * SHA-256 and SHA-512 based password encryption.
3349 Implemented by Ulrich Drepper.
3351 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3352 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3355 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3356 Implemented by Ulrich Drepper.
3360 * The following bugs are resolved with this release:
3362 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3363 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3364 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3365 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3366 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3367 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3368 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3371 * New Linux interfaces: epoll_pwait, sched_getcpu.
3373 * New generic interfaces: strerror_l.
3375 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3379 * The following bugs are resolved with this release:
3381 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3382 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3383 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3384 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3385 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3386 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3387 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3388 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3389 3225, 3251, 3252, 3253, 3273
3391 * For Linux, the sorting of addresses returned by getaddrinfo now also
3392 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3393 Implemented by Ulrich Drepper.
3395 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3396 Implemented by Ulrich Drepper.
3398 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3400 * New iconv module for MIK. Contributed by Alexander Shopov.
3402 * For sites with broken group and/or passwd database, the auto-propagate
3403 option of nscd can prevent creating ID lookup entries from the results
3404 of a name lookup and vice versa. This usually is no problem but some
3405 site might have problems with the default behavior.
3406 Implemented by Ulrich Drepper.
3408 * Iterating over entire database in NIS can be slow. With the
3409 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3410 to trade time for memory. The entire database will be read at once.
3411 Implemented by Ulrich Drepper.
3413 * The interfaces introduced in RFC 3542 have been implemented by
3416 * Support for the new ELF hash table format was added by Ulrich Drepper.
3418 * Support for priority inheritance mutexes added by Jakub Jelinek and
3421 * Support for priority protected mutexes added by Jakub Jelinek.
3423 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3424 or_IN, csb_PL, fy_NL, sr_ME.
3428 * More overflow detection functions.
3430 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3431 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3433 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3434 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3435 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3436 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3437 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3438 by Masahide Washizawa.
3440 * It is now possible to install audit libraries for the dynamic linker using
3441 LD_AUDIT. Implemented by Ulrich Drepper.
3443 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3444 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3445 For a libc and libpthread that works well on Linux 2.4 kernels, we
3446 recommend using the stable 2.3 branch.
3448 * The new function open_wmemstream acts like open_memstream,
3449 but uses a wchar_t wide character buffer.
3451 * The new function ppoll is an improved version of poll, similar to pselect.
3453 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3454 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3455 renameat, unlinkat, linkat, symlinkat, readlinkat.
3457 * New Linux kernel system calls: unshare,
3458 inotify_init, inotify_add_watch, inotify_rm_watch.
3460 * The euidaccess function is now also known by the alias eaccess,
3461 for compatibility with some other systems.
3463 * Timezone data updated to 2006b version.
3467 * The following bugs are resolved with this release:
3469 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3470 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3471 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3472 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3473 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3474 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3476 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3478 * As of this release, GCC 4 can be used to compile the C Library.
3480 * Timezone data updated to 2005m version.
3484 * The following bugs are resolved with this release:
3486 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3487 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3488 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3489 777, 787, 821, 822, 823, 825
3491 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3495 * Support for RFC 3678. Real implementations exist only for Linux so far.
3496 Implemented by Ulrich Drepper.
3498 * nscd can now cache entries persistently. Expiring entries are reloaded.
3499 For speedups the cache can be shared in memory with client processes.
3500 Implemented by Ulrich Drepper.
3502 * nscd can now perform SELinux checks.
3503 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3505 * getaddrinfo queries are now cached. Canonical name lookup is performed
3507 Implemented by Ulrich Drepper.
3509 * The nothrow function attribute is used when headers are used by gcc when
3510 compiling C code. This can avoid significant amounts of exception
3513 * The malloc functions perform more error checking and are stricter when
3514 it comes to reacting on errors. The default action is to terminate
3515 the process after showing an error message. Implemented by Ulrich Drepper.
3517 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3518 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3520 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3521 namespaces using the new function dlmopen(). This feature is of course,
3522 like most other dynamic loading functionality, not available in statically
3523 linked applications. Implemented by Ulrich Drepper.
3525 * Low-overhead boundary checking variants of string and some stdio functions
3526 were added. These are to be used in conjunction with a gcc patch by
3527 Jakub Jelinek which adds calls to these functions if possible.
3528 Implemented by Jakub Jelinek and Ulrich Drepper.
3530 * Old code for several operating systems and machine architectures that
3531 have not been in working condition in a long time have been removed from
3532 the main source tree maintained by the GNU C Library's maintainers.
3533 These files are now reside in the separate `ports' source module
3534 that is usable as an add-on when building the library.
3538 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3539 interrogate the dynamic linker, compatible with the Solaris interface.
3541 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3542 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3544 * getifaddrs now uses the netlink interface on Linux to get its information.
3545 Implemented by Thorsten Kukuk.
3547 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3548 Implemented by Ulrich Drepper.
3550 * support for non-executable stacks on x86 has been added. Changes mostly
3553 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3556 * getaddrinfo now performs destination address selection according to
3561 * Thread-safe interfaces for many functions that access locale data
3562 were added in version 2.3, but these features were omitted from NEWS.
3563 Many functions have variants with an `_l' suffix that take a `locale_t'
3564 object as a parameter rather than consulting the current locale.
3565 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3566 create and maintain `locale_t' objects. Additionally, the new function
3567 `uselocale' sets "the current locale" (as used by functions not so
3568 parameterized) set for an individual thread. These features were added
3569 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3571 * The functions getresuid, getresgid, setresuid, and setresgid, which
3572 have long been available on Linux, are now declared in <unistd.h>
3573 and are now also available on the Hurd.
3575 * ELF thread-local storage support (TLS) now works on x86-64.
3577 * The new dynamic string token $LIB is expanded in shared library names.
3578 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3580 * Aldy Hernandez contributed complete software floating point support for
3581 PowerPC machines with no FPU.
3583 * fexecve is implemented on Linux.
3585 * The `btowc' function should work at least twice as fast due to
3586 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3588 * With approriate thread add-ons cancelable functions are now implemented
3589 in libc.so as well. No need to call the function in libpthread. This
3590 change allowed to finally disable the incorrect and expensive handling
3591 of weak definition in ld.so.
3593 * Yet more PLT entries in libc.so have been removed. We finally arrived
3594 at the bare minimum. Startup times improved appropriately.
3596 * Support for the new Linux/x86 system call interface was added. The
3597 AT_SYSINFO auxiliary vector entry is recognized and handled.
3601 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3604 * iconv (the program and the interface) now accepts empty names (excluding
3605 options like //TRANSLIT) to mean "use charset of current locale".
3607 * localedef can now transliterate characters in strings which are not in
3608 the provided charmap. The information from the input locale is used.
3610 * Prelinking support was added for ELF targets. This requires additional
3611 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3613 * Read-only stdio streams now use mmap to speed up operation by eliminating
3614 copying and buffer underflows. To use add 'm' to the mode string of
3615 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3617 * The malloc functions were completely rewritten by Wolfram Gloger based
3618 on Doug Lea's malloc-2.7.0.c.
3620 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3621 implementation of regex.
3623 * Bruno Haible upgraded the iconv and locale implementation to support
3626 * Contents of the LC_* and LANG environment variables in the CEN style are
3627 not recognized anymore. It never was used. Change by Ulrich Drepper.
3629 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3630 thread-local storage (TLS) ABI on some platforms.
3631 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3633 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3634 EUC-JISX0213, and TSCII.
3636 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3637 BSD-compatible interface for getting all network interface addresses.
3638 Implementation for IPv4 by Roland McGrath.
3640 * Loading of locale data is faster due to the introduction of a locale
3641 archive. Implemented by Roland McGrath and Ulrich Drepper.
3643 * Startup times are significantly reduced by not using exported functions
3644 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3647 * Steven Munroe contributed a port to PowerPC64/Linux.
3651 * The Hurd now uses the GNU libio implementation of stdio rather than the
3652 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3654 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3655 with the same behavior as the Linux system call of the same name.
3659 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3660 128-bit long double format.
3662 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3663 IBM1161, and IBM1162 charsets.
3665 * Andreas Jaeger contributed a port to x86-64/Linux.
3667 * Peter Bruin contributed a port to PowerPC/Hurd.
3669 * libc functions using I/O streams now can handle wide-oriented streams
3672 * optimizations in the dynamic linker. Binaries created by recent binutils
3673 versions start up quicker due to reduced time spent on relocations.
3675 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3679 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3680 asinh, atanh, j0 for the 128-bit long double format.
3682 * Bruno Haible updated all the code handling Unicode in some form to
3683 support Unicode 3.1.
3685 * Speed of regex for single-byte locales is back to previous levels.
3686 Patch by Isamu Hasegawa.
3688 * Alpha, SPARC, and IA-64 now also using floating stacks.
3690 * Startup time of internationalized applications greatly improved through
3691 iconv cache. Use iconvconfig to generate the cache file.
3692 Contributed by Ulrich Drepper.
3694 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3695 severe performance problems. Patch by David Mosberger.
3697 * The Hurd port got a lot more functionality like AIO, various stdio
3698 extensions, etc. Mainly done by Roland McGrath.
3700 * mtrace can now lookup symbols in shared libraries.
3704 * Intel's IA-64 math library is largely integrated. It provides fast and
3705 accurate implementatations for most basic and standard math functions
3706 in float, double, and long double format.
3708 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3709 and asin for the 96-bit long double format and asin, log, tan for the
3710 128-bit long double format.
3712 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3713 The basic double functions exist today. Contributed by Abraham Ziv
3714 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3715 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3717 * An asynchronous name lookup library was added. The interface is designed
3718 after POSIX AIO. The proposal was circulated beforehand to get comments.
3719 No negative ones came in. Implemented by Ulrich Drepper.
3721 * Port to S390/64bit contributed by Martin Schwidefsky
3722 <schwidefsky@de.ibm.com>.
3724 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3725 of functions for Linux/IA-64.
3727 * The RPC code is now thread safe. Threads can now use the same service
3728 of different services at the same time. Patch by Eric Norum
3729 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3731 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3732 family of functions for Linux/S390.
3734 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3735 of functions for Linux/x86.
3737 * Port to Linux/CRIS contributed by Axis Communications.
3741 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3742 we can now check for namespace violations and missing declarations. The
3743 result is that almost all headers are now Unix-compliant (as defined in
3744 the upcoming XPG6). The negative side is that some programs might need
3745 corrections, too, if they depend on the incorrect form of the headers in
3746 previous versions which defined too many symbols and included too many
3749 * regex now handles multibyte character sets correctly.
3750 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3752 * iconv (the program) does now conform to the upcoming XPG6 and handles
3753 charmaps. Instead of the charset names the path of charmaps can be
3754 provided and the conversion happens based on this data.
3755 Contributed by Ulrich Drepper.
3757 * The locale program now provides detailed information about the installed
3762 only lists the names of the supported locales
3766 provides details such as country, language, and codeset name.
3767 Contributed by Ulrich Drepper.
3771 * The gencat program now parses the input file according to the charset
3772 selected by the LC_CTYPE category. This is important for stateful
3773 character sets. To make generating catalogs easier there is a way
3774 to overwrite the charset selected by the locale: before the first
3775 message or $ quote line the catalog can contain a line like
3777 $ codeset=ISO-8859-2
3779 to select the charset (ISO-8859-2 in this case).
3781 Implemented by Shinya Hanataka and Ulrich Drepper.
3783 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3784 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3785 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3787 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3788 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3790 * The LANGUAGE environment variable is now ignored unless the locale is
3791 changed from the default "C" locale.
3793 * The usual bug fixes.
3797 * Greg McGary added runtime support for bounds checking using gcc's
3798 new -fbounded-pointers option. ix86 target is complete. PowerPC
3801 * Thorsten Kukuk added secure mode support to nscd.
3803 * The Berkeley db libraries have been removed.
3805 Related, the nss_db module is now in a separate package since it
3806 obviously requires a database library being available.
3808 * Wide character I/O streams implemented by Ulrich Drepper.
3810 * Functions from the extended socket API added by Ulrich Drepper.
3812 * Functions feenableexcept and fedisableexcept to control the
3813 behaviour of individual exceptions have been added by Andreas Jaeger.
3815 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3817 * The resolver code has been updated from bind 8.2.3-T5B which supports
3818 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3821 This change could in some situations effect backward compatibility. Since
3822 now `_res' is a thread-local instead of a global variable, modifying it
3823 in one thread does not have any effect in other threads.
3825 The resolver library was also extended to allow IPv6 as the transport
3826 protocol for the requests. This work was done by Stig Venaas.
3828 * Compatibility code for K&R C compilers has been removed from the
3829 header files. A ISO C compiler is needed to use the library
3830 (conforming to either C89 or C99 standard).
3832 * Complete rewrite of the localedef program to support multibyte character
3833 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3834 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3835 Implemented by Ulrich Drepper.
3837 Bruno Haible significantly improved the generation and use of the data
3838 structures for the wide character tables.
3840 * Plural handling in gettext implemented by Ulrich Drepper.
3842 * The utmp daemon has been removed.
3844 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3846 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3849 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3851 * POSIX barriers implemented by Kaz Kylheku.
3853 * POSIX spawn function family implemented by Ulrich Drepper.
3855 * POSIX spinlocks are now available.
3857 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3859 * the configure option --enable-kernel=X.Y.Z allows to strip out
3860 compatibility for kernel versions before X.Y.Z. This is currently only
3861 implemented for Linux.
3863 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3864 field sin6_scope_id. This means that all programs using IPv6 should be
3865 recompiled. Don't expect binary compatibility with previous glibc
3868 * various conversion modules for IBM character sets contributed by
3871 * IA-64 port by Jes Sorensen and HJ Lu.
3885 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3887 * New xdr functions are added; some rpc functions are now 64bit clean.
3889 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3891 * Fixed known incompatibilities with glibc 2.0.
3893 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3895 * Optimized a number of functions (especially the ELF dynamic loader).
3897 * Update timezone data files.
3899 * lots of charmaps corrections
3901 * some new locale definitions and charmaps
3906 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3907 conform to POSIX member type requirements. Symbol versions have been
3908 adjusted as needed within the library, and for direct use by applications,
3909 but there is potential for problems if third-party libraries use
3910 struct timeval as part of their interface. This does not present
3911 a problem for X and other "essential" system libraries.
3913 * An additional locale model to support C++ Standard Library locale
3914 model and probably more was implemented by Ulrich Drepper.
3916 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3919 * Miles Bader provided the `argp' function family to support hierachical
3920 command line argument parsing, layered on top of getopt.
3922 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3924 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3927 * scanf recognizes the %a and %A format for scanning floating point numbers.
3929 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3930 information and interfaces for the available integer types.
3932 * about 130 new math functions were added to implement the ISO C9x math
3935 * the new header <complex.h> contains definitions of the complex math
3936 functions from ISO C 9X.
3938 * the new header <tgmath.h> defines generic macros to use complex or
3939 real valued functions.
3941 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3943 * Andreas Jaeger provided a test suite for the math library.
3945 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3947 * Ulrich Drepper added character set conversion functions (iconv).
3949 * Optimized string functions have been added.
3951 * The localedata addon is now part of glibc.
3953 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3955 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3956 daemon for NSS (nscd).
3958 Missing a better place here are some numbers on improvements. Under
3959 Linux 2.1.125 un-tar-ing the kernel sources takes
3963 using local files 12.19s 6.88s 22.91s
3965 using NIS 13.92s 8.91s 26.34s
3967 using NIS & nscd 10.37s 7.34s 25.30s
3969 using NIS+ 27.57s 30.37s 640.46s
3971 using NIS+ & nscd 10.25s 7.83s 26.51s
3973 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3975 Keep in mind that non-namelookup related operations dominate above times.
3976 It was just a common complain that using NIS+ unpacking the kernel is
3979 [1] The old nscd implementation is not available anymore since it was
3980 distributed with glibc up to version 2.0.98 and thus is now replaced.
3982 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3984 * Mark Kettenis provided a Hesiod NSS module.
3986 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3987 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3989 * rcmd can now handle netgroups (Dick Streefland).
3991 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3994 * Support for the IPv6 protocol has been added to the socket API, as per the
3995 latest draft standards.
3997 * Support for Linux 2.2 has been added.
3999 * Interface changes relative to the latest 2.0.x release:
4000 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4001 addseverity NEW: Unix98
4002 alphasort64 NEW: LFS
4003 argp_err_exit_status NEW: argp, GNU ext
4004 argp_error NEW: argp, GNU ext
4005 argp_failure NEW: argp, GNU ext
4006 argp_help NEW: argp, GNU ext
4007 argp_parse NEW: argp, GNU ext
4008 argp_program_bug_address NEW: argp, GNU ext
4009 argp_program_version NEW: argp, GNU ext
4010 argp_program_version_hook NEW: argp, GNU ext
4011 argp_state_help NEW: argp, GNU ext
4012 argp_usage NEW: argp, GNU ext
4013 authdes_create NEW: Secure RPC
4014 authdes_getucred NEW: Secure RPC
4015 authdes_pk_create NEW: Secure RPC
4016 backtrace NEW: GNU ext.
4017 backtrace_symbols NEW: GNU ext.
4018 backtrace_symbols_fd NEW: GNU ext.
4020 cacosf NEW: ISO C 9x
4021 cacosh NEW: ISO C 9x
4022 cacoshf NEW: ISO C 9x
4023 cacoshl NEW: ISO C 9x
4024 cacosl NEW: ISO C 9x
4031 casinf NEW: ISO C 9x
4032 casinh NEW: ISO C 9x
4033 casinhf NEW: ISO C 9x
4034 casinhl NEW: ISO C 9x
4035 casinl NEW: ISO C 9x
4037 catanf NEW: ISO C 9x
4038 catanh NEW: ISO C 9x
4039 catanhf NEW: ISO C 9x
4040 catanhl NEW: ISO C 9x
4041 catanl NEW: ISO C 9x
4042 cbc_crypt NEW: Secure RPC
4046 ccoshf NEW: ISO C 9x
4047 ccoshl NEW: ISO C 9x
4053 cimagf NEW: ISO C 9x
4054 cimagl NEW: ISO C 9x
4055 clearerr_locked REMOVED
4056 clntunix_create NEW: sunrpc ext
4058 clog10 NEW: ISO C 9x
4059 clog10f NEW: ISO C 9x
4060 clog10l NEW: ISO C 9x
4070 cprojf NEW: ISO C 9x
4071 cprojl NEW: ISO C 9x
4073 crealf NEW: ISO C 9x
4074 creall NEW: ISO C 9x
4079 csinhf NEW: ISO C 9x
4080 csinhl NEW: ISO C 9x
4083 csqrtf NEW: ISO C 9x
4084 csqrtl NEW: ISO C 9x
4088 ctanhf NEW: ISO C 9x
4089 ctanhl NEW: ISO C 9x
4091 des_setparity NEW: Secure RPC
4092 ecb_crypt NEW: Secure RPC
4093 endutxent NEW: Unix98
4095 exp10f NEW: ISO C 9x
4096 exp10l NEW: ISO C 9x
4100 fattach NEW: STREAMS
4101 fdetach NEW: STREAMS
4105 feclearexcept NEW: ISO C 9x
4106 fegetenv NEW: ISO C 9x
4107 fegetexceptflag NEW: ISO C 9x
4108 fegetround NEW: ISO C 9x
4109 feholdexcept NEW: ISO C 9x
4111 feraiseexcept NEW: ISO C 9x
4112 ferror_locked REMOVED
4113 fesetenv NEW: ISO C 9x
4114 fesetexceptflag NEW: ISO C 9x
4115 fesetround NEW: ISO C 9x
4116 fetestexcept NEW: ISO C 9x
4117 feupdateenv NEW: ISO C 9x
4118 fflush_locked REMOVED
4122 fgets_unlocked NEW: GNU ext.
4123 fileno_locked REMOVED
4135 fputc_locked REMOVED
4136 fputs_unlocked NEW: GNU ext.
4137 fread_unlocked NEW: GNU ext.
4142 fstatvfs NEW: Unix98
4146 ftruncate64 NEW: LFS
4148 fwrite_unlocked NEW: GNU ext.
4149 gai_strerror NEW: IPv6
4153 getchar_locked REMOVED
4155 getdate_err NEW: Unix98
4156 getdate_r NEW: GNU ext.
4158 getnameinfo NEW: IPv6
4159 getnetname NEW: Secure RPC
4160 getpmsg NEW: STREAMS
4161 getpt NEW: Unix98 PTY
4162 getrlimit64 NEW: LFS
4163 getutxent NEW: Unix98
4164 getutxid NEW: Unix98
4165 getutxline NEW: Unix98
4166 glob64 NEW: GNU ext.
4167 globfree64 NEW: GNU ext.
4168 gnu_get_libc_release NEW: GNU ext.
4169 gnu_get_libc_version NEW: GNU ext.
4170 grantpt NEW: Unix98 PTY
4171 host2netname NEW: Secure RPC
4173 iconv_close NEW: iconv
4174 iconv_open NEW: iconv
4175 if_freenameindex NEW: IPv6
4176 if_indextoname NEW: IPv6
4177 if_nameindex NEW: IPv6
4178 if_nametoindex NEW: IPv6
4179 in6addr_any NEW: IPv6
4180 in6addr_loopback NEW: IPv6
4181 inet6_isipv4mapped NEW: IPv6
4182 isastream NEW: STREAMS
4183 iswblank NEW: GNU ext.
4184 key_decryptsession NEW: Secure RPC
4185 key_decryptsession_pk NEW: Secure RPC
4186 key_encryptsession NEW: Secure RPC
4187 key_encryptsession_pk NEW: Secure RPC
4188 key_gendes NEW: Secure RPC
4189 key_get_conv NEW: Secure RPC
4190 key_secretkey_is_set NEW: Secure RPC
4191 key_setnet NEW: Secure RPC
4192 key_setsecret NEW: Secure RPC
4193 llrint NEW: ISO C 9x
4194 llrintf NEW: ISO C 9x
4195 llrintl NEW: ISO C 9x
4196 llround NEW: ISO C 9x
4197 llroundf NEW: ISO C 9x
4198 llroundl NEW: ISO C 9x
4203 lrintf NEW: ISO C 9x
4204 lrintl NEW: ISO C 9x
4205 lround NEW: ISO C 9x
4206 lroundf NEW: ISO C 9x
4207 lroundl NEW: ISO C 9x
4209 makecontext NEW: Unix98
4210 mempcpy NEW: GNU ext.
4213 modify_ldt NEW: kernel
4217 nearbyint NEW: ISO C 9x
4218 nearbyintf NEW: ISO C 9x
4219 nearbyintl NEW: ISO C 9x
4220 netname2host NEW: Secure RPC
4221 netname2user NEW: Secure RPC
4222 nexttoward NEW: ISO C 9x
4223 nexttowardf NEW: ISO C 9x
4224 nexttowardl NEW: ISO C 9x
4228 passwd2des NEW: Secure RPC
4230 pow10f NEW: GNU ext.
4231 pow10l NEW: GNU ext.
4234 printf_size NEW: GNU ext.
4235 printf_size_info NEW: GNU ext.
4236 profil_counter REMOVED
4237 pthread_mutexattr_getkind_np REPLACED
4238 pthread_mutexattr_setkind_np REPLACED
4239 ptsname NEW: Unix98 PTY
4240 ptsname_r NEW: Unix98 PTY
4242 putchar_locked REMOVED
4243 putgrent NEW: GNU ext.
4245 putpmsg NEW: STREAMS
4246 pututxline NEW: Unix98
4250 readdir64_r NEW: LFS
4251 remquo NEW: ISO C 9x
4252 remquof NEW: ISO C 9x
4253 remquol NEW: ISO C 9x
4255 roundf NEW: ISO C 9x
4256 roundl NEW: ISO C 9x
4258 scalbln NEW: ISO C 9x
4259 scalblnf NEW: ISO C 9x
4260 scalblnl NEW: ISO C 9x
4262 sendfile NEW: kernel
4263 setcontext NEW: Unix98
4264 setrlimit64 NEW: LFS
4265 setutxent NEW: Unix98
4267 sigignore NEW: Unix98
4268 sigqueue NEW: POSIX.1b
4269 sigrelse NEW: Unix98
4270 sigset NEW: POSIX.1b
4271 sigtimedwait NEW: POSIX.1b
4272 sigwaitinfo NEW: POSIX.1b
4273 sincos NEW: GNU ext.
4274 sincosf NEW: GNU ext.
4275 sincosl NEW: GNU ext.
4279 strcasestr NEW: GNU ext.
4280 strtoimax NEW: ISO C 9x
4281 strtoumax NEW: ISO C 9x
4282 strverscmp NEW: GNU ext.
4283 svcauthdes_stats NEW: Secure RPC
4284 svcunix_create NEW: sunrpc etx
4285 svcunixfd_create NEW: sunrpc ext
4286 swapcontext NEW: Unix98
4287 tcgetsid NEW: Unix98 PTY
4288 tdestroy NEW: GNU ext.
4289 tgamma NEW: ISO C 9x
4290 tgammaf NEW: ISO C 9x
4291 tgammal NEW: ISO C 9x
4295 truncf NEW: ISO C 9x
4296 truncl NEW: ISO C 9x
4298 unlockpt NEW: Unix98 PTY
4299 updwtmpx NEW: Unix98
4300 user2netname NEW: Secure RPC
4301 utmpxname NEW: Unix98
4302 versionsort NEW: GNU ext.
4303 versionsort64 NEW: GNU ext.
4305 wcscasecmp NEW: GNU ext.
4306 wcsncasecmp NEW: GNU ext.
4307 wcsnlen NEW: GNU ext.
4308 wcstoimax NEW: ISO C 9x
4309 wcstoll NEW: ISO C 9x
4310 wcstoull NEW: ISO C 9x
4311 wcstoumax NEW: ISO C 9x
4313 wordexp NEW: POSIX.2
4314 wordfree NEW: POSIX.2
4315 write_profiling REMOVED
4316 xdecrypt NEW: Secure RPC
4317 xdr_authdes_cred NEW: Secure RPC
4318 xdr_authdes_verf NEW: Secure RPC
4319 xdr_cryptkeyarg NEW: Secure RPC
4320 xdr_cryptkeyarg2 NEW: Secure RPC
4321 xdr_cryptkeyres NEW: Secure RPC
4322 xdr_getcredres NEW: Secure RPC
4323 xdr_key_netstarg NEW: Secure RPC
4324 xdr_key_netstres NEW: Secure RPC
4325 xdr_keybuf NEW: Secure RPC
4326 xdr_keystatus NEW: Secure RPC
4327 xdr_netnamestr NEW: Secure RPC
4328 xdr_sizeof NEW: Secure RPC
4329 xdr_unixcred NEW: sunrpc ext
4330 xencrypt NEW: Secure RPC
4331 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4342 * inet_ntoa is thread-safe
4344 * updwtmp is moved from libutil to libc
4346 * rewrite of cbrt function
4348 * update of timezone data
4362 * add atoll function
4364 * fix complex problems in Berkeley DB code
4366 * fix math functions
4370 * fixed lots of header problems (especially Linux/GNU specific)
4372 * dynamic loader preserves all registers
4374 * Roland McGrath provided support for handling of auxiliary objects in
4375 the ELF dynamic loader.
4377 * support for parallel builds is improved
4381 * GNU extensions are no longer declared by default. To enable them you
4382 must define the macro `_GNU_SOURCE' in your program or compile with
4385 * The library has changed from using GNU ld symbol aliases to using weak
4386 symbols where available. The ELF object file format supports weak
4387 symbols; GNU ld also supports weak symbols in the a.out format. (There
4388 is also now support for other GNU ld extensions in ELF. Use the
4389 `--with-elf' option to configure to indicate you have ELF, and
4390 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4391 of many files which contained only symbol aliases, reducing the size of
4392 the source and the compiled library; many other files were renamed to
4393 less cryptic names previously occupied by the symbol alias files.
4394 There is a new header file <elf.h> for programs which operate on
4395 files in the ELF format.
4397 * Converted to Autoconf version 2, so `configure' has more options.
4398 Run `configure --help' to see the details.
4400 * The library can now be configured to build profiling, highly-optimized
4401 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4402 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4403 `configure' enable building these extra libraries. The shared library is
4404 built by default when using both ELF and GNU ld. When shared libraries
4405 are enabled, the new library `-ldl' is available for arbitrary run-time
4406 loading of shared objects; its interface is defined in <dlfcn.h>. The
4407 new header file <link.h> gives access to the internals of the run-time
4408 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4409 application of same name on other systems and it provides information
4410 about dynamically linked binaries.
4412 * The C library now provides the run-time support code for profiling
4413 executables compiled with `-pg'. Programs can control the profiling code
4414 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4415 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4416 the support for this file format was contributed by David Mosberger-Tang.
4418 * The math code has been replaced with a math library based on fdlibm from
4419 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4420 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4421 functions. The math functions now reside in a separate library, so
4422 programs using them will need to use `-lm' their linking commands.
4424 * John C. Bowman contributed optimized ix87 assembler inline functions.
4426 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4427 mechanism similar to that found in Solaris 2. This is now used for the
4428 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4429 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4430 file controls what services are used for each individual database. This
4431 works by loading shared libraries with names specified in `nsswitch.conf',
4432 so service modules can be changed or added at any time without even
4433 relinking any program. Currently there are the file, db, and NIS based
4434 NSS services available.
4436 * The new functions `strtoq' and `strtouq' parse integer values from
4437 strings, like `strtol' and `strtoul', but they return `long long int' and
4438 `unsigned long long int' values, respectively (64-bit quantities).
4440 * The new functions `strtof' and `strtold' parse floating-point values from
4441 strings, like `strtod', but they return `float' and `long double' values,
4442 respectively (on some machines `double' and `long double' are the same).
4444 * Ulrich Drepper has contributed new implementations of the floating-point
4445 printing and reading code used in the `printf' family of functions and
4446 `strtod', `strtof', and `strtold'. These new functions are perfectly
4447 accurate, and much faster than the old ones.
4449 * The implementation of the POSIX locale model was completely rewritten by
4450 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4451 to compile the POSIX locale definition.
4453 * The former dummy implementations of the strcoll and strxfrm function are
4454 now replaced by fully functional code contributed by Ulrich Drepper. The
4455 collation information comes from the POSIX locale definitions.
4457 * The new header <langinfo.h> defines an interface for accessing
4458 various locale-dependent data (using the locale chosen with `setlocale').
4460 * Ulrich Drepper has contributed a new suite of functions for operation on
4461 wide-character and multibyte-character strings, in <wchar.h>;
4462 and classification and case conversion of wide characters, in <wctype.h>.
4463 These new functions are conforming to the ISO C, Amendement 1 specification.
4465 * There is now a second implementation of the standard I/O library available.
4466 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4467 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4469 * You can now use positional parameter specifications in format strings
4470 for the `printf' and `scanf' families of functions. For example,
4471 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4472 ``Number 6, Mr Jones''. This is mainly useful when providing different
4473 format strings for different languages, whose grammars may dictate
4474 different orderings of the values being printed. To support this
4475 feature, the interface for `register_printf_handler' has changed; see
4476 the header file <printf.h> for details.
4478 * The `printf' and `scanf' families of functions now understand a new
4479 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4480 to group numbers as indicated by the locale; for `scanf' and friends, this
4481 says to accept as valid only a number with all the proper grouping
4482 separators in the right places. In the default "C" locale, numbers are
4483 not grouped; but locales for specific countries will define the usual
4484 conventions (i.e. separate thousands with `,' in the US locale).
4486 * The pgrp functions have been regularized, slightly incompatibly but much
4487 less confusingly. The core functions are now `getpgid' and `setpgid',
4488 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4489 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4490 provided for compatibility. There is no longer an incompatible `getpgrp'
4491 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4493 * The new header file <fts.h> and suite of functions simplify programs that
4494 operate on directory trees. This code comes from 4.4 BSD.
4496 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4497 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4498 NSS scheme used in glibc.
4500 * There is a new malloc debugging hook `__memalign_hook'.
4502 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4503 `unsigned int' in <sys/types.h>. These are for compatibility only and
4504 their use is discouraged.
4506 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4507 done differently, so that it works even without GNU ld.
4509 * New function `euidaccess' checks allowed access to a file like `access',
4510 but using the effective IDs instead of the real IDs.
4512 * The time zone data files have been updated for the latest and greatest
4513 local time conventions of the countries of the world.
4515 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4518 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4519 for formatting floating-point numbers. They are provided only for
4520 compatibility; new programs should use `sprintf' instead. There are
4521 also equivalent function for the `long double' floating-point type and
4522 all functions also exist in a reentrant form.
4524 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4525 functions for maintaining the login-record files (primarily of use to
4526 system programs such as `login'), and convenient functions for
4527 allocating and initializing a pseudo-terminal (pty) device.
4529 * Ulrich Drepper has contributed new support for System V style
4530 shared memory and IPC on systems that support it.
4532 * Ulrich Drepper has contributed several miscellaneous new functions found
4533 in System V: The `hsearch' family of functions provide an effective
4534 implementation of hash tables; `a64l' and `l64a' provide a very simple
4535 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4538 * Ulrich Drepper has contributed new reentrant counterparts for the
4539 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4541 * Ulrich Drepper has contributed new, highly-optimized versions of several
4542 string functions for the i486/Pentium family of processors.
4544 * Ulrich Drepper has updated the Linux-specific code, based largely
4545 on work done in Hongjiu Lu's version of GNU libc for Linux.
4546 The GNU library now supports Linux versions 2.0.10 and later,
4547 using the ELF object file format (i[3456]86-*-linux).
4549 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4551 * David Mosberger-Tang and Richard Henderson have ported the C library
4552 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4553 dynamic linking support for ELF/Alpha.
4555 * Richard Henderson contributed several Alpha optimized assembler function
4556 for arithmetic and string handling.
4558 * Ulrich Drepper has contributed a new set of message catalog functions to
4559 support multiple languages using the <libintl.h> interface, for use with
4560 his new package GNU gettext. Translation volunteers have contributed
4561 catalogs of the library's messages in Spanish, German, and Korean.
4563 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4564 program and the `catgets' function for reading the catalog files it
4565 creates. (The <libintl.h> interface is preferred; we include the
4566 <nl_types.h> interface using `catgets' only for source compatibility with
4567 programs already written to use it.)
4569 * New header file <values.h> gives SVID-compatible names for <limits.h>
4572 * Various new macros, declarations, and small header files for compatibility
4575 * New function `group_member' is a convenient way to check if a process has
4576 a given effective group ID.
4578 * When using GCC 2.7 and later, the socket functions are now declared in a
4579 special way so that passing an argument of type `struct sockaddr_in *',
4580 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4581 `struct sockaddr *' type, does not generate a type-clash warning.
4583 * New function `error' declared in header file <error.h> is a convenient
4584 function for printing error messages and optionally exiting; this is the
4585 canonical function used in GNU programs. The new functions `err', `warn',
4586 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4587 doing the same thing.
4589 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4590 POSIX.2 `glob' function to do ~ and {...} expansion.
4592 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4593 `clearenv' which is used in POSIX.9 is also available.
4595 * New function `getsid' returns session ID number on systems that support it.
4597 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4598 files <db.h> and <mpool.h> provide a rich set of functions for several
4599 types of simple databases stored in memory and in files, and <ndbm.h> is
4600 an old `ndbm'-compatible interface using the `db' functions. Link with
4601 `-ldb' to get these functions.
4603 * New macro `strdupa' copies a string like `strdup', but uses local stack
4604 space from `alloca' instead of dynamic heap space from `malloc'.
4606 * New function `strnlen' is like `strlen' but searches only a given maximum
4607 number of characters for the null terminator. `stpncpy', `strndup' and
4608 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4611 * New function `statfs' in header <sys/statfs.h>.
4613 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4614 provide convenient functions for operating on blocks of null-terminated
4617 * A new suite of functions in <utmp.h> handle all the details of reading
4618 and writing the utmp file.
4620 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4623 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4624 completely POSIX compliant and also implemented the extended functionality
4625 to handle alternate digit representation and alternate era date formats.
4627 * Ulrich Drepper provided an implementation of the `strptime' function
4628 defined in XPG4.2 which transforms a string into a `struct tm' value.
4630 * Paul Eggert provided the tzselect shell script as part of the timezone
4631 code. The shell script makes it easy to select the correct timezone
4634 * The implementation of the malloc family of functions is completely replaced
4635 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4636 The implementation uses the mmap function (if available) and it is
4637 optimized for the use in multi threaded programs.
4639 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4640 functions. This new functionality is usable by specifying a special
4641 salt string and it is compatible with implementation on *BSD systems.
4643 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4644 `getsubopt' to handle second level command line options, `bsd_signal'
4645 to access BSD style `signal' functionality, the obsolete `regexp' style
4648 * the `lchown' function is available on system which support this
4651 * The implementation of the shadow password handling function was contributed
4654 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4656 * POSIX.1g support was added. The <sys/select.h> header is available,
4657 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4658 implementation of `getaddrinfo'.
4662 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4664 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4665 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4667 * The new function `daemon' from 4.4 BSD is useful for server programs that
4668 want to put themselves in the background.
4670 * Joel Sherrill has contributed support for several standalone boards that
4671 run without an operating system.
4673 * `printf', `scanf' and friends now accept a `q' type modifier for long
4674 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4676 * All of the code taken from BSD (notably most of the math and networking
4677 routines) has been updated from the BSD 4.4-Lite release.
4679 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4681 * The new functions `getdomainname' and `setdomainname' fetch or change the
4682 YP/NIS domain name. These are system calls which exist on systems which
4685 * The time zone data files have been updated for the latest international
4688 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4689 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4693 * The C library now includes support for Sun RPC, from Sun's free
4694 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4695 are included. (There is still no support for YP.)
4697 * Tom Quinn has contributed a port of the C library to SGI machines running
4698 Irix 4 (mips-sgi-irix4).
4700 * The new `lockf' function is a simplified interface to the locking
4701 facilities of `fcntl', included for compatibility.
4703 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4705 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4707 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4710 * The error message printed by `assert' for a failed assertion now includes
4711 the name of the program (if using GNU ld) and the name of the calling
4712 function (with versions of GCC that support this).
4714 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4716 * The library now includes the <sys/mman.h> header file and memory
4717 management functions `mmap', `munmap', `mprotect', `msync', and
4718 `madvise', on systems that support those facilities.
4720 * The interface for `mcheck' has changed slightly: the function called to
4721 abort the program when an allocation inconsistency is detected now takes
4722 an argument that indicates the type of failure. The new function
4723 `mprobe' lets you request a consistency check for a particular block at
4724 any time (checks are normally done only when you call `free' or `realloc'
4727 * It is now possible to easily cross-compile the C library, building on one
4728 system a library to run on another machine and/or operating system. All
4729 you need to do is set the variable `HOST_CC' in `configparms' to the
4730 native compiler for programs to run on the machine you are building on (a
4731 few generator programs are used on Unix systems); set `CC' to the
4734 * The new function `fexecve' (only implemented on the GNU system) executes
4735 a program file given a file descriptor already open on the file.
4739 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4740 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4742 * You can set the variable `libprefix' in `configparms' to specify a prefix
4743 to be prepended to installed library files; this makes it easy to install
4744 the GNU C library to be linked as `-lgnuc' or whatever.
4746 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4747 copies a limited number of characters from a string, and returns the
4748 address of the last character written.
4750 * You no longer need to check for whether the installed `stddef.h' is
4751 compatible with the GNU C library. configure now checks for you.
4753 * You can now define a per-stream `fileno' function to convert the
4754 stream's cookie into an integral file descriptor.
4756 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4757 allocates zero bytes of storage, and returns a unique pointer which
4758 you can pass to `realloc' or `free'. The behavior is undefined if
4759 you dereference this pointer.
4761 * The C library now runs on Sony NEWS m68k machines running either
4762 NewsOS 3 or NewsOS 4.
4764 * The new `syscall' function is a system-dependent primitive function
4765 for invoking system calls. It has the canonical behavior on Unix
4766 systems, including unreliable return values for some calls (such as
4767 `pipe', `fork' and `getppid').
4769 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4770 to `EAGAIN', which is the preferred name. On systems whose kernels
4771 use two distinct codes, the C library now translates EWOULDBLOCK to
4772 EAGAIN in every system call function.
4776 * The GNU C Library Reference Manual is now distributed with the library.
4777 `make dvi' will produce a DVI file of the printed manual.
4778 `make info' will produce Info files that you can read on line using C-h i
4779 in Emacs or the `info' program.
4780 Please send comments on the manual to bug-glibc-manual@gnu.org.
4782 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4784 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4786 * Jason Merrill has contributed a port to the Sequent Symmetry running
4787 Dynix version 3 (i386-sequent-dynix).
4789 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4790 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4792 * New function `memory_warnings' lets you arrange to get warnings when
4793 malloc is running out of memory to allocate, like Emacs gives you.
4795 * The C library now contains the relocating allocator used in Emacs 19 for
4796 its editing buffers. This allocator (ralloc) minimizes allocation
4797 overhead and fragmentation by moving allocated regions around whenever it
4798 needs to. You always refer to a ralloc'd region with a "handle" (a
4799 pointer to a pointer--an object of type `void **').
4801 * There is a new `printf' format: `%m' gives you the string corresponding
4802 to the error code in `errno'.
4804 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4805 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4806 pass, the `a' modifier says to fill in a `char **' you pass with a
4809 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4810 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4811 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4813 * `mkstemp' is a traditional Unix function to atomically create and open a
4814 uniquely-named temporary file.
4818 * The standard location for the file that says what the local timezone is
4819 has changed again. It is now `/usr/local/etc/localtime' (or more
4820 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4822 * The distribution no longer contains any files with names longer than 14
4825 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4826 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4828 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4832 * The configuration process has changed quite a bit. The `configure'
4833 script is now used just like the configuration scripts for other GNU
4834 packages. The `sysdeps' directory hierarchy is much rearranged.
4835 The file `INSTALL' explains the new scheme in detail.
4837 * The header files no longer need to be processed into ANSI C and
4838 traditional C versions. There is just one set of files to install, and
4839 it will work with ANSI or old C compilers (including `gcc -traditional').
4841 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4842 MIPS DECStation running Ultrix 4.
4844 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4845 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4846 made itself into a shared library.
4848 * Yet further improved support for the i386, running 4.3 BSD-like systems
4849 (such as Mach 3 with the Unix single-server), or System V.
4851 * New function `strncasecmp' to do case-insensitive string comparison
4852 with limited length.
4854 * New function `strsep' is a reentrant alternative to `strtok'.
4856 * New functions `scandir' and `alphasort' for searching directories.
4858 * New function `setenv' is a better interface to `putenv'.
4860 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4861 function for traversing a directory tree.
4863 * The GNU obstack package is now also part of the C library.
4864 The new function `open_obstack_stream' creates a stdio stream that
4865 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4866 formatted output directly to an obstack.
4868 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4869 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4871 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4873 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4874 things to your strings.
4876 * There are some new test programs: test-fseek, testmb, and testrand.
4878 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4879 These ports are not finished, but are a good starting place for really
4880 supporting those systems.
4882 * `/etc/localtime' is now the standard location for the file that says what
4883 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4884 This follows the general principle that `/etc' is the place for all local
4885 configuration files.
4887 * The C library header files now use `extern "C"' when used by the C++
4888 compiler, so the C library should now work with C++ code.
4890 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4891 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4894 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4895 function which uses the `merge sort' algorithm, and is said to be
4896 significantly faster than the old GNU `qsort' function. Merge sort is
4897 now the standard `qsort' function. The new algorithm can require a lot
4898 of temporary storage; so, the old sorting function is called when the
4899 required storage is not available.
4901 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4902 provides the Unix `crypt' function, plus some other entry points.
4904 * The code and header files taken from 4.4 BSD have been updated with the
4905 latest files released from Berkeley.
4907 ----------------------------------------------------------------------
4908 Copyright information:
4910 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4912 Permission is granted to anyone to make or distribute verbatim copies
4913 of this document as received, in any medium, provided that the
4914 copyright notice and this permission notice are preserved,
4915 thus giving the recipient permission to redistribute in turn.
4917 Permission is granted to distribute modified versions
4918 of this document, or of portions of it,
4919 under the above conditions, provided also that they
4920 carry prominent notices stating who last changed them.
4923 version-control: never