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 * The GNU C Library can now be compiled with support for Intel CET, AKA
13 Intel Control-flow Enforcement Technology. When the library is built
14 with --enable-cet, the resulting glibc is protected with indirect
15 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
16 compatible with all existing executables and shared libraries. This
17 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
18 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
19 capable of multi-byte NOPs, like x86-64 processors as well as Intel
20 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
21 x86_64 and x32 on non-CET processors. --enable-cet has been tested
22 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
25 * The GNU C Library now has correct support for ABSOLUTE symbols
26 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
27 relocated incorrectly or in some cases discarded. The GNU linker can
28 make use of the newer semantics, but it must communicate it to the
29 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
30 field) to indicate such support is required.
32 * Unicode 11.0.0 Support: Character encoding, character type info, and
33 transliteration tables are all updated to Unicode 11.0.0, using
34 generator scripts contributed by Mike FABIAN (Red Hat).
36 * <math.h> functions that round their results to a narrower type are added
37 from TS 18661-1:2014 and TS 18661-3:2015:
39 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
42 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
45 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
48 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
51 * Nominative and genitive month names are now supported for the following
52 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
53 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
54 support abbreviated alternative month names: Catalan, Greek, and Kashubian.
56 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
57 include the support of the nominative and genitive month names.
59 * Building and running on GNU/Hurd systems now works without out-of-tree
62 * The renameat2 function has been added, a variant of the renameat function
63 which has a flags argument. If the flags are zero, the renameat2 function
64 acts like renameat. If the flag is not zero and there is no kernel
65 support for renameat2, the function will fail with an errno value of
66 EINVAL. This is different from the existing gnulib function renameatu,
67 which performs a plain rename operation in case of a RENAME_NOREPLACE
68 flags and a non-existing destination (and therefore has a race condition
69 that can clobber the destination inadvertently).
71 * The statx function has been added, a variant of the fstatat64
72 function with an additional flags argument. If there is no direct
73 kernel support for statx, glibc provides basic stat support based on
74 the fstatat64 function.
76 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
77 library if installed. libidn2 version 2.0.5 or later is recommended. If
78 libidn2 is not available, internationalized domain names are not encoded
79 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
80 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
81 with an encoding error.) Flags which used to change the IDN encoding and
82 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
83 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
84 deprecated. They no longer have any effect.
86 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
87 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
88 range of ELF gABI expressions including such constructs as
89 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
90 have been further restricted, and where in the past a dynamic string
91 token sequence may have been interpreted as a literal string it will
92 now cause a load failure. These load failures were always considered
93 unspecified behaviour from the perspective of the dynamic loader, and
94 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
95 results in a load failure now.
97 Deprecated and removed features, and other changes affecting compatibility:
99 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
100 installed. Software that was using either header should be updated to
101 use standard <stdio.h> interfaces instead.
103 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
104 This was never required by the C standard, and the macros just expanded
105 to call alternative names for the same functions. If you hoped getc and
106 putc would provide performance improvements over fgetc and fputc, instead
107 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
108 necessary, flockfile and funlockfile.
110 * All stdio functions now treat end-of-file as a sticky condition. If you
111 read from a file until EOF, and then the file is enlarged by another
112 process, you must call clearerr or another function with the same effect
113 (e.g. fseek, rewind) before you can read the additional data. This
114 corrects a longstanding C99 conformance bug. It is most likely to affect
115 programs that use stdio to read interactive input from a terminal.
118 * The macros 'major', 'minor', and 'makedev' are now only available from
119 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
120 headers that happen to include <sys/types.h>. These macros are rarely
121 used, not part of POSIX nor XSI, and their names frequently collide with
122 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
125 <sys/sysmacros.h> is a GNU extension. Portable programs that require
126 these macros should first include <sys/types.h>, and then include
127 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
129 * The tilegx*-*-linux-gnu configurations are no longer supported.
131 * The obsolete function ustat is no longer available to newly linked
132 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
133 function has been deprecated in favor of fstatfs and statfs.
135 * The obsolete function nfsservctl is no longer available to newly linked
136 binaries. This function was specific to systems using the Linux kernel
137 and could not usefully be used with the GNU C Library on systems with
138 version 3.1 or later of the Linux kernel.
140 * The obsolete function name llseek is no longer available to newly linked
141 binaries. This function was specific to systems using the Linux kernel
142 and was not declared in a header. Programs should use the lseek64 name
143 for this function instead.
145 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
146 getaddrinfo and getnameinfo functions have been deprecated. The behavior
147 previously selected by them is now always enabled.
149 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
150 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
151 restriction (rejecting '_' in host names, among other things) has been
152 removed, for increased compatibility with non-IDN name resolution.
154 * The fcntl function now have a Long File Support variant named fcntl64. It
155 is added to fix some Linux Open File Description (OFD) locks usage on non
156 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
157 fcntl and LFS support is handled transparently. Also for Linux, the OFD
158 locks act as a cancellation entrypoint.
160 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
161 ecb_crypt, and des_setparity are no longer available to newly linked
162 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
163 longer installed. These functions encrypted and decrypted data with the
164 DES block cipher, which is no longer considered secure. Software that
165 still uses these functions should switch to a modern cryptography library,
168 * Reflecting the removal of the encrypt and setkey functions above, the
169 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
170 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
173 * The obsolete function fcrypt is no longer available to newly linked
174 binaries. It was just another name for the standard function crypt,
175 and it has not appeared in any header file in many years.
177 * We have tentative plans to hand off maintenance of the passphrase-hashing
178 library, libcrypt, to a separate development project that will, we hope,
179 keep up better with new passphrase-hashing algorithms. We will continue
180 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
181 'crypt_r' should not need to change at all; however, distributions will
182 need to install <crypt.h> and libcrypt from a separate project.
184 In this release, if the configure option --disable-crypt is used, glibc
185 will not install <crypt.h> or libcrypt, making room for the separate
186 project's versions of these files. The plan is to make this the default
187 behavior in a future release.
189 Changes to build and runtime requirements:
191 GNU make 4.0 or later is now required to build glibc.
193 Security related changes:
195 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
196 been fixed by removing the glibc-internal IDNA implementation and using
197 the system-provided libidn2 library instead. Originally reported by Hanno
198 Böck and Christian Weisgerber.
200 CVE-2017-18269: An SSE2-based memmove implementation for the i386
201 architecture could corrupt memory. Reported by Max Horn.
203 CVE-2018-11236: Very long pathname arguments to realpath function could
204 result in an integer overflow and buffer overflow. Reported by Alexey
207 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
208 architecture could write beyond the target buffer, resulting in a buffer
209 overflow. Reported by Andreas Schwab.
211 The following bugs are resolved with this release:
213 [The release manager will add the list generated by
214 scripts/list-fixed-bugs.py just before the release.]
221 * The GNU C Library can now be compiled with support for building static
222 PIE executables (See --enable-static-pie in INSTALL). These static PIE
223 executables are like static executables but can be loaded at any address
224 and provide additional security hardening benefits at the cost of some
225 memory and performance. When the library is built with --enable-static-pie
226 the resulting libc.a is usable with GCC 8 and above to create static PIE
227 executables using the GCC option '-static-pie'. This feature is currently
228 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
229 aarch64 with binutils 2.30 or later.
231 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
232 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
235 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
237 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
239 * In order to support faster and safer process termination the malloc API
240 family of functions will no longer print a failure address and stack
241 backtrace after detecting heap corruption. The goal is to minimize the
242 amount of work done after corruption is detected and to avoid potential
243 security issues in continued process execution. Reducing shutdown time
244 leads to lower overall process restart latency, so there is benefit both
245 from a security and performance perspective.
247 * The abort function terminates the process immediately, without flushing
248 stdio streams. Previous glibc versions used to flush streams, resulting
249 in deadlocks and further data corruption. This change also affects
250 process aborts as the result of assertion failures.
252 * On platforms where long double has the IEEE binary128 format (aarch64,
253 alpha, mips64, riscv, s390 and sparc), the math library now implements
254 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
255 These are the same interfaces added in version 2.26 for some platforms where
256 this format is supported but is not the format of long double.
258 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
259 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
260 implements interfaces for that type, as defined by ISO/IEC TS
261 18661-3:2015. These are corresponding interfaces to those supported for
264 * The math library now implements interfaces for the _Float32, _Float64 and
265 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
266 corresponding interfaces to those supported for _Float128.
268 * glibc now implements the memfd_create and mlock2 functions on Linux.
270 * Support for memory protection keys was added. The <sys/mman.h> header now
271 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
274 * The copy_file_range function was added.
276 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
278 * The ldconfig utility now processes `include' directives using the C/POSIX
279 collation ordering. Previous glibc versions used locale-specific
280 ordering, the change might break systems that relied on that.
282 * Support for two grammatical forms of month names has been added.
283 In a call to strftime, the "%B" and "%b" format specifiers will now
284 produce the grammatical form required when the month is used as part
285 of a complete date. New "%OB" and "%Ob" specifiers produce the form
286 required when the month is named by itself. For instance, in Greek
287 and in many Slavic and Baltic languages, "%B" will produce the month
288 in genitive case, and "%OB" will produce the month in nominative case.
290 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
291 are all valid and will all accept any known form of month
292 name---standalone or complete, abbreviated or full. In a call to
293 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
294 the strings used by "%B" and "%b", respectively. New query
295 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
296 used by "%OB" and "%Ob", respectively.
298 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
299 define the strings for %OB and %Ob, respectively; these have the
300 same syntax as "mon" and "abmon". These arrays are optional; if they
301 are not provided then they have the same content as "mon" and "abmon",
304 These features are provided for locales which define "alt_mon" and/or
305 "ab_alt_mon" in their locale source data. This release includes such
306 alternative month name data for the following languages: Belarusian,
307 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
309 This feature is currently a GNU extension, but it is expected to
310 be added to the next revision of POSIX, and it is also already
311 available on some BSD-derived operating systems.
313 This feature will cause existing statically compiled applications
314 to fail to load locales and fall back to the builtin C/POSIX locales.
315 See notes below for other changes affecting compatibility.
317 * Support for the RISC-V ISA running on Linux has been added. This port
318 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
319 for the following ISA and ABI pairs:
325 Deprecated and removed features, and other changes affecting compatibility:
327 * Statically compiled applications attempting to load locales compiled for the
328 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
329 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
330 support for two grammatical forms of the month names, also extends the locale
331 data binary format. Static applications needing locale support must be
332 recompiled to match the runtime and data they are deployed with. In some
333 distributions there is an upgrade window where dynamically linked applications
334 may use a new library but the old locale data and also fall back to the
335 builtin C/POSIX locales; restarting the application process is sufficient to
338 * Support for statically linked applications which call dlopen is deprecated
339 and will be removed in a future version of glibc. Applications which call
340 dlopen need to be linked dynamically instead.
342 * Support for old programs which use internal stdio data structures and
343 functions is deprecated. This includes programs which use the C++ streams
344 provided by libstdc++ in GCC 2.95. Programs which use the internal
345 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
346 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
347 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
348 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
349 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
350 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
351 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
352 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
353 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
354 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
355 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
356 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
357 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
358 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
359 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
360 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
361 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
362 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
363 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
364 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
365 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
366 future version of glibc. Unlike other symbol removals, these old
367 applications will not be supported using compatibility symbols.
369 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
370 defined by <sys/ptrace.h>.
372 * libm no longer supports SVID error handling (calling a user-provided
373 matherr function on error) or the _LIB_VERSION variable to control error
374 handling. (SVID error handling and the _LIB_VERSION variable still work
375 for binaries linked against older versions of the GNU C Library.) The
376 libieee.a library is no longer provided. math.h no longer defines struct
377 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
378 TLOSS, PLOSS and HUGE.
380 * The libm functions pow10, pow10f and pow10l are no longer supported for
381 new programs. Programs should use the standard names exp10, exp10f and
382 exp10l for these functions instead.
384 * The mcontext_t type is no longer the same as struct sigcontext. On
385 platforms where it was previously the same, this changes the C++ name
386 mangling for interfaces involving this type.
388 * The add-ons mechanism for building additional packages at the same time as
389 glibc has been removed. The --enable-add-ons configure option is now
392 * The --without-fp configure option is now ignored. Whether hardware
393 floating-point instructions are used is now configured based on whether
394 the compiler used at configure time (without any options implied by a
395 --with-cpu= configure option) uses such instructions.
397 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
398 the specified string can be parsed as a domain name.
400 * In the malloc_info output, the <heap> element may contain another <aspace>
401 element, "subheaps", which contains the number of sub-heaps.
403 * The libresolv function p_secstodate is no longer supported for new
406 * The tilepro-*-linux-gnu configuration is no longer supported.
408 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
409 and will be removed in a future release. Software that is still using
410 either header should be updated to use standard <stdio.h> interfaces
413 libio.h was originally the header for a set of supported GNU extensions,
414 but they have not been maintained as such in many years, they are now
415 standing in the way of improvements to stdio, and we don't think there are
416 any remaining external users. _G_config.h was never intended for public
417 use, but predates the bits convention.
419 Changes to build and runtime requirements:
421 * bison version 2.7 or later is required to generate code in the 'intl'
424 Security related changes:
426 CVE-2009-5064: The ldd script would sometimes run the program under
427 examination directly, without preventing code execution through the
428 dynamic linker. (The glibc project disputes that this is a security
429 vulnerability; only trusted binaries must be examined using the ldd
432 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
433 suffered from a one-byte overflow during ~ operator processing (either
434 on the stack or the heap, depending on the length of the user name).
435 Reported by Tim Rühsen.
437 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
438 would sometimes fail to free memory allocated during ~ operator
439 processing, leading to a memory leak and, potentially, to a denial
442 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
443 without GLOB_NOESCAPE, could write past the end of a buffer while
444 unescaping user names. Reported by Tim Rühsen.
446 CVE-2017-17426: The malloc function, when called with an object size near
447 the value SIZE_MAX, would return a pointer to a buffer which is too small,
448 instead of NULL. This was a regression introduced with the new malloc
449 thread cache in glibc 2.26. Reported by Iain Buclaw.
451 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
452 to the allocation of too much memory. (This is not a security bug per se,
453 it is mentioned here only because of the CVE assignment.) Reported by
456 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
457 of the number of search path components. (This is not a security
458 vulnerability per se because no trust boundary is crossed if the fix for
459 CVE-2017-1000366 has been applied, but it is mentioned here only because
460 of the CVE assignment.) Reported by Qualys.
462 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
463 for AT_SECURE or SUID binaries could be used to load libraries from the
466 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
467 succeeds without returning an absolute path due to unexpected behaviour
468 of the Linux kernel getcwd syscall. Reported by halfdog.
470 CVE-2018-6485: The posix_memalign and memalign functions, when called with
471 an object size near the value of SIZE_MAX, would return a pointer to a
472 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
474 CVE-2018-6551: The malloc function, when called with an object size near
475 the value of SIZE_MAX, would return a pointer to a buffer which is too
476 small, instead of NULL.
478 The following bugs are resolved with this release:
480 [866] glob: glob should match dangling symlinks
481 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
482 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
484 [5997] math: Very slow execution of sinf function
485 [10580] localedata: hr_HR: updated locale
486 [10871] locale: 'mon' array should contain both nominative and genitive
488 [12349] localedata: eu_ES: incorrect thousands separator
489 [13605] localedata: shn_MM: new Shan locale
490 [13805] localedata: ru_RU: currency should use ',' as radix point
491 [13953] localedata: km_KH: locale update
492 [13994] localedata: mjw_IN: new locale
493 [14121] build: make writes .mo files in po directory
494 [14333] libc: Fix the race between atexit() and exit()
495 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
496 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
498 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
499 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
500 width Latin characters
501 [15332] localedata: es_CU: locale update
502 [15436] stdio: Don't close or flush stdio streams on abort
503 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
505 [16148] localedata: ca_ES: incorrect thousands separator
506 [16750] dynamic-link: ldd should not try to execute the binaries
508 [16777] localedata: pl_PL: incorrect thousands separator in locale
509 [16905] localedata: hanzi: new collation
510 [17563] localedata: cmn_TW: add hanzi collation
511 [17750] localedata: wrong collation order of diacritics in most locales
512 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
514 [17956] build: Build fails on missing definitions from header file
515 nss/nss.h when Mozilla NSS is used for cryptography
516 [18203] libc: realpath() does not handle unreachable paths correctly
517 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
518 [18812] localedata: kab_DZ: new Kabyle Algeria locale
519 [18822] libc: Internal functions are called via PLT
520 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
521 [19170] libc: __gmon_start__ defined in hppa in crtn.S
522 [19574] libc: glibc should support building static PIE binaries
523 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
525 [19971] glob: glob: Do not skip entries with zero d_ino values
526 [19982] localedata: fr.po: spelling mistake for error code EXDEV
527 [20008] localedata: km_KH: convert to translit_neutral
528 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
529 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
530 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
531 [20482] localedata: de_CH: abbreviated weekdays should be two letters
532 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
534 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
535 they're set, wrong errors returned by gaih_inet when lookup functions
537 [20756] localedata: [PATCH] Use Unicode wise thousands separator
538 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
540 [20952] localedata: yuw_PG: new locale
541 [21084] localedata: charmaps/IBM858: new codepage
542 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
543 [21242] libc: assert gives pedantic warning in old gcc versions
544 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
545 __regcall calling convention
546 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
547 [21326] libc: C99 functions are not declared for C++11 and later if
548 _GNU_SOURCE is not predefined by g++
549 [21457] libc: sys/ucontext.h namespace
550 [21530] libc: tmpfile() should be implemented using O_TMPFILE
551 [21660] math: GCC fails to compile a formula with tgmath.h
552 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
553 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
554 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
555 [21684] math: tgmath.h handling of complex integers
556 [21685] math: tgmath.h handling of bit-fields
557 [21686] math: tgmath.h handling of __int128
558 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
560 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
562 [21750] localedata: column width of characters incompatible with classical
564 [21754] malloc: malloc: Perform as little work as possible after heap
565 consistency check failures
566 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
567 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
568 [21791] string: Unused XXX_chk_XXX functions in libc.a
569 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
571 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
573 [21845] localedata: Added new Locale bho_NP
574 [21853] localedata: Fix abday Which looks same as day in zh_SG
575 [21854] localedata: Added New Locale en_SC
576 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
577 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
578 _dl_runtime_resolve_avx_slow
579 [21885] network: getaddrinfo: gethosts does not release resolver context
580 on memory allocation failure
581 [21899] libc: XPG4.2 sigaction namespace
582 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
584 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
586 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
587 errno value, causing getaddrinfo to fail
588 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
589 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
590 instead of EAI_NODATA
591 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
592 constant PTRACE_SEIZE_DEVEL
593 [21930] math: C-only gcc builtins used in <math.h> isinf
594 [21932] network: Unpaired __resolv_context_get in generic get*_r
596 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
597 [21944] libc: sigval namespace
598 [21951] localedata: Update hanzi collation by stroke
599 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
600 sysdeps/x86_64/fpu/e_expf.S
601 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
602 allocation in syscall loops)
603 [21959] localedata: Fix Country name for xh_ZA
604 [21960] localedata: Fix abmon for bem_ZM
605 [21966] math: AVX2 mathvec functions use FMA without checking
606 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
607 [21971] localedata: Added New Locale for mfe_MU
608 [21972] libc: assert macro requires operator== (int) for its argument type
609 [21973] math: [sparc] libm missing sqrtl compat symbol
610 [21974] libc: Remove __bb_init_func and __bb_exit_func
611 [21982] string: stratcliff.c: error: assuming signed overflow does not
613 [21986] stdio: __guess_grouping is called incorrectly
614 [21987] math: [sparc32] wrong bits/long-double.h installed
615 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
617 [22022] localedata: Missing country_name for mni_IN
618 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
620 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
622 [22026] locale: iconv_open: heap overflow on gconv_init failure
623 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
624 [22035] math: [m68k] bits/math-inline.h macro namespace
625 [22038] localedata: Fix abbreviated weeks and months for Somali
626 [22044] localedata: Remove redundant data for Limburgish Language
627 [22050] malloc: Linking with -lmcheck does not hook
628 __malloc_initialize_hook correctly
629 [22051] libc: zero terminator in the middle of glibc's .eh_frame
630 [22052] malloc: malloc failed to compile with GCC 7 and -O3
631 [22070] localedata: charmaps/UTF-8: wcwidth for
632 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
633 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
634 Jungseong and Jongseong) should be 0
635 [22078] nss: nss_files performance issue in multi mode
636 [22082] math: bits/math-finite.h exp10 condition
637 [22086] libc: pcprofiledump incorrect cross-endian condition
638 [22093] dynamic-link: ld.so no longer searches in .../x86_64
639 [22095] network: Name server address allocation memory leak in resolv.conf
641 [22096] network: __resolv_conf_attach can incorrectly free passed conf
643 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
644 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
646 [22111] malloc: malloc: per thread cache is not returned when thread exits
647 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
648 [22134] libc: [linux] implement fexecve with execveat
649 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
651 [22145] libc: ttyname() gives up too early in the face of namespaces
652 [22146] math: C++ build issue with float128 on x86_64
653 [22153] nptl: nptl: save error code before process termination
654 [22156] libc: [hppa,ia64,microblaze] Executable stack default
655 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
656 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
657 [22165] libc: [hppa] Text relocations in libc.so
658 [22180] libc: destructor registered via __cxa_atexit is called twice
659 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
660 symlinks") cause "make" segfaults
661 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
663 [22207] libc: FAIL: stdlib/test-atexit-race
664 [22225] math: nearbyint arithmetic moved before feholdexcept
665 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
666 [22235] math: iscanonical in C++ and float128
667 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
668 the svid compat wrapper
669 [22244] math: ynf and yn are wrong without the svid compat wrapper
670 [22273] libc: Improper assert in Linux posix_spawn implementation
671 [22284] libc: -pg -pie doesn't work
672 [22292] locale: localedef exits with error 4 when it should be error 1
673 [22294] locale: Allow "" for int_currency_symbol definition in locales.
674 [22295] locale: Don't warn on non-symbolic characters in locale sources in
676 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
677 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
679 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
680 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
681 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
682 [22322] libc: [mips64] wrong bits/long-double.h installed
683 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
684 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
686 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
687 [22343] malloc: Integer overflow in posix_memalign
688 [22347] libc: getrandom() returns the number of bytes that were copied to
689 the buffer even though the comments say "Return 0 on success and -1 on
691 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
692 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
694 [22370] dynamic-link: Incorrect note padding check
695 [22375] libc: malloc returns pointer from tcache_get when should return
696 NULL (CVE-2017-17426)
697 [22377] math: iseqsig, float128 and C++
698 [22382] localedata: Error in tpi_PG locale
699 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
700 inside the ASCII printable range
701 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
703 [22403] localedata: Slash needs escaping in some locales
704 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
705 [22409] network: res_hnok does not accept some host names used on the
707 [22412] network: res_dnok, res_hnok should perform syntax checks
708 [22413] network: ns_name_pton ignores syntactically invalid trailing
710 [22415] stdio: setvbuf can lead to invalid free/segfault
711 [22432] build: Non-deterministic build
712 [22439] malloc: malloc_info should compute summary statistics for all sub-
714 [22442] network: if_nametoindex could report index for the wrong
716 [22446] build: aliasing violation calling readlink in handle_request
717 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
718 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
719 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
720 __nscd_hash/__nss_hash
721 [22463] network: p_secstodate overflow handling
722 [22469] localedata: pl_PL LC_COLLATE does not use i18n
723 [22478] libc: sigwait can fail with EINTR
724 [22505] libc: ldconfig processes include directive in locale-specific
726 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
727 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
728 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
729 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
730 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
731 [22534] localedata: Collation rules for Serbian and Bosnian should be the
733 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
734 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
735 [22577] libc: missing newline after "cannot allocate TLS data structures
737 [22588] manual: manual/conf.texi: missing underscore in front of
739 [22593] math: nextafter and nexttoward are declared with const attribute
740 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
741 [22603] string: ia64 memchr overflows internal pointer check
742 [22605] libc: SH clone does not set the exit code correctly
743 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
745 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
746 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
747 modified in case of success
748 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
749 [22615] manual: manual: ambiguous wording about errno value in case of
751 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
753 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
754 binaries or if /proc is not mounted (CVE-2017-16997)
755 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
756 [22630] build: $(no-pie-ldflag) is no longer effective
757 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
758 [22635] nptl: pthread_self returns NULL before libpthread is loaded
759 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
760 [22637] nptl: guard size is subtracted from thread stack size instead of
762 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
763 [22657] localedata: hu_HU: Avoid double space in date
764 [22660] math: fmax, fmin sNaN handling on alpha
765 [22664] libc: New warning of GCC8
766 [22665] math: alpha: ceil and floor raise inexact exceptions
767 [22666] math: alpha: trunc raise inexact exceptions
768 [22667] libc: makecontext lacks stack alignment on i386
769 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
770 [22679] libc: getcwd(3) can succeed without returning an absolute path
772 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
774 [22687] math: [powerpc-nofpu] complex long double functions spurious
776 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
777 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
778 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
780 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
781 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
783 [22701] nis: Incomplete removal of libnsl
784 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
785 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
786 [22715] dynamic-link: FAIL: elf/tst-audit10
787 [22719] libc: Backtrace tests fail on hppa
788 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
789 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
790 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
791 before the first call to crypt_r () in crypt/badsalttest.c
798 * A per-thread cache has been added to malloc. Access to the cache requires
799 no locks and therefore significantly accelerates the fast path to allocate
800 and free small amounts of memory. Refilling an empty cache requires locking
801 the underlying arena. Performance measurements show significant gains in a
802 wide variety of user workloads. Workloads were captured using a special
803 instrumented malloc and analyzed with a malloc simulator. Contributed by
804 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
806 * Unicode 10.0.0 Support: Character encoding, character type info, and
807 transliteration tables are all updated to Unicode 10.0.0, using
808 generator scripts contributed by Mike FABIAN (Red Hat).
809 These updates cause user visible changes, especially the changes in
810 wcwidth for many emoji characters cause problems when emoji sequences
811 are rendered with pango, see for example:
812 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
814 * Collation of Hungarian has been overhauled and is now consistent with "The
815 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
818 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
820 - The GNU C Library will now detect when /etc/resolv.conf has been
821 modified and reload the changed configuration. The new resolver option
822 “no-reload” (RES_NORELOAD) disables this behavior.
824 - The GNU C Library now supports an arbitrary number of search domains
825 (configured using the “search” directive in /etc/resolv.conf);
826 previously, there was a hard limit of six domains. For backward
827 compatibility, applications that directly modify the ‘_res’ global
828 object are still limited to six search domains.
830 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
831 Library will now randomly pick a name server from the configuration as a
832 starting point. (Previously, the second name server was always used.)
834 * The tunables feature is now enabled by default. This allows users to tweak
835 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
837 * New function reallocarray, which resizes an allocated block (like realloc)
838 to the product of two sizes, with a guaranteed clean failure upon integer
839 overflow in the multiplication. Originally from OpenBSD, contributed by
840 Dennis Wölfing and Rüdiger Sonderfeld.
842 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
843 These are extended versions of preadv and pwritev, respectively, taking an
844 additional flags argument. The set of supported flags depends on the
845 running kernel; full support currently requires kernel 4.7 or later.
847 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
848 create a new session ID for the spawned process. This feature is
849 scheduled to be added to the next major revision of POSIX; for the time
850 being, it is available under _GNU_SOURCE.
852 * errno.h is now safe to use from C-preprocessed assembly language on all
853 supported operating systems. In this context, it will only define the
854 Exxxx constants, as preprocessor macros expanding to integer literals.
856 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
857 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
858 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
859 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
861 To compile programs that use this feature, the compiler must support
862 128-bit floating point with the type name _Float128 (as defined by TS
863 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
864 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
865 must be defined to make the new interfaces visible.
867 The new functions and macros correspond to those present for other
868 floating-point types (except for a few obsolescent interfaces not
869 supported for the new type), with F128 or f128 suffixes; for example,
870 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
871 printf or scanf formats for the new type; the strfromf128 and strtof128
872 interfaces should be used instead.
874 Deprecated and removed features, and other changes affecting compatibility:
876 * The synchronization that pthread_spin_unlock performs has been changed to
877 now be equivalent to a C11 atomic store with release memory order to the
878 spin lock's memory location. Previously, several (but not all)
879 architectures used stronger synchronization (e.g., containing what is
880 often called a full barrier). This change can improve performance, but
881 may affect odd fringe uses of spin locks that depend on the previous
882 behavior (e.g., using spin locks as atomic variables to try to implement
883 Dekker's mutual exclusion algorithm).
885 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
888 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
889 will only be built and installed when the GNU C Library is configured with
890 --enable-obsolete-rpc. This allows alternative RPC implementations, such
891 as TIRPC or rpcsvc-proto, to be used.
893 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
894 libnss_compat, are deprecated, and will not be built or installed by
897 The NIS(+) support library, libnsl, is also deprecated. By default, a
898 compatibility shared library will be built and installed, but not headers
899 or development libraries. Only a few NIS-related programs require this
900 library. (In particular, the GNU C Library has never required programs
901 that use 'gethostbyname' to be linked with libnsl.)
903 Replacement implementations based on TIRPC, which additionally support
904 IPv6, are available from <https://github.com/thkukuk/>. The configure
905 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
906 name service modules, to be built and installed.
908 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
909 support is enabled, the configured recursive resolver must support EDNS.
910 (Responding to EDNS-enabled queries with responses which are not
911 EDNS-enabled is fine, but FORMERR responses are not.)
913 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
914 servers have not supported this opcode for a long time.
916 * The _res_opcodes variable has been removed from libresolv. It had been
917 exported by accident.
919 * <string.h> no longer includes inline versions of any string functions,
920 as this kind of optimization is better done by the compiler. The macros
921 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
923 * The nonstandard header <xlocale.h> has been removed. Most programs should
924 use <locale.h> instead. If you have a specific need for the definition of
925 locale_t with no other declarations, please contact
926 libc-alpha@sourceware.org and explain.
928 * The obsolete header <sys/ultrasound.h> has been removed.
930 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
932 * The obsolete function cfree has been removed. Applications should use
935 * The stack_t type no longer has the name struct sigaltstack. This changes
936 the C++ name mangling for interfaces involving this type.
938 * The ucontext_t type no longer has the name struct ucontext. This changes
939 the C++ name mangling for interfaces involving this type.
941 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
942 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
943 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
944 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
945 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
946 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
947 fpregset_t type no longer has the name struct fpu. This changes the C++
948 name mangling for interfaces involving those types.
950 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
951 synced with the kernel:
953 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
954 are not supported on this architecture and have been removed.
956 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
957 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
958 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
960 Programs that assume the GET/SETREGS ptrace requests are universally
961 available will now fail to build, instead of malfunctioning at runtime.
963 Changes to build and runtime requirements:
965 * Linux kernel 3.2 or later is required at runtime, on all architectures
966 supported by that kernel. (This is a change from version 2.25 only for
969 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
971 * On most architectures, GCC 4.9 or later is required to build the GNU C
972 Library. On powerpc64le, GCC 6.2 or later is required.
974 Older GCC versions and non-GNU compilers are still supported when
975 compiling programs that use the GNU C Library. (We do not know exactly
976 how old, and some GNU extensions to C may be _de facto_ required. If you
977 are interested in helping us make this statement less vague, please
978 contact libc-alpha@sourceware.org.)
980 Security related changes:
982 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
983 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
985 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
986 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
988 * Avoid printing a backtrace from the __stack_chk_fail function since it is
989 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
992 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
993 fixed (CVE-2017-12133).
995 The following bugs are resolved with this release:
997 [984] network: Respond to changed resolv.conf in gethostbyname
998 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
999 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1001 [12189] libc: __stack_chk_fail should not attempt a backtrace
1003 [14096] time: Race condition on timezone/tst-timezone.out
1004 [14172] localedata: az_IR: new locale
1005 [14995] build: glibc fails to build if gold is the default linker, even if
1007 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1008 [16637] network: inet_pton function is accepting IPv6 with bad format
1009 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1010 x86 and other generic code
1011 [16875] localedata: ko_KR: fix lang_name
1012 [17225] localedata: ar_SY: localized month names for May and June are
1014 [17297] localedata: da_DK: wrong date_fmt string
1015 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1016 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1017 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1018 [19066] localedata: ar_SA abbreviated day and month names are in English
1019 [19569] network: resolv: Support an arbitrary number of search domains
1020 [19570] network: Implement random DNS server selection in the stub
1022 [19838] locale: localedef fails on PA-RISC
1023 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1024 order of 0D36 and 0D37
1025 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1027 [20098] libc: FAIL: debug/backtrace-tst on hppa
1028 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1030 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1032 [20313] localedata: Update locale data to Unicode 9.0
1033 [20424] manual: Document how to provide a malloc replacement
1034 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1036 [20686] locale: Add el_GR@euro to SUPPORTED.
1037 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1038 failures consistently
1039 [21015] dynamic-link: Document and fix --enable-bind-now
1040 [21016] nptl: pthread_cond support is broken on hppa
1041 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1042 frame-pointer on i386
1043 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1045 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1046 [21088] libc: Build fails with --enable-static-nss
1047 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1048 generic c code is used
1049 [21109] libc: Tunables broken on big-endian
1050 [21112] math: powf has large ulp errors with base close to 1 and exponent
1052 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1054 [21120] malloc: glibc malloc is incompatible with GCC 7
1055 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1057 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1058 (-0.0) when linking with -lieee
1059 [21171] math: log10, log2 and lgamma return incorrect results
1060 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1061 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1062 [21207] localedata: ce_RU: update weekdays from CLDR
1063 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1064 [21217] localedata: Update months from CLDR-31
1065 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1066 [21243] libc: support_delete_temp_file should issue warning for failed
1068 [21244] libc: support resolv_test_start() socket fd close should be
1070 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1071 new posix_spawn implementation
1072 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1073 leads to lower CPU frequency
1074 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1075 [21261] libc: [sparc64] bits/setjmp.h namespace
1076 [21267] network: [mips] bits/socket.h IOC* namespace
1077 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1078 [21270] libc: mmap64 silently truncates large offset values
1079 [21275] libc: posix_spawn always crashes on ia64 now
1080 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1081 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1082 [21289] libc: Incorrect declaration for 32-bit platforms with
1083 _FILE_OFFSET_BITS=64 causes build error
1084 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1085 not support gethostbyname4_r
1086 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1088 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1090 [21340] libc: Support POSIX_SPAWN_SETSID
1091 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1092 [21359] network: ns_name_pack needs additional byte in destination buffer
1093 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1094 against fragmentation attacks (CVE-2017-12132)
1095 [21369] network: resolv: Remove EDNS fallback
1096 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1098 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1099 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1100 [21393] stdio: Missing dup3 error check in freopen, freopen64
1101 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1102 [21399] localedata: Bad description for U00EC in
1103 localedata/charmaps/CP1254
1104 [21411] malloc: realloc documentation error
1105 [21426] network: sys/socket.h uio.h namespace
1106 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1107 [21445] libc: signal.h bsd_signal namespace
1108 [21455] network: Network headers stdint.h namespace
1109 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1111 [21475] network: resolv: Overlong search path is truncated mid-label
1112 [21511] libc: sigstack namespace
1113 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1114 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1115 bits/syscall.h: No such file or directory
1116 [21517] libc: struct sigaltstack namespace
1117 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1118 [21533] localedata: Update locale data to Unicode 10.0
1120 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1121 junk at end of line, first unrecognized character is `@'
1122 [21538] libc: SIG_HOLD missing for XPG4
1123 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1124 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1125 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1126 [21543] libc: sigevent namespace
1127 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1128 size for general purpose registers in mcontext_t structure
1129 [21550] libc: sigwait namespace
1130 [21552] libc: XPG4 bsd_signal namespace
1131 [21554] libc: sigpause namespace
1132 [21560] libc: sys/wait.h signal.h namespace
1133 [21561] libc: waitid namespace
1134 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1135 '/usr/include/stdlib.h': Permission denied
1136 [21575] libc: sys/wait.h missing struct rusage definition
1137 [21584] libc: sigaltstack etc namespace
1138 [21597] libc: siginterrupt namespace
1139 [21607] math: hppa: FAIL: math/test-tgmath
1140 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1142 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1143 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1144 stack and heap (CVE-2017-1000366)
1145 [21625] libc: wait3 namespace
1146 [21654] nss: Incorrect pointer alignment in NSS group merge result
1148 [21657] network: Parse interface zone id for node-local multicast
1149 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1151 [21666] libc: .symver is used on common symbol
1152 [21668] network: resolv: res_init cross-thread broadcast introduces race
1154 [21687] math: tgmath.h totalorder, totalordermag return type
1155 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1157 [21696] libc: Incorrect assumption of of __cpu_mask in
1158 posix/sched_cpucount.c
1159 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1160 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1162 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1163 leading to relocation crash
1164 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1165 `resolv_conf_matches (resp, conf)' failed.
1166 [21710] localedata: Added Samoan language locale for Samoa
1167 [21711] localedata: Pashto yesstr/nostr locale are missing
1168 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1169 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1170 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1171 [21723] localedata: yesstr/nostr missing for Chinese language locale
1172 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1174 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1176 [21728] localedata: New Locale for Tongan language
1177 [21729] localedata: incorrect LC_NAME fields for hi_IN
1178 [21733] localedata: yesstr and nostr are missing for zh_HK
1179 [21734] localedata: Missing yesstr and nostr are for kw_GB
1180 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1181 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1182 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1183 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1185 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1187 [21749] localedata: Wrong abbreviated day name (“abday”) for
1189 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1190 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1191 [21759] localedata: missing yesstr and nostr for Tigrinya
1192 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1193 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1194 [21767] localedata: Missing Bislama locales
1195 [21768] localedata: Missing yesstr and nostr for aa_ET
1196 [21770] localedata: Missing Field in li_NL
1197 [21778] nptl: Robust mutex may deadlock
1198 [21779] libc: MicroBlaze segfaults when loading libpthread
1199 [21783] localedata: Fix int_select international_call_prefixes
1200 [21784] localedata: Inconsistency in country_isbn
1201 [21788] localedata: Missing Country Postal Abbreviations
1202 [21794] localedata: Added-country_isbn-for-Italy
1203 [21795] localedata: Add/Fix country_isbn for France
1204 [21796] localedata: Added country_isbn for Republic of Korea
1205 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1206 [21799] localedata: Added int_select international_call_prefixes
1207 [21801] localedata: Added int_select international_call_prefixes
1208 [21804] nptl: Double semicolon in thread-shared-types.h
1209 [21807] localedata: LC_ADDRESS fix for pap_CW
1210 [21808] localedata: Fix LC_ADDRESS for pap_AW
1211 [21821] localedata: Added country_name in mai_IN
1212 [21822] localedata: Fix LC_TIME for mai_IN
1213 [21823] localedata: missing yesstr, nostr for sa_IN
1214 [21825] localedata: Fix name_mrs for mag_IN
1215 [21828] localedata: 2.26 changelog should mention user visible changes
1217 [21835] localedata: Added Maithili language locale for Nepal
1218 [21838] localedata: Removed redundant data for the_NP
1219 [21839] localedata: Fix LC_MONETARY for ta_LK
1220 [21844] localedata: Fix Latin characters and Months Sequence.
1221 [21848] localedata: Fix mai_NP Title Name
1226 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1227 24731-2:2010, is supported to enable declarations of functions from that
1228 TR. Note that not all functions from that TR are supported by the GNU C
1231 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1232 18661-1:2014, is supported to enable declarations of functions and macros
1233 from that TS. Note that not all features from that TS are supported by
1236 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1237 18661-4:2015, is supported to enable declarations of functions and macros
1238 from that TS. Note that most features from that TS are not supported by
1241 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1242 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1243 Since the GNU C Library defaults to a much newer revision of POSIX, this
1244 will only affect programs that specifically request an old conformance
1245 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1246 see a change in the visible declarations, but a program compiled with
1247 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1250 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1251 defined by all multithreaded code, but glibc has not required this for
1254 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1255 means that in a future release, the macros “major”, “minor”, and “makedev”
1256 will only be available from <sys/sysmacros.h>.
1258 These macros are not part of POSIX nor XSI, and their names frequently
1259 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1260 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1261 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1264 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1265 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1266 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1268 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1269 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1270 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1271 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1272 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1273 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1274 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1275 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1276 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1277 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1278 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1281 * New <math.h> features are added from TS 18661-1:2014:
1283 - Signaling NaN macros: SNANF, SNAN, SNANL.
1285 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1286 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1287 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1289 - llogb functions: the llogb, llogbf and llogbl functions, and the
1290 FP_LLOGB0 and FP_LLOGBNAN macros.
1292 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1295 - Comparison macros: iseqsig.
1297 - Classification macros: iscanonical, issubnormal, iszero.
1299 - Total order functions: totalorder, totalorderf, totalorderl,
1300 totalordermag, totalordermagf, totalordermagl.
1302 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1304 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1305 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1307 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1308 are added to libc. They convert a floating-point number into string.
1310 * Most of glibc can now be built with the stack smashing protector enabled.
1311 It is recommended to build glibc with --enable-stack-protector=strong.
1312 Implemented by Nick Alcock (Oracle).
1314 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1315 intended to be used instead of memset() to erase sensitive data after use;
1316 the compiler will not optimize out calls to explicit_bzero even if they
1317 are "unnecessary" (in the sense that no _correct_ program can observe the
1318 effects of the memory clear).
1320 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1321 to float instead of double. This does not affect the ABI of any libraries
1322 that are part of the GNU C Library, but may affect the ABI of other
1323 libraries that use this type in their interfaces.
1325 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1326 float_t and double_t types are now defined to long double instead of float
1327 and double. These options are not the default, and this does not affect
1328 the ABI of any libraries that are part of the GNU C Library, but it may
1329 affect the ABI of other libraries that use this type in their interfaces,
1330 if they are compiled or used with those options.
1332 * The getentropy and getrandom functions, and the <sys/random.h> header file
1335 * The buffer size for byte-oriented stdio streams is now limited to 8192
1336 bytes by default. Previously, on Linux, the default buffer size on most
1337 file systems was 4096 bytes (and thus remains unchanged), except on
1338 network file systems, where the buffer size was unpredictable and could be
1339 as large as several megabytes.
1341 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1342 for the Linux quota interface which predates kernel version 2.4.22 has
1345 * The malloc_get_state and malloc_set_state functions have been removed.
1346 Already-existing binaries that dynamically link to these functions will
1347 get a hidden implementation in which malloc_get_state is a stub. As far
1348 as we know, these functions are used only by GNU Emacs and this change
1349 will not adversely affect already-built Emacs executables. Any undumped
1350 Emacs executables, which normally exist only during an Emacs build, should
1351 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1353 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1354 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1355 “no-ip6-dotint” had already been the default, and support for the
1356 “ip6-dotint” option was removed from the Internet in 2006.
1358 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1359 flag from <resolv.h> have been removed. The option relied on a
1360 backwards-incompatible DNS extension which was never deployed on the
1363 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1364 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1365 They were already unimplemented.
1367 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1368 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1369 removed again from the IETF name lookup interface specification in RFC
1370 2553. Applications should use getaddrinfo instead.
1372 * DNSSEC-related declarations and definitions have been removed from the
1373 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1374 decode the data part of DNSSEC record types. Previous versions of glibc
1375 only implemented minimal support for the previous version of DNSSEC, which
1376 is incompatible with the currently deployed version.
1378 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1379 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1380 <arpa/nameser.h> header file because the distinction between RR types and
1381 meta-RR types is not officially standardized, subject to revision, and
1382 thus not suitable for encoding in a macro.
1384 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1385 and rhook members of the res_state type in <resolv.h> have been removed.
1386 The glibc stub resolver did not support these hooks, but the header file
1387 did not reflect that.
1389 * For multi-arch support it is recommended to use a GCC which has
1390 been built with support for GNU indirect functions. This ensures
1391 that correct debugging information is generated for functions
1392 selected by IFUNC resolvers. This support can either be enabled by
1393 configuring GCC with '--enable-gnu-indirect-function', or by
1394 enabling it by default by setting 'default_gnu_indirect_function'
1395 variable for a particular architecture in the GCC source file
1398 * GDB pretty printers have been added for mutex and condition variable
1399 structures in POSIX Threads. When installed and loaded in gdb these pretty
1400 printers show various pthread variables in human-readable form when read
1401 using the 'print' or 'display' commands in gdb.
1403 * Tunables feature added to allow tweaking of the runtime for an application
1404 program. This feature can be enabled with the '--enable-tunables' configure
1405 flag. The GNU C Library manual has details on usage and README.tunables has
1406 instructions on adding new tunables to the library.
1408 * A new version of condition variables functions have been implemented in
1409 the NPTL implementation of POSIX Threads to provide stronger ordering
1412 * A new version of pthread_rwlock functions have been implemented to use a more
1413 scalable algorithm primarily through not using a critical section anymore to
1416 Security related changes:
1418 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1419 have been created with makecontext could fail to terminate due to a
1420 missing .cantunwind annotation. This has been observed to lead to a hang
1421 (denial of service) in some Go applications compiled with gccgo. Reported
1422 by Andreas Schwab. (CVE-2016-6323)
1424 * The DNS stub resolver functions would crash due to a NULL pointer
1425 dereference when processing a query with a valid DNS question type which
1426 was used internally in the implementation. The stub resolver now uses a
1427 question type which is outside the range of valid question type values.
1430 The following bugs are resolved with this release:
1432 [4099] stdio: Overly agressive caching by stream i/o functions.
1433 [7065] build: Support building glibc with -fstack-protector or -fstack-
1435 [9842] localedata: en_CA: incorrect date format
1436 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1437 before it started waiting
1438 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1439 [15765] nptl: sem_open is wrongly a cancellation point
1440 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1441 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1443 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1444 library linked with pthread
1445 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1446 "FIXME: Ingo" issue)
1447 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1448 [17252] libc: getrandom and getentropy syscall
1449 [17730] malloc: thread-local storage is sometimes improperly free()'d
1450 after being __libc_memalign()'d
1451 [18241] stdio: failed fseek on memstream does not set errno and can fail
1453 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1455 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1456 PTHREAD_PRIO_INHERIT mutex on ARM
1457 [18784] network: res_query and related function crash for special record
1458 type queries (CVE-2015-5180)
1459 [19380] math: strtod does not raise "inexact"
1460 [19387] string: Integer overflow in memchr
1461 [19390] string: Integer overflow in strncat
1462 [19398] build: linknamespace tests fail with massively parallel build
1463 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1465 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1466 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1467 [19514] libc: [PATCH] Fix spelling errors in spelling
1468 "implement"/"implementation" in several places
1469 [19582] network: Deprecate RES_USE_INET6
1470 [19673] manual: clog10 docs appear to be erroneous
1471 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1472 causes a segmentation fault
1473 [19826] libc: invalid pointer returned from __tls_get_addr with static
1475 [20016] network: resolv: Remove hooks support from the API
1476 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1478 [20033] math: [x86_64] vectorized math function don't call the __finite
1480 [20116] nptl: use after free in pthread_create
1481 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1482 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1484 [20311] nptl: please install proc_service.h
1485 [20366] libc: Compilation errors in installed headers in strict-compliance
1487 [20370] malloc: malloc: Arena free list management is still racy
1488 (incorrect fix in bug 19243)
1489 [20386] libc: assert (X = 0) does not result in GCC warning
1490 [20432] malloc: malloc: Minimize interface required for interposition
1491 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1492 in _Unwind_Backtrace (CVE-2016-6323)
1493 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1494 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1496 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1497 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1498 [20477] network: resolv: incorrect double-checked locking related to
1500 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1502 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1504 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1505 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1506 cause transition penalty
1507 [20517] math: sparcv9 missing fdiml compat symbol
1508 [20524] manual: strverscmp is inconsistent
1509 [20525] libc: <sys/quota.h> should be based on kernel headers
1510 [20539] math: GCC 7 -static -lm fails to link at -O3
1511 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1513 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1515 [20591] network: Remove obsolete DNSSEC support
1516 [20592] network: DNS resource record type classification macros in
1517 <arpa/nameser.h> are incorrect
1518 [20593] network: Update DNS RR type definitions
1519 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1520 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1522 [20629] network: libresolv: Remove support for bitstring labels
1524 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1525 assertion: "maxidx >= 2"
1526 [20660] math: [arm] Use VSQRT
1527 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1528 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1529 [20689] libc: Test for FMA should also check for AVX.
1530 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1531 [20715] math: iszero macro breaks existing code
1532 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1533 [20728] libc: powerpc: Missing TOC stub in clone
1534 [20729] build: glibc-2.24 fails to build for i486 with -Os
1535 [20750] build: Build fails with default PIE enabled gcc-6.x
1536 [20768] math: [alpha] sqrt fegetenv namespace
1537 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1539 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1540 __FLT_EVAL_METHOD__ is 2
1541 [20790] network: rpcgen buffer overrun in get_prog_declaration
1542 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1543 [20829] libc: crypt snprintf namespace
1544 [20847] libc: tst-vfork3 failure
1545 [20855] math: Default bits/mathdef.h has inappropriate float_t
1546 [20859] math: [sh4] FP_ILOGB0 invalid
1547 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1548 U+20AC), not same as GBK
1549 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1550 [20916] math: pow handling of sNaN arguments
1551 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1552 [20919] math: Bad pow (qNaN, 0) result with -lieee
1553 [20940] math: hypot sNaN handling
1554 [20947] math: fmax, fmin sNaN handling
1555 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1556 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1558 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1559 [20973] nptl: robust mutexes: Lost wake-ups
1560 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1561 [20978] nis: strlen on null pointer in nss_nisplus
1562 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1564 [21014] string: i686 memchr overflows internal pointer check
1565 [21019] libc: [mips] n32 lseek incorrect on overflow
1566 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1567 [21026] libc: [MIPS] readahead syscall is broken on n64
1568 [21028] math: Fallback fesetexceptflag should always succeed
1569 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1570 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1572 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1573 [21061] librt: [microblaze] librt lost clock_* exports
1574 [21073] libc: tunables: insecure environment variables passed to
1575 subprocesses with AT_SECURE
1576 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1580 * The minimum Linux kernel version that this version of the GNU C Library
1581 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1582 kernel version 2.6.32 or later suffices (on architectures that already
1583 required kernel versions more recent than 3.2, those requirements remain
1584 unchanged). Linux 3.2 or later kernel headers are required on all
1587 * The pap_AN locale has been deleted. This has been deprecated for a long
1588 time. It has been replaced by pap_AW & pap_CW, both of which have long
1589 been included in previous releases.
1591 * The readdir_r and readdir64_r functions have been deprecated. It is
1592 recommended to use readdir and readdir64 instead.
1594 * The type “union wait” has been removed. It was deprecated in the early
1595 1990s and never part of POSIX. Application code should use the int type
1596 instead of “union wait”.
1598 * A new NSS action is added to facilitate large distributed system
1599 administration. The action, MERGE, allows remote user stores like LDAP
1600 to be merged into local user stores like /etc/groups in order to provide
1601 easy to use, updated, and managed sets of merged credentials. The new
1602 action can be used by configuring it in /etc/nsswitch.conf:
1603 group: files [SUCCESS=merge] nis
1604 Implemented by Stephen Gallagher (Red Hat).
1606 * The deprecated __malloc_initialize_hook variable has been removed from the
1609 * The long unused localedef --old-style option has been removed. It hasn't
1610 done anything in over 16 years. Scripts using this option can safely
1613 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1614 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1615 return the next representable value in the direction of positive infinity
1616 and the nextdown functions return the next representable value in the
1617 direction of negative infinity. These are currently enabled as GNU
1620 Security related changes:
1622 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1623 could result in a stack overflow when getnetbyname was called with an
1624 overly long name. (CVE-2016-3075)
1626 * Previously, getaddrinfo copied large amounts of address data to the stack,
1627 even after the fix for CVE-2013-4458 has been applied, potentially
1628 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1629 instead. Reported by Michael Petlan. (CVE-2016-3706)
1631 * The glob function suffered from a stack-based buffer overflow when it was
1632 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1633 Reported by Alexander Cherepanov. (CVE-2016-1234)
1635 * The Sun RPC UDP client could exhaust all available stack space when
1636 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1637 alloca plugin for GCC. (CVE-2016-4429)
1639 * The IPv6 name server management code in libresolv could result in a memory
1640 leak for each thread which is created, performs a failing naming lookup,
1641 and exits. Over time, this could result in a denial of service due to
1642 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1644 The following bugs are resolved with this release:
1646 [1170] localedata: ne_NP: update Nepali locale definition file
1647 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1649 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1650 [6796] math: fdim() does not set errno on overflow
1651 [10354] libc: posix_spawn should use vfork() in more cases than presently
1652 [11213] localedata: localedata: add copyright disclaimer to locale files
1653 [12143] localedata: chr_US: new Cherokee locale
1654 [12450] localedata: sgs_LT: new locale
1655 [12676] localedata: ln_CD: new locale
1656 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1658 [13304] math: fma, fmaf, fmal produce wrong results
1659 [14259] build: --localedir arg to configure is ignored
1660 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1661 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1662 [14934] localedata: es_CL: wrong first weekday chilean locale
1663 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1665 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1667 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1668 [15368] nptl: raise() is not async-signal-safe
1669 [15479] math: ceil, floor, round and trunc raise inexact exception
1670 [15578] localedata: kk_KZ: various updates
1671 [16003] localedata: pap_AN: punt old locale
1672 [16137] localedata: iw_IL: punt old locale
1673 [16190] localedata: eo: new esperanto locale
1674 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1675 [16742] malloc: race condition: pthread_atfork() called before first
1676 malloc() results in unexpected locking behaviour/deadlocks
1677 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1679 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1680 [17565] localedata: pt_PT: wrong (work-)week start
1681 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1682 [17950] build: Build fails with -msse
1683 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1684 [18433] libc: posix_spawn does not return correctly upon failure to
1686 [18453] localedata: charmaps/IBM875: incorrect codes
1687 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1689 [18896] localedata: he_IL: improvements for currency
1690 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1691 Romanian locale data
1692 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1694 [19056] libc: Deprecate readdir_r
1695 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1697 [19198] localedata: nl_NL: small improvements for Dutch locales
1698 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1699 nameservers (CVE-2016-5417)
1700 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1702 [19400] locale: Language missing in "iso-639.def", trivial fix in
1704 [19431] malloc: Deadlock between fflush, getdelim, and fork
1705 [19505] libc: Incorrect file descriptor validity checks in
1706 posix_spawn_file_actions_add{open,close,dup2}
1707 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1708 when using RTLD_NEXT
1709 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1710 `intl/{gettextP,loadinfo}.h'
1711 [19534] libc: execle, execlp may use malloc
1712 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1713 [19573] network: res_nclose and __res_maybe_init disagree about name
1714 server initialization, breaking Hesiod
1715 [19575] localedata: Status of GB18030 tables
1716 [19581] localedata: sr_* date_fmt string contains additional newline
1717 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1719 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1720 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1721 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1722 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1724 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1726 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1728 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1729 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1730 [19633] locale: strfmon_l applies global locale to number formatting
1731 [19642] network: Memory leak in getnameinfo
1732 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1733 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1735 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1736 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1738 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1739 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1741 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1743 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1745 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1746 pointers and lengths in error-case.
1747 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1748 report errors on UTF-16 surrogates values.
1749 [19755] nscd: nscd assertion failure in gc
1750 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1751 [19759] libc: mempcpy shouldn't be inlined
1752 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1753 [19765] libc: s390 needs an optimized mempcpy
1754 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1755 NAME_MAX limit assumption (CVE-2016-1234)
1756 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1757 [19787] network: Missing and incorrect truncation checks in getnameinfo
1758 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1760 [19791] network: Assertion failure in res_query.c with un-connectable name
1762 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1763 [19822] math: libm.so install clobbers old version
1764 [19825] network: resolv: send_vc can return uninitialized data in second
1765 response to getaddrinfo
1766 [19830] network: nss_dns: should check RDATA length against buffer length
1767 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1768 confronted with A/AAAA records of invalid size
1769 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1770 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1771 [19853] stdio: Printing IBM long double in decimal with high precision is
1773 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1774 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1775 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1777 [19865] network: Assertion failure or memory leak in
1778 _nss_dns_getcanonname_r
1779 [19868] network: nss_dns: netent code does not skip over non-PTR records
1780 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1782 [19881] string: Improve x86-64 memset
1783 [19907] string: Incorrect memcpy tests
1784 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1786 [19925] libc: termios.h XCASE namespace
1787 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1788 [19929] libc: limits.h NL_NMAX namespace
1789 [19931] stdio: Memory leak in vfprintf
1790 [19957] libc: clone(CLONE_VM) access invalid parent memory
1791 [19963] localedata: en_IL: New locale
1792 [19989] stdio: stdio.h cuserid namespace
1793 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1795 [19996] locale: langinfo.h nl_langinfo_l namespace
1796 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1797 [20010] network: getaddrinfo: Stack overflow in hostent translation
1799 [20012] stdio: libio: fmemopen append mode failure
1800 [20014] stdio: stdio.h namespace for pre-threads POSIX
1801 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1802 [20023] libc: fcntl.h timespec namespace
1803 [20024] math: [x86_64] vectorized sincos trashes the stack
1804 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1805 [20041] time: sys/time.h timespec namespace
1806 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1807 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1808 [20051] libc: ttyslot in wrong header under wrong conditions
1809 [20054] libc: gethostname not declared for XPG4
1810 [20055] libc: termios.h missing tcgetsid for XPG4
1811 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1813 [20073] libc: sys/stat.h fchmod namespace
1814 [20074] libc: stdlib.h rand_r namespace
1815 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1816 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1818 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1819 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1821 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1822 [20119] libc: Wrong mask for processors level type from CPUID
1823 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1824 AS not supporting AVX512
1825 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1827 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1828 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1830 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1831 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1832 [20178] libc: posix_spawn{p} should not call exit
1833 [20191] stdio: libio: vtables hardening
1834 [20195] string: FMA4 detection requires CPUID execution with register
1836 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1837 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1839 [20212] math: acos (sNaN) returns sNaN
1840 [20213] math: asin (sNaN) returns sNaN
1841 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1842 [20218] math: [i386] asinhl (sNaN) returns sNaN
1843 [20219] math: [i386] atanhl (sNaN) returns sNaN
1844 [20222] stdio: fopencookie: Mangle function pointers
1845 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1846 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1847 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1849 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1850 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1851 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1852 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1853 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1854 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1855 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1856 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1857 [20240] math: modf (sNaN) returns sNaN
1858 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1859 [20250] math: frexp (sNaN) returns sNaN
1860 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1861 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1863 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1864 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1865 subscript is below array bounds [-Werror=array-bounds]
1866 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1867 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1869 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1870 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1872 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1873 "invalid" exceptions
1874 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1875 [20316] localedata: id_ID: Februari instead of Pebruari
1876 [20327] string: POWER8 strcasecmp returns incorrect result
1877 [20347] math: Failure: Test: j0_downward (0xap+0)
1878 [20348] libc: FAIL: misc/tst-preadvwritev64
1879 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1880 [20350] libc: There is no test for p{read,write}64
1881 [20357] math: Incorrect cos result for 1.5174239687223976
1882 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1886 * Unicode 8.0.0 Support: Character encoding, character type info, and
1887 transliteration tables are all updated to Unicode 8.0.0, using new
1888 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1889 These updates cause user visible changes, such as the fixes for bugs
1890 89, 16061, and 18568.
1892 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1893 kernel-internal CPU set size. This means that requests that change the
1894 CPU affinity which failed before (for example, an all-ones CPU mask) will
1895 now succeed. Applications that need to determine the effective CPU
1896 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1897 setting it because the kernel can adjust it (and the previous size check
1898 would not detect this in the majority of cases).
1900 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1901 the following new symbols are used: fts64_children, fts64_close,
1902 fts64_open, fts64_read and fts64_set.
1904 * getaddrinfo now detects certain invalid responses on an internal netlink
1905 socket. If such responses are received, an affected process will
1906 terminate with an error message of "Unexpected error <number> on netlink
1907 descriptor <number>" or "Unexpected netlink response of size <number> on
1908 descriptor <number>". The most likely cause for these errors is a
1909 multi-threaded application which erroneously closes and reuses the netlink
1910 file descriptor while it is used by getaddrinfo.
1912 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1913 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1914 unnecessary serialization of memory allocation requests across threads.
1915 The defect is now corrected. Users should see a substantial increase in
1916 the concurent throughput of allocation requests for applications which
1917 trigger this bug. Affected applications typically create and
1918 destroy threads frequently. (Bug 19048 was reported and analyzed by
1921 * There is now a --disable-timezone-tools configure option for disabling the
1922 building and installing of the timezone related utilities (zic, zdump, and
1923 tzselect). This is useful for people who build the timezone data and code
1924 independent of the GNU C Library.
1926 * The obsolete header <regexp.h> has been removed. Programs that require
1927 this header must be updated to use <regex.h> instead.
1929 * The obsolete functions bdflush, create_module, get_kernel_syms,
1930 query_module and uselib are no longer available to newly linked binaries;
1931 the header <sys/kdaemon.h> has been removed. These functions and header
1932 were specific to systems using the Linux kernel and could not usefully be
1933 used with the GNU C Library on systems with version 2.6 or later of the
1936 * Optimized string, wcsmbs and memory functions for IBM z13.
1937 Implemented by Stefan Liebler.
1939 * Newly linked programs that define a variable called signgam will no longer
1940 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1941 require signgam to be set by those functions must ensure that they use the
1942 variable provided by the GNU C Library and declared in <math.h>, without
1943 defining their own copy.
1945 * The minimum GCC version that can be used to build this version of the GNU
1946 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1947 still be used to compile programs using the GNU C Library.
1949 Security related changes:
1951 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1952 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1954 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1955 the pointer guard feature. It is always enabled. Previously,
1956 LD_POINTER_GUARD could be used to disable security hardening in binaries
1957 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1960 * An integer overflow in hcreate and hcreate_r could lead to an
1961 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1963 * The catopen function no longer has unbounded stack usage. Reported by
1964 Max. (CVE-2015-8779)
1966 * The nan, nanf and nanl functions no longer have unbounded stack usage
1967 depending on the length of the string passed as an argument to the
1968 functions. Reported by Joseph Myers. (CVE-2014-9761)
1970 * A stack-based buffer overflow was found in libresolv when invoked from
1971 libnss_dns, allowing specially crafted DNS responses to seize control
1972 of execution flow in the DNS client. The buffer overflow occurs in
1973 the functions send_dg (send datagram) and send_vc (send TCP) for the
1974 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1975 family. The use of AF_UNSPEC triggers the low-level resolver code to
1976 send out two parallel queries for A and AAAA. A mismanagement of the
1977 buffers used for those queries could result in the response of a query
1978 writing beyond the alloca allocated buffer created by
1979 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1980 the overflow. Thanks to the Google Security Team and Red Hat for
1981 reporting the security impact of this issue, and Robert Holiday of
1982 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1984 The following bugs are resolved with this release:
1986 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1987 [887] math: Math library function "logb" and "nextafter" inconsistent
1988 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1989 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1990 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1991 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1993 [4404] localedata: German translation of "Alarm clock" is misleading
1994 [6799] math: nextafter() and nexttoward() doen't set errno on
1995 overflow/underflow errors
1996 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1998 [10432] nis: _nss_nis_setnetgrent assertion failure
1999 [11460] libc: fts has no LFS support
2000 [12926] network: getaddrinfo()/make_request() may spin forever
2001 [13065] nptl: Race condition in pthread barriers
2002 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2003 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2005 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2006 [14912] libc: Rename non-installed bits/*.h headers
2007 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2008 [15367] math: Let gcc use __builtin_isinf
2009 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2010 [15421] math: lgamma wrongly sets signgam for ISO C
2011 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2012 FE_INVALID with argument out of range
2013 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2015 [15786] dynamic-link: ifunc resolver functions can smash function
2017 [15918] math: Unnecessary check for equality in hypotf()
2018 [16061] localedata: Review / update transliteration data
2019 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2021 [16141] time: strptime %z offset restriction
2022 [16171] math: drem should be alias of remainder
2023 [16296] math: fegetround is pure?
2024 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2025 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2027 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2028 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2030 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2031 "Exception "Inexact" set" on ppc32
2032 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2033 [16517] math: Missing underflow exception from tanf/tan/tanl
2034 [16519] math: Missing underflow exception from sinhf
2035 [16520] math: Missing underflow exception from tanhf
2036 [16521] math: Missing underflow exception from exp2
2037 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2039 [16734] stdio: fopen calls mmap to allocate its buffer
2040 [16961] math: nan function incorrect handling of bad sequences
2041 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2042 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2043 [16985] locale: localedef: confusing error message when opening output
2045 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2046 [17197] locale: Redundant shift character in iconv conversion output at
2048 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2049 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2050 [17250] dynamic-link: static linking breaks nss loading
2051 (getaddrinfo/getpwnam/etc...)
2052 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2054 [17441] math: isnan() should use __builtin_isnan() in GCC
2055 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2056 (related to lock elision)
2057 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2058 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2059 [17887] time: strptime should be able to parse "+01:00" style timezones
2060 [17905] libc: catopen() Multiple unbounded stack allocations
2062 [18084] libc: backtrace (..., 0) dumps core on x86
2063 [18086] libc: nice() sets errno to 0 on success
2064 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2065 is too large (CVE-2015-8778)
2066 [18251] dynamic-link: SONAME missing when audit modules provides path
2067 [18265] libc: add attributes for wchar string and memory functions
2068 [18370] math: csqrt missing underflows
2069 [18421] libc: [hppa] read-only segment has dynamic relocations
2070 [18472] libc: Obsolete syscall wrappers should be compat symbols
2071 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2072 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2073 [18525] localedata: Remove locale timezone information
2074 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2075 [18568] localedata: Update locale data to Unicode 8.0
2076 [18589] locale: sort-test.sh fails at random
2077 [18595] math: ctan, ctanh missing underflows
2078 [18604] libc: assert macro-expands its argument
2079 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2080 contains a vector instruction exception.
2081 [18611] math: j1, jn missing errno setting on underflow
2082 [18618] localedata: sync Chechen locale definitions with other *_RU
2084 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2085 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2086 [18665] network: In send_dg, the recvfrom function is NOT always using the
2087 buffer size of a newly created buffer (CVE-2015-7547)
2088 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2090 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2092 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2093 [18699] math: tilegx cproj() for various complex infinities does not yield
2095 [18724] libc: Harden put*ent functions against data injection
2096 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2097 [18755] build: build errors with -DNDEBUG
2098 [18757] stdio: fmemopen fails to set errno on failure
2099 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2101 [18781] libc: openat64 lacks O_LARGEFILE
2102 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2103 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2104 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2105 [18790] math: [ldbl-128ibm] tanhl inaccurate
2106 [18795] libc: stpncpy fortification misses buffer lengths that are
2107 statically too large
2108 [18796] build: build fails for --disable-mathvec
2109 [18803] math: hypot missing underflows
2110 [18820] stdio: fmemopen may leak memory on failure
2111 [18823] math: csqrt spurious underflows
2112 [18824] math: fma spurious underflows
2113 [18825] math: pow missing underflows
2114 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2115 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2116 [18870] build: sem_open.c fails to compile with missing symbol
2118 [18872] stdio: Fix memory leak in printf_positional
2119 [18873] libc: posix_fallocate overflow check ineffective
2120 [18875] math: Excess precision leads incorrect libm
2121 [18877] libc: arm: mmap offset regression
2122 [18887] libc: memory corruption when using getmntent on blank lines
2123 [18918] localedata: hu_HU: change time to HH:MM:SS format
2124 [18921] libc: Regression: extraneous stat() and fstat() performed by
2126 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2127 binaries (CVE-2015-8777)
2128 [18951] math: tgamma missing underflows
2129 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2131 [18953] localedata: lt_LT: change currency symbol to the euro
2132 [18956] math: powf inaccuracy
2133 [18961] math: [i386] exp missing underflows
2134 [18966] math: [i386] exp10 missing underflows
2135 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2136 [18969] build: multiple string test failures due to missing locale
2138 [18970] libc: Reference of pthread_setcancelstate in libc.a
2139 [18977] math: float / long double Bessel functions not in XSI POSIX
2140 [18980] math: i386 libm functions return with excess range and precision
2141 [18981] math: i386 scalb*, ldexp return with excess range and precision
2142 [18982] stdio: va_list and vprintf
2143 [18985] time: Passing out of range data to strftime() causes a segfault
2145 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2146 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2147 [19012] locale: iconv_open leaks memory on error path
2148 [19016] math: clog, clog10 inaccuracy
2149 [19018] nptl: Mangle function pointers in tls_dtor_list
2150 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2151 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2152 [19048] malloc: malloc: arena free list can become cyclic, increasing
2154 [19049] math: [powerpc] erfc incorrect zero sign
2155 [19050] math: [powerpc] log* incorrect zero sign
2156 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2157 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2158 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2159 [19074] network: Data race in _res_hconf_reorder_addrs
2160 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2161 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2162 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2164 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2166 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2167 [19086] manual: posix_fallocate64 documented argument order is wrong.
2168 [19088] math: lround, llround missing exceptions close to overflow
2170 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2171 [19095] math: dbl-64 lrint incorrect for 64-bit long
2172 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2173 [19124] dynamic-link: ld.so failed to build with older assmebler
2174 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2175 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2176 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2177 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2178 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2179 pthread_setaffinity_np
2180 [19156] math: [ldbl-128] j0l spurious underflows
2181 [19164] nptl: tst-getcpu fails with many possible CPUs
2182 [19168] math: math/test-ildoubl and math/test-ldouble failure
2183 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2184 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2186 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2187 (FE_NOMASK_ENV) do not clear SSE exceptions
2188 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2189 _int_new_arena/reused_arena
2190 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2191 [19201] math: dbl-64 remainder incorrect sign of zero result
2192 [19205] math: bits/math-finite.h conditions do not match math.h and
2194 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2195 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2196 for C99-based standards
2197 [19212] libc: features.h not -Wundef clean
2198 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2200 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2201 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2202 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2204 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2205 "inexact" exceptions
2206 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2208 [19242] libc: strtol incorrect in Turkish locales
2209 [19243] malloc: reused_arena can pick an arena on the free list, leading
2210 to an assertion failure and reference count corruption
2211 [19253] time: tzset() ineffective when temporary TZ did not include DST
2213 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2214 [19270] math: [hppa] Shared libm missing __isnanl
2215 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2216 MAP_HUGETLB and MAP_STACK defines
2217 [19313] nptl: Wrong __cpu_mask for x32
2218 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2219 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2220 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2221 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2222 [19363] time: x32: times() return value wrongly truncates/sign extends
2224 [19367] dynamic-link: Improve branch prediction on Silvermont
2225 [19369] network: Default domain name not reset by res_ninit when "search"
2226 / "domain" entry is removed from resolv.conf
2227 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2229 [19385] localedata: bg_BG: time separator should be colon, not comma
2230 [19408] libc: linux personality syscall wrapper may erroneously return an
2231 error on 32-bit architectures
2232 [19415] libc: dladdr returns wrong names on hppa
2233 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2234 IBM905, IBM907, and IBM909
2235 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2236 [19443] build: build failures with -DDEBUG
2237 [19451] build: Make check fails on test-double-vlen2
2238 [19462] libc: Glibc failed to build with -Os
2239 [19465] math: Wrong code with -Os
2240 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2242 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2244 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2245 [19486] math: S390: Math tests fail with "Exception Inexact set".
2246 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2247 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2249 [19590] math: Fail to build shared objects that use libmvec.so functions.
2253 * The following bugs are resolved with this release:
2255 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2256 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2257 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2258 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2259 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2260 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2261 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2262 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2263 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2264 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2265 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2266 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2267 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2268 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2269 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2270 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2271 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2272 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2273 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2274 18657, 18676, 18694, 18696, 18887.
2276 * Cache information can be queried via sysconf() function on s390 e.g. with
2277 _SC_LEVEL1_ICACHE_SIZE as argument.
2279 * A buffer overflow in gethostbyname_r and related functions performing DNS
2280 requests has been fixed. If the NSS functions were called with a
2281 misaligned buffer, the buffer length change due to pointer alignment was
2282 not taken into account. This could result in application crashes or,
2283 potentially arbitrary code execution, using crafted, but syntactically
2284 valid DNS responses. (CVE-2015-1781)
2286 * The time zone file parser has been made more robust against crafted time
2287 zone files, avoiding heap buffer overflows related to the processing of
2288 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2289 large time zone data files. Overly long time zone specifiers in the TZ
2290 variable no longer result in stack overflows and crashes.
2292 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2293 for LD and GD on x86 and x86-64, has been implemented. You will need
2294 binutils-2.24 or later to enable this optimization.
2296 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2297 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2298 Hat). These updates cause user visible changes, such as the fix for bug
2301 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2302 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2303 condition in some applications.
2305 * Added vector math library named libmvec with the following vectorized x86_64
2306 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2308 The library can be disabled with --disable-mathvec. Use of the functions is
2309 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2310 Shared library libmvec.so is linked in as needed when using -lm (no need to
2311 specify -lmvec explicitly for not static builds).
2312 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2314 * A new fmemopen implementation has been added with the goal of POSIX
2315 compliance. The new implementation fixes the following long-standing
2316 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2317 old implementation is still present for use be by existing binaries.
2319 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2320 releases. It has been fixed to match 2.19 and older, but binaries built
2321 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2323 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2324 Contributed by Roland McGrath (Google).
2326 * The header <regexp.h> is deprecated, and will be removed in a future
2327 release. Use of this header will trigger a deprecation warning.
2328 Application developers should update their code to use <regex.h> instead.
2330 This header was formerly part of SUS, but was deprecated in 1994 and
2331 removed from the standard in 2001. Also, the glibc implementation
2332 leaks memory. See BZ#18681 for more details.
2336 * The following bugs are resolved with this release:
2338 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2339 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2340 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2341 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2342 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2343 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2344 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2345 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2346 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2347 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2350 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2351 too little memory for the to-be-scanned arguments and overflow the
2352 allocated buffer. The implementation now correctly computes the required
2353 buffer size when using malloc, and switches to malloc from alloca as
2356 * A new semaphore algorithm has been implemented in generic C code for all
2357 machines. Previous custom assembly implementations of semaphore were
2358 difficult to reason about or ensure that they were safe. The new version
2359 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2360 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2361 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2363 * Port to Altera Nios II has been contributed by Mentor Graphics.
2365 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2366 implementations for powerpc64/powerpc64le.
2367 Implemented by Adhemerval Zanella (IBM).
2369 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2370 and powerpc64le. This may improve lock scaling of existing programs on
2371 HTM capable systems. The lock elision code is only enabled with
2372 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2373 powerpc will issue a transaction abort on every syscall to avoid side
2374 effects being visible outside transactions.
2376 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2377 AArch64. Contributed by ARM Ltd.
2379 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2381 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2382 under certain input conditions resulting in the execution of a shell for
2383 command substitution when the application did not request it. The
2384 implementation now checks WRDE_NOCMD immediately before executing the
2385 shell and returns the error WRDE_CMDSUB as expected.
2387 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2388 processing format strings with a large number of format specifiers.
2390 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2391 infinite loop if the DNS response contained a PTR record of an unexpected
2394 * The minimum GCC version that can be used to build this version of the GNU
2395 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2396 still be used to compile programs using the GNU C Library.
2398 * The GNU C Library is now built with -Werror by default. This can be
2399 disabled by configuring with --disable-werror.
2401 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2403 * The obsolete sigvec function has been removed. This was the original
2404 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2405 programs have been using instead for about 25 years. Of course, ABI
2406 compatibility for old binaries using sigvec remains intact.
2408 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2409 with newer versions of bison.
2411 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2412 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2413 registers overlay two consecutive single-precision registers. MIPS32R2
2414 introduced a new FPU mode (FR=1) where double-precision registers extend the
2415 corresponding single-precision registers which is incompatible with the
2416 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2417 require the use of FR=1 making a transition necessary. New o32 ABI
2418 extensions enable users to migrate over time from the original o32 ABI
2419 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2420 tracks the ABI of any loaded object and verifies that new objects are
2421 compatible. Mode transitions will also be requested as required and
2422 unsupportable objects will be rejected. The ABI checks include both soft and
2423 hard float ABIs for o32, n32 and n64.
2425 GCC 5 with GNU binutils 2.25 onwards:
2426 It is strongly recommended that all o32 system libraries are built using the
2427 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2428 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2429 compiler using --with-fp-32=xx to set this by default.
2433 * The following bugs are resolved with this release:
2435 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2436 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2437 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2438 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2439 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2440 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2441 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2442 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2443 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2444 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2445 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2446 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2447 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2448 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2449 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2451 * Reverted change of ABI data structures for s390 and s390x:
2452 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2453 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2454 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2455 are preserved pointing straight to the same implementation as the old ones.
2456 Given that, new callers will simply provide a too-big buffer to these
2457 functions. Any applications/libraries out there that embed jmp_buf or
2458 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2459 against 2.19 headers will have to rebuilt again. This is necessary in any
2460 case to revert the breakage in their ABI caused by the glibc change.
2462 * Support for file description locks is added to systems running the
2463 Linux kernel. The standard file locking interfaces are extended to
2464 operate on file descriptions, not file descriptors, via the use of
2465 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2466 are associated with an open file instead of a process.
2468 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2470 * The minimum Linux kernel version that this version of the GNU C Library
2471 can be used with is 2.6.32.
2473 * Running the testsuite no longer terminates as soon as a test fails.
2474 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2475 with PASS or FAIL lines for individual tests. A summary of the results is
2476 printed, including a list of failing lists, and "make check" exits with
2477 error status if there were any unexpected failures. "make check
2478 stop-on-test-failure=y" may be used to keep the old behavior.
2480 * The am33 port, which had not worked for several years, has been removed
2483 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2484 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2485 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2486 conflicted with POSIX (support for which was removed in 2.19), the
2487 interfaces those macros enabled remain available when compiling with
2488 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2489 test macros defined.
2491 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2493 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2494 This may improve lock scaling of existing programs on TX capable systems.
2495 The lock elision code is only built with --enable-lock-elision=yes and
2496 then requires a GCC version supporting the TX builtins. With lock elision
2497 default mutexes are elided via __builtin_tbegin, if the cpu supports
2498 transactions. By default lock elision is not enabled and the elision code
2501 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2502 copy the path argument. This allowed programs to cause posix_spawn to
2503 deference a dangling pointer, or use an unexpected pathname argument if
2504 the string was modified after the posix_spawn_file_actions_addopen
2507 * All supported architectures now use the main glibc sysdeps directory
2508 instead of some being in a separate "ports" directory (which was
2509 distributed separately before glibc 2.17).
2511 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2512 On configurations that support it (all Linux configurations), it's now
2513 used regardless of the --enable-add-ons switch to configure. It is no
2514 longer possible to build such configurations without pthreads support.
2516 * Locale names, including those obtained from environment variables (LANG
2517 and the LC_* variables), are more tightly checked for proper syntax.
2518 setlocale will now fail (with EINVAL) for locale names that are overly
2519 long, contain slashes without starting with a slash, or contain ".." path
2520 components. (CVE-2014-0475) Previously, some valid locale names were
2521 silently replaced with the "C" locale when running in AT_SECURE mode
2522 (e.g., in a SUID program). This is no longer necessary because of the
2525 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2526 with application code using Intel MPX instructions. (With all previous
2527 versions, the MPX register state could be clobbered when making calls
2528 into or out of a shared library.) Note that while the new dynamic
2529 linker is compatible with all known x86 hardware whether or not it
2530 supports Intel MPX, some x86 instruction-set emulators might fail to
2531 handle the new instruction encodings. This is known to affect Valgrind
2532 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2533 release), and might affect other tools that do instruction emulation.
2535 * Support for loadable gconv transliteration modules has been removed.
2536 The support for transliteration modules has been non-functional for
2537 over a decade, and the removal is prompted by security defects. The
2538 normal gconv conversion modules are still supported. Transliteration
2539 with //TRANSLIT is still possible, and the //IGNORE specifier
2540 continues to be supported. (CVE-2014-5119)
2542 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2543 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2544 resulting a denial-of-service security vulnerability in applications which
2545 use functions related to iconv. (CVE-2014-6040)
2549 * The following bugs are resolved with this release:
2551 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2552 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2553 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2554 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2555 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2556 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2557 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2558 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2559 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2560 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2561 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2562 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2563 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2564 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2565 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2566 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2567 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2568 16431, 16453, 16474, 16506, 16510, 16529
2570 * Slovenian translations for glibc messages have been contributed by the
2571 Translation Project's Slovenian team of translators.
2573 * The public headers no longer use __unused nor __block. This change is to
2574 support compiling programs that are derived from BSD sources and use
2575 __unused internally, and to support compiling with Clang's -fblock
2576 extension which uses __block.
2578 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2579 large collation sequences to optimize multiple passes. This cache
2580 computation may overflow for large collation sequences and may cause a
2581 stack or buffer overflow. This is now fixed to use a slower algorithm
2582 which does not use a cache if there is an integer overflow.
2584 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2585 rules for large collation sequences to optimize multiple passes and falls
2586 back to alloca if malloc fails, resulting in a possible stack overflow.
2587 The implementation now falls back to an uncached collation sequence lookup
2590 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2591 initialized for static applications resulting in the security feature
2592 being disabled. The pointer guard is now correctly initialized to a
2593 random value for static applications. Existing static applications need
2594 to be recompiled to take advantage of the fix (bug 15754).
2596 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2597 to the d_name member of struct dirent, or omit the terminating NUL
2598 character. (Bugzilla #14699).
2600 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2601 aligned_alloc functions could allocate too few bytes or corrupt the
2602 heap when passed very large allocation size values (Bugzilla #15855,
2605 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2606 for AF_INET6 has been fixed (Bugzilla #16072).
2608 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2609 pap_CW, quz_PE, the_NP.
2611 * Substantially revised locales: gd_GB, ht_HT
2613 * The LC_ADDRESS field was updated to support country_car for almost all
2616 * ISO 1427 definitions were updated.
2618 * ISO 3166 definitions were updated.
2620 * The localedef utility now supports --big-endian and --little-endian
2621 command-line options to generate locales for a different system from that
2622 for which the C library was built.
2624 * Binary locale files now only depend on the endianness of the system for
2625 which they are generated and not on other properties of that system. As a
2626 consequence, binary files generated with new localedef may be incompatible
2627 with old versions of the GNU C Library, and binary files generated with
2628 old localedef may be incompatible with this version of the GNU C Library,
2629 in the following circumstances:
2631 + Locale files may be incompatible on m68k systems.
2633 + Locale archive files (but not separate files for individual locales) may
2634 be incompatible on systems where plain "char" is signed.
2636 * The configure option --disable-versioning has been removed. Builds with
2637 --disable-versioning had not worked for several years.
2639 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2640 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2642 * SystemTap probes for malloc have been introduced.
2644 * SystemTap probes for slow multiple precision fallback paths of
2645 transcendental functions have been introduced.
2647 * Support for powerpc64le has been added.
2649 * The soft-float powerpc port now supports e500 processors.
2651 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2653 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2654 set of header declarations that are enabled by default, even when other
2655 feature test macros or compiler options such as -std=c99 would otherwise
2656 disable some of those declarations.
2658 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2659 conflict with POSIX. The libbsd-compat library (which was a dummy library
2660 that did nothing) has also been removed.
2662 * Preliminary documentation about Multi-Thread, Async-Signal and
2663 Async-Cancel Safety has been added.
2665 * Change of ABI data structures for s390 and s390x:
2666 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2667 allow for future hardware extensions. All library functions that accept or
2668 return these structures were versioned in order to provide backwards
2669 compatibility with existing code. However, not all cases can be handled
2670 with interface versioning. If an application embeds either structure into
2671 another structure and that structure is passed to another compilation unit
2672 which expects the newer larger-sized object then any access to the new
2673 fields will result in undefined behaviour. Similarly any access to fields
2674 that were moved by the enlarging of the structures will cause undefined
2675 behaviour. To fix the undefined behaviour all objects that are part of the
2676 final in-memory image must agree on the size of structures, and this may
2677 require recompilation.
2681 * The following bugs are resolved with this release:
2683 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2684 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2685 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2686 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2687 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2688 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2689 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2690 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2691 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2692 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2693 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2694 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2695 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2698 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2699 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2700 Distributions can re-enable building and using pt_chown via the new configure
2701 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2702 considerable security risks and should only be used if the distribution
2703 understands and accepts the risks.
2705 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2708 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2709 fixed (Bugzilla #15330).
2711 * Add support for calling C++11 thread_local object destructors on thread
2712 and program exit. This needs compiler support for offloading C++11
2713 destructor calls to glibc.
2715 * Improved worst case performance of libm functions with double inputs and
2718 * Support for priority inherited mutexes in pthread condition variables on
2719 non-x86 architectures.
2721 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2723 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2725 * Optimized string functions for ARM. Implemented by Will Newton and
2728 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2730 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2733 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2734 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2736 * Added a benchmark framework to track performance of functions in glibc.
2738 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2739 It is based on draft TS 18661 and currently enabled as a GNU extension.
2741 * On Linux, the clock function now uses the clock_gettime system call
2742 for improved precision, rather than old times system call.
2744 * Added support for version-3 tz format files. This is needed when using
2745 the tz database release 2013e or later, and affects a few unusual cases --
2746 currently only TZ='America/Godthab' for time stamps after 2037.
2748 * Added new API functions pthread_getattr_default_np and
2749 pthread_setattr_default_np to get and set the default pthread
2750 attributes of a process.
2752 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2753 This may improve lock scaling of existing programs on TSX capable systems.
2754 When the --enable-lock-elision=yes parameter is specified at configure
2755 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2758 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2759 availability). Implemented by Ryan S. Arnold.
2761 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2765 * The following bugs are resolved with this release:
2767 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2768 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2769 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2770 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2771 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2772 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2773 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2774 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2775 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2776 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2777 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2778 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2779 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2780 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2781 14879, 14889, 14893, 14898, 14914.
2783 * Optimization of memcpy for MIPS.
2785 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2786 EMFILE has been fixed (Bugzilla #14889).
2788 * The add-on ports collection is now distributed in the "ports" subdirectory
2789 of the main GNU C Library distribution, rather than separately.
2791 * Port to ARM AArch64 contributed by Linaro.
2793 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2794 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2796 Implemented by Andreas Krebbel.
2798 * The new function secure_getenv allows secure access to the environment,
2799 returning NULL if running in a SUID/SGID process. This function replaces
2800 the internal function __secure_getenv.
2802 * SystemTap static probes have been added into the dynamic linker.
2803 Implemented by Gary Benson.
2805 * Optimizations of string functions strstr, strcasestr and memmem.
2806 Implemented by Maxim Kuvyrkov.
2808 * The minimum Linux kernel version that this version of the GNU C Library
2809 can be used with is 2.6.16.
2811 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2812 powerpc POWER7. Implemented by Will Schmidt.
2814 * New configure option --disable-nscd builds the C library such that it
2815 never attempts to contact the Name Service Caching Daemon (nscd).
2816 New configure option --disable-build-nscd avoids building nscd itself;
2817 this is the default if --disable-nscd is used.
2819 * Improved support for cross-compilation, including cross-testing and
2820 bootstrap builds without a previously built glibc.
2822 * Several testsuite tests are now able to test multiple IFUNC variants of an
2823 interface, rather than just testing the one that would be chooen by
2826 * New configure options --with-bugurl and --with-pkgversion, for
2827 distributors to use to embed their bug-reporting and package version
2828 information in --help and --version output.
2830 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2831 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2832 allows creation of chroots without the procfs mounted on /proc.
2834 * The `crypt' function now fails if passed salt bytes that violate the
2835 specification for those values. On Linux, the `crypt' function will
2836 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2837 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2838 when the mode is enabled.
2840 * The `clock_*' suite of functions (declared in <time.h>) is now available
2841 directly in the main C library. Previously it was necessary to link with
2842 -lrt to use these functions. This change has the effect that a
2843 single-threaded program that uses a function such as `clock_gettime' (and
2844 is not linked with -lrt) will no longer implicitly load the pthreads
2845 library at runtime and so will not suffer the overheads associated with
2846 multi-thread support in other code such as the C++ runtime library.
2848 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2853 * The following bugs are resolved with this release:
2855 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2856 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2857 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2858 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2859 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2860 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2861 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2862 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2863 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2864 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2865 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2866 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2867 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2868 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2869 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2870 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2871 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2872 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2873 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2874 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2875 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2876 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2879 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2880 configuring glibc with:
2881 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2882 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2883 Implemented by H.J. Lu.
2887 + define static_assert
2889 + do not declare gets
2891 + declare at_quick_exit and quick_exit also for ISO C11
2893 + aligned_alloc. NB: The code is deliberately allows the size parameter
2894 to not be a multiple of the alignment. This is a moronic requirement
2895 in the standard but it is only a requirement on the caller, not the
2898 + timespec_get added
2900 + uchar.h support added
2902 + CMPLX, CMPLXF, CMPLXL added
2904 Implemented by Ulrich Drepper.
2906 * Support for the IA-64 has been moved to ports.
2908 * Remove support for anything but ELF binary format
2910 * Checking versions of poll, ppoll added.
2911 Implemented by Ulrich Drepper.
2913 * More generic and 64-bit performance optimizations to math functions.
2914 Implemented by Ulrich Drepper.
2916 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2917 headers and functions available at compile time as they were before
2918 version 2.14. This option will be removed at some time in the future
2919 after the TI-RPC library becomes fully sufficient for the needs of
2920 existing applications.
2922 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2923 Note that glibc is not expected to work with any Linux kernel version
2926 * New header <sys/auxv.h> and function getauxval allowing easy access to
2927 the AT_* key-value pairs passed from the Linux kernel. The header also
2928 defines the HWCAP_* bits associated with the AT_HWCAP key.
2930 * New locales: mag_IN
2932 * New configure option --enable-systemtap builds SystemTap static probes
2933 into libc for setjmp and longjmp and into libpthread for various operations.
2934 So far the setjmp/longjmp probes and some of the libpthread probes are
2935 provided only for i*86 and x86_64.
2936 Implemented by Roland McGrath and Rayson Ho.
2938 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2940 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2943 * More optimized functions for SPARC. Implemented by David S. Miller.
2945 * Improved support for cross-compilation, especially bootstrap builds
2946 without a previously built glibc.
2948 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2949 Chris Metcalf from Tilera.
2951 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2952 now supported for ARM processors.
2954 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2955 as the name of the dynamic linker, to distinguish it from the
2956 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2958 * Support for CRIS has been removed from ports.
2960 * A new class of installed header has been documented for low-level
2961 platform-specific functionality. PowerPC added the first instance with a
2962 function to provide time base register access. Contributed by Tulio
2963 Magno Quites Machado Filho.
2965 * ix86 configurations now install header files that are consistent with
2966 what x86-64 configurations install. These same header files can be used
2967 for -m32, -m64, or -mx32 builds.
2968 Contributed by H.J. Lu.
2970 * Math library bug fixes. A thorough audit of all open math library bugs was
2971 conducted by Joseph Myers. Significant progress was made on many math
2972 library bugs resulting in more accurate exceptions and function results.
2973 Many thanks to all those that contributed including Andreas Jaeger for his
2974 patch review and work on the x87 trigonometric instruction issues.
2976 * Timezone data is no longer installed. Timezone-related binaries and scripts
2977 will continue to be installed. Users should obtain their timezone data from
2978 their distribution provider or from the tzdata package at
2979 <ftp://munnari.oz.au/pub/>.
2983 * The following bugs are resolved with this release:
2985 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2986 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2987 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2988 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2989 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2990 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2991 13484, 13506, 13515, 13523, 13524, 13538, 13540
2993 * New program pldd to list loaded object of a process
2994 Implemented by Ulrich Drepper.
2996 * Add nss_db support back to glibc. No more dependency on Berkeley db
2997 and support for initgroups lookups.
2998 Implemented by Ulrich Drepper.
3000 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3001 Contributed by HJ Lu.
3003 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3004 Contributed by HJ Lu.
3006 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3007 on x86-32 and x86-64.
3008 Contributed by Liubov Dmitrieva.
3010 * Optimized strchr and strrchr for SSE on x86-32.
3011 Contributed by Liubov Dmitrieva.
3013 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3014 for x86-64 and x86-32.
3015 Contributed by Liubov Dmitrieva.
3017 * New interfaces: scandirat, scandirat64
3018 Implemented by Ulrich Drepper.
3020 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3021 Implemented by Ulrich Drepper.
3023 * nscd now also caches the netgroup database.
3024 Implemented by Ulrich Drepper.
3026 * Integrate libm with gcc's -ffinite-math-only option.
3027 Implemented by Ulrich Drepper.
3029 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3030 to math functions. Implemented by Ulrich Drepper.
3032 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3033 Implemented by Ulrich Drepper.
3035 * New Linux interfaces: process_vm_readv, process_vm_writev
3037 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3038 Implemented by Ulrich Drepper.
3040 * Optimized nearbyint and strcasecmp for PPC.
3041 Implemented by Adhemerval Zanella.
3043 * New locales: bho_IN, unm_US, es_CU, ta_LK
3047 * The following bugs are resolved with this release:
3049 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3050 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3051 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3052 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3053 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3054 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3055 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3056 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3057 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3058 12795, 12811, 12813, 12814, 12841
3060 * The RPC implementation in libc is obsoleted. Old programs keep working
3061 but new programs cannot be linked with the routines in libc anymore.
3062 Programs in need of RPC functionality must be linked against TI-RPC.
3063 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3065 Visible changes of this change include (obviously) the inability to link
3066 programs using RPC functions without referencing the TI-RPC library and the
3067 removal of the RPC headers from the glibc headers.
3068 Implemented by Ulrich Drepper.
3070 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3071 syncfs, setns, sendmmsg
3073 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3074 yue_HK, lij_IT, mhr_RU
3076 * New encodings: CP770, CP771, CP772, CP773, CP774
3078 * New program sotruss to trace calls through PLTs
3079 Implemented by Ulrich Drepper.
3081 * The malloc hook implementation is marked deprecated and will be removed
3082 from the default implementation in the next version. The design never
3083 worked ever since the introduction of threads. Even programs which do
3084 not create threads themselves can use multiple threads created internally.
3088 * The following bugs are resolved with this release:
3090 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3091 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3092 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3093 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3096 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3098 * POWER7 optimizations: memset, memcmp, strncmp
3100 * New optimized string functions for x86-64: strnlen (SSE2),
3101 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3102 Implemented by Ulrich Drepper.
3106 * The following bugs are resolved with this release:
3108 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3109 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3110 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3111 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3112 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3113 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3114 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3115 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3117 * New interfaces: pthread_getname_np, pthread_setname_np
3119 * New Linux interface: recvmmsg
3121 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3123 * The dynamic linker now recognizes supported ABI versions from the
3124 EI_ABIVERSION field in the ELF header.
3125 Implemented by Ulrich Drepper.
3127 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3128 will not be used to fill in password fields in the passwd.byname replies.
3129 Instead it is used to synthesize the shadow.byname table, should it be
3130 missing. This is a useful mode in some installations involving Solaris.
3131 Implemented by Ulrich Drepper.
3133 * New locales: kok_IN, sq_MK, cv_RU
3137 * The following bugs are resolved with this release:
3139 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3140 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3141 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3142 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3143 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3144 10789, 10817, 10823, 10840, 10847
3146 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3148 Implemented by Ulrich Drepper.
3150 * Checking version of longjmp added that fails if an uninitialized stack
3151 frame would be created. Implemented by Ulrich Drepper.
3153 * STT_GNU_IFUNC is now supported in static executables.
3154 Implemented by H.J. Lu.
3156 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3158 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3159 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3160 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3161 Contributed by H.J. Lu.
3163 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3164 Implemented by Ulrich Drepper.
3166 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3168 Contributed by H.J. Lu.
3170 * Support for fma instruction in AVX on x86-64.
3171 Implemented by H.J. Lu and Ulrich Drepper.
3173 * AVX support in x86-64 auditing support in ld.so.
3174 Implemented by H.J. Lu.
3176 * STB_GNU_UNIQUE support added.
3177 Implemented by Ulrich Drepper.
3179 * Implement second fallback mode for optimized DNS lookup for even more
3180 broken environments. If two requests from the same port are not
3181 handled correctly close the socket and open a new one before sending
3182 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3183 can be used to select this mode right away, instead of rediscovering the
3184 necessity is every process again.
3185 Implemented by Ulrich Drepper.
3187 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3188 Implemented by Adam Tkac.
3190 * Optimized iconv conversions for S390x.
3191 Implemented by Andreas Krebbel.
3193 * Using condvars with PI mutexes is now more efficient due to kernel
3194 support for requeueing to PI futexes. NPTL support added for x86-64.
3195 Implemented by Ulrich Drepper.
3197 * New locales: ps_AF, my_MM
3201 * The following bugs are resolved with this release:
3203 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3204 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3205 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3206 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3207 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3209 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3210 Implemented by Ulrich Drepper.
3212 * New Linux interfaces: accept4, fallocate, fallocate64.
3213 Implemented by Ulrich Drepper.
3215 * Correct declarations of string function when used in C++ code. This
3216 could lead to compile errors for invalid C++ code.
3218 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3221 * New POSIX 2008 interface: psiginfo
3222 Implemented by Ulrich Drepper.
3224 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3225 Implemented by Ulrich Drepper.
3227 * Support for selecting between multiple function definitions at runtime
3228 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3230 * The libcrypt library can now use the hash function implementations in
3231 NSS. Implemented by Ulrich Drepper.
3233 * The malloc implementation can be compiled to be less memory efficient
3234 but higher performing in multi-threaded programs.
3235 Implemented by Ulrich Drepper.
3237 * New locales: nan_TW@latin, ks_IN
3239 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3240 Implemented by Ulrich Drepper.
3242 * Extended printf hook support. It is possible to use user-defined types
3243 and extend existing format specifiers.
3244 Implemented by Ulrich Drepper.
3246 * Handling for group shadow files has been added.
3247 Implemented by Ulrich Drepper.
3249 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3250 servers (the case, e.g., for some people using the built-in DNS
3251 server in ADSL modems/routers). There is a once-per-process timeout
3252 in case of a broken server. To avoid it, users can run nscd or put
3253 'options single-request' in /etc/resolv.conf.
3254 Implemented by Ulrich Drepper.
3258 * The following bugs are resolved with this release:
3260 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3261 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3262 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3263 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3264 6968, 6974, 6980, 6995, 7008, 7009, 7029
3266 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3267 up at the same time. Implemented by Ulrich Drepper.
3269 * TLS descriptors for LD and GD on x86 and x86-64.
3270 Implemented by Alexandre Oliva.
3272 * getaddrinfo now handles DCCP and UDPlite.
3273 Implemented by Ulrich Drepper.
3275 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3276 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3277 Implemented by Ulrich Drepper.
3279 * New implementation of memmem, strstr, and strcasestr which is O(n).
3280 Implemented by Eric Blake.
3282 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3284 * Implement "e" option for popen to open file descriptor with the
3285 close-on-exec flag set. Implemented by Ulrich Drepper.
3287 * Implement "b" mode for fmemopen. In this mode writes writes don't
3288 implicitly add a NUL byte and seeks from the end of the buffer really
3289 use the buffer end, not the string length as the basis.
3290 Implemented by Ulrich Drepper.
3292 * Many functions, exported and internal, now atomically set the close-on-exec
3293 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3295 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3297 Implemented by Pravin Satpute.
3299 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3303 * The following bugs are resolved with this release:
3305 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3306 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3307 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3308 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3309 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3310 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3311 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3313 * New locales: bo_CN, bo_IN, shs_CA.
3315 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3317 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3318 Implemented by Pravin Satpute.
3320 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3322 * nscd caches now all timeouts for DNS entries
3323 Implemented by Ulrich Drepper.
3325 * nscd is more efficient and wakes up less often.
3326 Implemented by Ulrich Drepper.
3328 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3329 vdprintf, and obstack_vprintf.
3330 Implemented by Jakub Jelinek.
3332 * Faster memset for x86-64.
3333 Implemented by Harsha Jagasia and H.J. Lu.
3335 * Faster memcpy on x86.
3336 Implemented by Ulrich Drepper.
3338 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3339 Implemented by Ulrich Drepper.
3341 * Faster sqrt and sqrtf implemention for some PPC variants.
3342 Implemented by Steven Munroe.
3346 * The following bugs are resolved with this release:
3348 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3349 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3350 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3351 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3352 5071, 5103, 5104, 5112, 5113, 5184, 5186
3354 * More checking functions: fread, fread_unlocked, open*, mq_open.
3355 Implemented by Jakub Jelinek and Ulrich Drepper.
3357 * Extend fortification to C++. Implemented by Jakub Jelinek.
3359 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3360 by not recognizing 'a' as a modifier when those specs are requested.
3361 Implemented by Jakub Jelinek.
3363 * PPC optimizations to math and string functions.
3364 Implemented by Steven Munroe.
3366 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3367 options to be passed. Implemented by Ulrich Drepper.
3369 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3371 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3372 Implemented by Ulrich Drepper.
3374 * Handle private futexes in the NPTL implementation.
3375 Implemented by Jakub Jelinek and Ulrich Drepper.
3377 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3378 Implemented by Roland McGrath and Ulrich Drepper.
3380 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3382 * SHA-256 and SHA-512 based password encryption.
3383 Implemented by Ulrich Drepper.
3385 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3386 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3389 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3390 Implemented by Ulrich Drepper.
3394 * The following bugs are resolved with this release:
3396 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3397 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3398 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3399 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3400 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3401 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3402 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3405 * New Linux interfaces: epoll_pwait, sched_getcpu.
3407 * New generic interfaces: strerror_l.
3409 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3413 * The following bugs are resolved with this release:
3415 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3416 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3417 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3418 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3419 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3420 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3421 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3422 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3423 3225, 3251, 3252, 3253, 3273
3425 * For Linux, the sorting of addresses returned by getaddrinfo now also
3426 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3427 Implemented by Ulrich Drepper.
3429 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3430 Implemented by Ulrich Drepper.
3432 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3434 * New iconv module for MIK. Contributed by Alexander Shopov.
3436 * For sites with broken group and/or passwd database, the auto-propagate
3437 option of nscd can prevent creating ID lookup entries from the results
3438 of a name lookup and vice versa. This usually is no problem but some
3439 site might have problems with the default behavior.
3440 Implemented by Ulrich Drepper.
3442 * Iterating over entire database in NIS can be slow. With the
3443 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3444 to trade time for memory. The entire database will be read at once.
3445 Implemented by Ulrich Drepper.
3447 * The interfaces introduced in RFC 3542 have been implemented by
3450 * Support for the new ELF hash table format was added by Ulrich Drepper.
3452 * Support for priority inheritance mutexes added by Jakub Jelinek and
3455 * Support for priority protected mutexes added by Jakub Jelinek.
3457 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3458 or_IN, csb_PL, fy_NL, sr_ME.
3462 * More overflow detection functions.
3464 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3465 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3467 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3468 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3469 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3470 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3471 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3472 by Masahide Washizawa.
3474 * It is now possible to install audit libraries for the dynamic linker using
3475 LD_AUDIT. Implemented by Ulrich Drepper.
3477 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3478 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3479 For a libc and libpthread that works well on Linux 2.4 kernels, we
3480 recommend using the stable 2.3 branch.
3482 * The new function open_wmemstream acts like open_memstream,
3483 but uses a wchar_t wide character buffer.
3485 * The new function ppoll is an improved version of poll, similar to pselect.
3487 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3488 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3489 renameat, unlinkat, linkat, symlinkat, readlinkat.
3491 * New Linux kernel system calls: unshare,
3492 inotify_init, inotify_add_watch, inotify_rm_watch.
3494 * The euidaccess function is now also known by the alias eaccess,
3495 for compatibility with some other systems.
3497 * Timezone data updated to 2006b version.
3501 * The following bugs are resolved with this release:
3503 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3504 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3505 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3506 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3507 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3508 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3510 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3512 * As of this release, GCC 4 can be used to compile the C Library.
3514 * Timezone data updated to 2005m version.
3518 * The following bugs are resolved with this release:
3520 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3521 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3522 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3523 777, 787, 821, 822, 823, 825
3525 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3529 * Support for RFC 3678. Real implementations exist only for Linux so far.
3530 Implemented by Ulrich Drepper.
3532 * nscd can now cache entries persistently. Expiring entries are reloaded.
3533 For speedups the cache can be shared in memory with client processes.
3534 Implemented by Ulrich Drepper.
3536 * nscd can now perform SELinux checks.
3537 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3539 * getaddrinfo queries are now cached. Canonical name lookup is performed
3541 Implemented by Ulrich Drepper.
3543 * The nothrow function attribute is used when headers are used by gcc when
3544 compiling C code. This can avoid significant amounts of exception
3547 * The malloc functions perform more error checking and are stricter when
3548 it comes to reacting on errors. The default action is to terminate
3549 the process after showing an error message. Implemented by Ulrich Drepper.
3551 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3552 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3554 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3555 namespaces using the new function dlmopen(). This feature is of course,
3556 like most other dynamic loading functionality, not available in statically
3557 linked applications. Implemented by Ulrich Drepper.
3559 * Low-overhead boundary checking variants of string and some stdio functions
3560 were added. These are to be used in conjunction with a gcc patch by
3561 Jakub Jelinek which adds calls to these functions if possible.
3562 Implemented by Jakub Jelinek and Ulrich Drepper.
3564 * Old code for several operating systems and machine architectures that
3565 have not been in working condition in a long time have been removed from
3566 the main source tree maintained by the GNU C Library's maintainers.
3567 These files are now reside in the separate `ports' source module
3568 that is usable as an add-on when building the library.
3572 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3573 interrogate the dynamic linker, compatible with the Solaris interface.
3575 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3576 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3578 * getifaddrs now uses the netlink interface on Linux to get its information.
3579 Implemented by Thorsten Kukuk.
3581 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3582 Implemented by Ulrich Drepper.
3584 * support for non-executable stacks on x86 has been added. Changes mostly
3587 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3590 * getaddrinfo now performs destination address selection according to
3595 * Thread-safe interfaces for many functions that access locale data
3596 were added in version 2.3, but these features were omitted from NEWS.
3597 Many functions have variants with an `_l' suffix that take a `locale_t'
3598 object as a parameter rather than consulting the current locale.
3599 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3600 create and maintain `locale_t' objects. Additionally, the new function
3601 `uselocale' sets "the current locale" (as used by functions not so
3602 parameterized) set for an individual thread. These features were added
3603 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3605 * The functions getresuid, getresgid, setresuid, and setresgid, which
3606 have long been available on Linux, are now declared in <unistd.h>
3607 and are now also available on the Hurd.
3609 * ELF thread-local storage support (TLS) now works on x86-64.
3611 * The new dynamic string token $LIB is expanded in shared library names.
3612 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3614 * Aldy Hernandez contributed complete software floating point support for
3615 PowerPC machines with no FPU.
3617 * fexecve is implemented on Linux.
3619 * The `btowc' function should work at least twice as fast due to
3620 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3622 * With approriate thread add-ons cancelable functions are now implemented
3623 in libc.so as well. No need to call the function in libpthread. This
3624 change allowed to finally disable the incorrect and expensive handling
3625 of weak definition in ld.so.
3627 * Yet more PLT entries in libc.so have been removed. We finally arrived
3628 at the bare minimum. Startup times improved appropriately.
3630 * Support for the new Linux/x86 system call interface was added. The
3631 AT_SYSINFO auxiliary vector entry is recognized and handled.
3635 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3638 * iconv (the program and the interface) now accepts empty names (excluding
3639 options like //TRANSLIT) to mean "use charset of current locale".
3641 * localedef can now transliterate characters in strings which are not in
3642 the provided charmap. The information from the input locale is used.
3644 * Prelinking support was added for ELF targets. This requires additional
3645 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3647 * Read-only stdio streams now use mmap to speed up operation by eliminating
3648 copying and buffer underflows. To use add 'm' to the mode string of
3649 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3651 * The malloc functions were completely rewritten by Wolfram Gloger based
3652 on Doug Lea's malloc-2.7.0.c.
3654 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3655 implementation of regex.
3657 * Bruno Haible upgraded the iconv and locale implementation to support
3660 * Contents of the LC_* and LANG environment variables in the CEN style are
3661 not recognized anymore. It never was used. Change by Ulrich Drepper.
3663 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3664 thread-local storage (TLS) ABI on some platforms.
3665 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3667 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3668 EUC-JISX0213, and TSCII.
3670 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3671 BSD-compatible interface for getting all network interface addresses.
3672 Implementation for IPv4 by Roland McGrath.
3674 * Loading of locale data is faster due to the introduction of a locale
3675 archive. Implemented by Roland McGrath and Ulrich Drepper.
3677 * Startup times are significantly reduced by not using exported functions
3678 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3681 * Steven Munroe contributed a port to PowerPC64/Linux.
3685 * The Hurd now uses the GNU libio implementation of stdio rather than the
3686 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3688 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3689 with the same behavior as the Linux system call of the same name.
3693 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3694 128-bit long double format.
3696 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3697 IBM1161, and IBM1162 charsets.
3699 * Andreas Jaeger contributed a port to x86-64/Linux.
3701 * Peter Bruin contributed a port to PowerPC/Hurd.
3703 * libc functions using I/O streams now can handle wide-oriented streams
3706 * optimizations in the dynamic linker. Binaries created by recent binutils
3707 versions start up quicker due to reduced time spent on relocations.
3709 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3713 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3714 asinh, atanh, j0 for the 128-bit long double format.
3716 * Bruno Haible updated all the code handling Unicode in some form to
3717 support Unicode 3.1.
3719 * Speed of regex for single-byte locales is back to previous levels.
3720 Patch by Isamu Hasegawa.
3722 * Alpha, SPARC, and IA-64 now also using floating stacks.
3724 * Startup time of internationalized applications greatly improved through
3725 iconv cache. Use iconvconfig to generate the cache file.
3726 Contributed by Ulrich Drepper.
3728 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3729 severe performance problems. Patch by David Mosberger.
3731 * The Hurd port got a lot more functionality like AIO, various stdio
3732 extensions, etc. Mainly done by Roland McGrath.
3734 * mtrace can now lookup symbols in shared libraries.
3738 * Intel's IA-64 math library is largely integrated. It provides fast and
3739 accurate implementatations for most basic and standard math functions
3740 in float, double, and long double format.
3742 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3743 and asin for the 96-bit long double format and asin, log, tan for the
3744 128-bit long double format.
3746 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3747 The basic double functions exist today. Contributed by Abraham Ziv
3748 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3749 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3751 * An asynchronous name lookup library was added. The interface is designed
3752 after POSIX AIO. The proposal was circulated beforehand to get comments.
3753 No negative ones came in. Implemented by Ulrich Drepper.
3755 * Port to S390/64bit contributed by Martin Schwidefsky
3756 <schwidefsky@de.ibm.com>.
3758 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3759 of functions for Linux/IA-64.
3761 * The RPC code is now thread safe. Threads can now use the same service
3762 of different services at the same time. Patch by Eric Norum
3763 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3765 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3766 family of functions for Linux/S390.
3768 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3769 of functions for Linux/x86.
3771 * Port to Linux/CRIS contributed by Axis Communications.
3775 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3776 we can now check for namespace violations and missing declarations. The
3777 result is that almost all headers are now Unix-compliant (as defined in
3778 the upcoming XPG6). The negative side is that some programs might need
3779 corrections, too, if they depend on the incorrect form of the headers in
3780 previous versions which defined too many symbols and included too many
3783 * regex now handles multibyte character sets correctly.
3784 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3786 * iconv (the program) does now conform to the upcoming XPG6 and handles
3787 charmaps. Instead of the charset names the path of charmaps can be
3788 provided and the conversion happens based on this data.
3789 Contributed by Ulrich Drepper.
3791 * The locale program now provides detailed information about the installed
3796 only lists the names of the supported locales
3800 provides details such as country, language, and codeset name.
3801 Contributed by Ulrich Drepper.
3805 * The gencat program now parses the input file according to the charset
3806 selected by the LC_CTYPE category. This is important for stateful
3807 character sets. To make generating catalogs easier there is a way
3808 to overwrite the charset selected by the locale: before the first
3809 message or $ quote line the catalog can contain a line like
3811 $ codeset=ISO-8859-2
3813 to select the charset (ISO-8859-2 in this case).
3815 Implemented by Shinya Hanataka and Ulrich Drepper.
3817 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3818 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3819 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3821 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3822 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3824 * The LANGUAGE environment variable is now ignored unless the locale is
3825 changed from the default "C" locale.
3827 * The usual bug fixes.
3831 * Greg McGary added runtime support for bounds checking using gcc's
3832 new -fbounded-pointers option. ix86 target is complete. PowerPC
3835 * Thorsten Kukuk added secure mode support to nscd.
3837 * The Berkeley db libraries have been removed.
3839 Related, the nss_db module is now in a separate package since it
3840 obviously requires a database library being available.
3842 * Wide character I/O streams implemented by Ulrich Drepper.
3844 * Functions from the extended socket API added by Ulrich Drepper.
3846 * Functions feenableexcept and fedisableexcept to control the
3847 behaviour of individual exceptions have been added by Andreas Jaeger.
3849 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3851 * The resolver code has been updated from bind 8.2.3-T5B which supports
3852 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3855 This change could in some situations effect backward compatibility. Since
3856 now `_res' is a thread-local instead of a global variable, modifying it
3857 in one thread does not have any effect in other threads.
3859 The resolver library was also extended to allow IPv6 as the transport
3860 protocol for the requests. This work was done by Stig Venaas.
3862 * Compatibility code for K&R C compilers has been removed from the
3863 header files. A ISO C compiler is needed to use the library
3864 (conforming to either C89 or C99 standard).
3866 * Complete rewrite of the localedef program to support multibyte character
3867 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3868 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3869 Implemented by Ulrich Drepper.
3871 Bruno Haible significantly improved the generation and use of the data
3872 structures for the wide character tables.
3874 * Plural handling in gettext implemented by Ulrich Drepper.
3876 * The utmp daemon has been removed.
3878 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3880 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3883 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3885 * POSIX barriers implemented by Kaz Kylheku.
3887 * POSIX spawn function family implemented by Ulrich Drepper.
3889 * POSIX spinlocks are now available.
3891 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3893 * the configure option --enable-kernel=X.Y.Z allows to strip out
3894 compatibility for kernel versions before X.Y.Z. This is currently only
3895 implemented for Linux.
3897 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3898 field sin6_scope_id. This means that all programs using IPv6 should be
3899 recompiled. Don't expect binary compatibility with previous glibc
3902 * various conversion modules for IBM character sets contributed by
3905 * IA-64 port by Jes Sorensen and HJ Lu.
3919 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3921 * New xdr functions are added; some rpc functions are now 64bit clean.
3923 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3925 * Fixed known incompatibilities with glibc 2.0.
3927 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3929 * Optimized a number of functions (especially the ELF dynamic loader).
3931 * Update timezone data files.
3933 * lots of charmaps corrections
3935 * some new locale definitions and charmaps
3940 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3941 conform to POSIX member type requirements. Symbol versions have been
3942 adjusted as needed within the library, and for direct use by applications,
3943 but there is potential for problems if third-party libraries use
3944 struct timeval as part of their interface. This does not present
3945 a problem for X and other "essential" system libraries.
3947 * An additional locale model to support C++ Standard Library locale
3948 model and probably more was implemented by Ulrich Drepper.
3950 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3953 * Miles Bader provided the `argp' function family to support hierachical
3954 command line argument parsing, layered on top of getopt.
3956 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3958 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3961 * scanf recognizes the %a and %A format for scanning floating point numbers.
3963 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3964 information and interfaces for the available integer types.
3966 * about 130 new math functions were added to implement the ISO C9x math
3969 * the new header <complex.h> contains definitions of the complex math
3970 functions from ISO C 9X.
3972 * the new header <tgmath.h> defines generic macros to use complex or
3973 real valued functions.
3975 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3977 * Andreas Jaeger provided a test suite for the math library.
3979 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3981 * Ulrich Drepper added character set conversion functions (iconv).
3983 * Optimized string functions have been added.
3985 * The localedata addon is now part of glibc.
3987 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3989 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3990 daemon for NSS (nscd).
3992 Missing a better place here are some numbers on improvements. Under
3993 Linux 2.1.125 un-tar-ing the kernel sources takes
3997 using local files 12.19s 6.88s 22.91s
3999 using NIS 13.92s 8.91s 26.34s
4001 using NIS & nscd 10.37s 7.34s 25.30s
4003 using NIS+ 27.57s 30.37s 640.46s
4005 using NIS+ & nscd 10.25s 7.83s 26.51s
4007 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4009 Keep in mind that non-namelookup related operations dominate above times.
4010 It was just a common complain that using NIS+ unpacking the kernel is
4013 [1] The old nscd implementation is not available anymore since it was
4014 distributed with glibc up to version 2.0.98 and thus is now replaced.
4016 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4018 * Mark Kettenis provided a Hesiod NSS module.
4020 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4021 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4023 * rcmd can now handle netgroups (Dick Streefland).
4025 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4028 * Support for the IPv6 protocol has been added to the socket API, as per the
4029 latest draft standards.
4031 * Support for Linux 2.2 has been added.
4033 * Interface changes relative to the latest 2.0.x release:
4034 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4035 addseverity NEW: Unix98
4036 alphasort64 NEW: LFS
4037 argp_err_exit_status NEW: argp, GNU ext
4038 argp_error NEW: argp, GNU ext
4039 argp_failure NEW: argp, GNU ext
4040 argp_help NEW: argp, GNU ext
4041 argp_parse NEW: argp, GNU ext
4042 argp_program_bug_address NEW: argp, GNU ext
4043 argp_program_version NEW: argp, GNU ext
4044 argp_program_version_hook NEW: argp, GNU ext
4045 argp_state_help NEW: argp, GNU ext
4046 argp_usage NEW: argp, GNU ext
4047 authdes_create NEW: Secure RPC
4048 authdes_getucred NEW: Secure RPC
4049 authdes_pk_create NEW: Secure RPC
4050 backtrace NEW: GNU ext.
4051 backtrace_symbols NEW: GNU ext.
4052 backtrace_symbols_fd NEW: GNU ext.
4054 cacosf NEW: ISO C 9x
4055 cacosh NEW: ISO C 9x
4056 cacoshf NEW: ISO C 9x
4057 cacoshl NEW: ISO C 9x
4058 cacosl NEW: ISO C 9x
4065 casinf NEW: ISO C 9x
4066 casinh NEW: ISO C 9x
4067 casinhf NEW: ISO C 9x
4068 casinhl NEW: ISO C 9x
4069 casinl NEW: ISO C 9x
4071 catanf NEW: ISO C 9x
4072 catanh NEW: ISO C 9x
4073 catanhf NEW: ISO C 9x
4074 catanhl NEW: ISO C 9x
4075 catanl NEW: ISO C 9x
4076 cbc_crypt NEW: Secure RPC
4080 ccoshf NEW: ISO C 9x
4081 ccoshl NEW: ISO C 9x
4087 cimagf NEW: ISO C 9x
4088 cimagl NEW: ISO C 9x
4089 clearerr_locked REMOVED
4090 clntunix_create NEW: sunrpc ext
4092 clog10 NEW: ISO C 9x
4093 clog10f NEW: ISO C 9x
4094 clog10l NEW: ISO C 9x
4104 cprojf NEW: ISO C 9x
4105 cprojl NEW: ISO C 9x
4107 crealf NEW: ISO C 9x
4108 creall NEW: ISO C 9x
4113 csinhf NEW: ISO C 9x
4114 csinhl NEW: ISO C 9x
4117 csqrtf NEW: ISO C 9x
4118 csqrtl NEW: ISO C 9x
4122 ctanhf NEW: ISO C 9x
4123 ctanhl NEW: ISO C 9x
4125 des_setparity NEW: Secure RPC
4126 ecb_crypt NEW: Secure RPC
4127 endutxent NEW: Unix98
4129 exp10f NEW: ISO C 9x
4130 exp10l NEW: ISO C 9x
4134 fattach NEW: STREAMS
4135 fdetach NEW: STREAMS
4139 feclearexcept NEW: ISO C 9x
4140 fegetenv NEW: ISO C 9x
4141 fegetexceptflag NEW: ISO C 9x
4142 fegetround NEW: ISO C 9x
4143 feholdexcept NEW: ISO C 9x
4145 feraiseexcept NEW: ISO C 9x
4146 ferror_locked REMOVED
4147 fesetenv NEW: ISO C 9x
4148 fesetexceptflag NEW: ISO C 9x
4149 fesetround NEW: ISO C 9x
4150 fetestexcept NEW: ISO C 9x
4151 feupdateenv NEW: ISO C 9x
4152 fflush_locked REMOVED
4156 fgets_unlocked NEW: GNU ext.
4157 fileno_locked REMOVED
4169 fputc_locked REMOVED
4170 fputs_unlocked NEW: GNU ext.
4171 fread_unlocked NEW: GNU ext.
4176 fstatvfs NEW: Unix98
4180 ftruncate64 NEW: LFS
4182 fwrite_unlocked NEW: GNU ext.
4183 gai_strerror NEW: IPv6
4187 getchar_locked REMOVED
4189 getdate_err NEW: Unix98
4190 getdate_r NEW: GNU ext.
4192 getnameinfo NEW: IPv6
4193 getnetname NEW: Secure RPC
4194 getpmsg NEW: STREAMS
4195 getpt NEW: Unix98 PTY
4196 getrlimit64 NEW: LFS
4197 getutxent NEW: Unix98
4198 getutxid NEW: Unix98
4199 getutxline NEW: Unix98
4200 glob64 NEW: GNU ext.
4201 globfree64 NEW: GNU ext.
4202 gnu_get_libc_release NEW: GNU ext.
4203 gnu_get_libc_version NEW: GNU ext.
4204 grantpt NEW: Unix98 PTY
4205 host2netname NEW: Secure RPC
4207 iconv_close NEW: iconv
4208 iconv_open NEW: iconv
4209 if_freenameindex NEW: IPv6
4210 if_indextoname NEW: IPv6
4211 if_nameindex NEW: IPv6
4212 if_nametoindex NEW: IPv6
4213 in6addr_any NEW: IPv6
4214 in6addr_loopback NEW: IPv6
4215 inet6_isipv4mapped NEW: IPv6
4216 isastream NEW: STREAMS
4217 iswblank NEW: GNU ext.
4218 key_decryptsession NEW: Secure RPC
4219 key_decryptsession_pk NEW: Secure RPC
4220 key_encryptsession NEW: Secure RPC
4221 key_encryptsession_pk NEW: Secure RPC
4222 key_gendes NEW: Secure RPC
4223 key_get_conv NEW: Secure RPC
4224 key_secretkey_is_set NEW: Secure RPC
4225 key_setnet NEW: Secure RPC
4226 key_setsecret NEW: Secure RPC
4227 llrint NEW: ISO C 9x
4228 llrintf NEW: ISO C 9x
4229 llrintl NEW: ISO C 9x
4230 llround NEW: ISO C 9x
4231 llroundf NEW: ISO C 9x
4232 llroundl NEW: ISO C 9x
4237 lrintf NEW: ISO C 9x
4238 lrintl NEW: ISO C 9x
4239 lround NEW: ISO C 9x
4240 lroundf NEW: ISO C 9x
4241 lroundl NEW: ISO C 9x
4243 makecontext NEW: Unix98
4244 mempcpy NEW: GNU ext.
4247 modify_ldt NEW: kernel
4251 nearbyint NEW: ISO C 9x
4252 nearbyintf NEW: ISO C 9x
4253 nearbyintl NEW: ISO C 9x
4254 netname2host NEW: Secure RPC
4255 netname2user NEW: Secure RPC
4256 nexttoward NEW: ISO C 9x
4257 nexttowardf NEW: ISO C 9x
4258 nexttowardl NEW: ISO C 9x
4262 passwd2des NEW: Secure RPC
4264 pow10f NEW: GNU ext.
4265 pow10l NEW: GNU ext.
4268 printf_size NEW: GNU ext.
4269 printf_size_info NEW: GNU ext.
4270 profil_counter REMOVED
4271 pthread_mutexattr_getkind_np REPLACED
4272 pthread_mutexattr_setkind_np REPLACED
4273 ptsname NEW: Unix98 PTY
4274 ptsname_r NEW: Unix98 PTY
4276 putchar_locked REMOVED
4277 putgrent NEW: GNU ext.
4279 putpmsg NEW: STREAMS
4280 pututxline NEW: Unix98
4284 readdir64_r NEW: LFS
4285 remquo NEW: ISO C 9x
4286 remquof NEW: ISO C 9x
4287 remquol NEW: ISO C 9x
4289 roundf NEW: ISO C 9x
4290 roundl NEW: ISO C 9x
4292 scalbln NEW: ISO C 9x
4293 scalblnf NEW: ISO C 9x
4294 scalblnl NEW: ISO C 9x
4296 sendfile NEW: kernel
4297 setcontext NEW: Unix98
4298 setrlimit64 NEW: LFS
4299 setutxent NEW: Unix98
4301 sigignore NEW: Unix98
4302 sigqueue NEW: POSIX.1b
4303 sigrelse NEW: Unix98
4304 sigset NEW: POSIX.1b
4305 sigtimedwait NEW: POSIX.1b
4306 sigwaitinfo NEW: POSIX.1b
4307 sincos NEW: GNU ext.
4308 sincosf NEW: GNU ext.
4309 sincosl NEW: GNU ext.
4313 strcasestr NEW: GNU ext.
4314 strtoimax NEW: ISO C 9x
4315 strtoumax NEW: ISO C 9x
4316 strverscmp NEW: GNU ext.
4317 svcauthdes_stats NEW: Secure RPC
4318 svcunix_create NEW: sunrpc etx
4319 svcunixfd_create NEW: sunrpc ext
4320 swapcontext NEW: Unix98
4321 tcgetsid NEW: Unix98 PTY
4322 tdestroy NEW: GNU ext.
4323 tgamma NEW: ISO C 9x
4324 tgammaf NEW: ISO C 9x
4325 tgammal NEW: ISO C 9x
4329 truncf NEW: ISO C 9x
4330 truncl NEW: ISO C 9x
4332 unlockpt NEW: Unix98 PTY
4333 updwtmpx NEW: Unix98
4334 user2netname NEW: Secure RPC
4335 utmpxname NEW: Unix98
4336 versionsort NEW: GNU ext.
4337 versionsort64 NEW: GNU ext.
4339 wcscasecmp NEW: GNU ext.
4340 wcsncasecmp NEW: GNU ext.
4341 wcsnlen NEW: GNU ext.
4342 wcstoimax NEW: ISO C 9x
4343 wcstoll NEW: ISO C 9x
4344 wcstoull NEW: ISO C 9x
4345 wcstoumax NEW: ISO C 9x
4347 wordexp NEW: POSIX.2
4348 wordfree NEW: POSIX.2
4349 write_profiling REMOVED
4350 xdecrypt NEW: Secure RPC
4351 xdr_authdes_cred NEW: Secure RPC
4352 xdr_authdes_verf NEW: Secure RPC
4353 xdr_cryptkeyarg NEW: Secure RPC
4354 xdr_cryptkeyarg2 NEW: Secure RPC
4355 xdr_cryptkeyres NEW: Secure RPC
4356 xdr_getcredres NEW: Secure RPC
4357 xdr_key_netstarg NEW: Secure RPC
4358 xdr_key_netstres NEW: Secure RPC
4359 xdr_keybuf NEW: Secure RPC
4360 xdr_keystatus NEW: Secure RPC
4361 xdr_netnamestr NEW: Secure RPC
4362 xdr_sizeof NEW: Secure RPC
4363 xdr_unixcred NEW: sunrpc ext
4364 xencrypt NEW: Secure RPC
4365 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4376 * inet_ntoa is thread-safe
4378 * updwtmp is moved from libutil to libc
4380 * rewrite of cbrt function
4382 * update of timezone data
4396 * add atoll function
4398 * fix complex problems in Berkeley DB code
4400 * fix math functions
4404 * fixed lots of header problems (especially Linux/GNU specific)
4406 * dynamic loader preserves all registers
4408 * Roland McGrath provided support for handling of auxiliary objects in
4409 the ELF dynamic loader.
4411 * support for parallel builds is improved
4415 * GNU extensions are no longer declared by default. To enable them you
4416 must define the macro `_GNU_SOURCE' in your program or compile with
4419 * The library has changed from using GNU ld symbol aliases to using weak
4420 symbols where available. The ELF object file format supports weak
4421 symbols; GNU ld also supports weak symbols in the a.out format. (There
4422 is also now support for other GNU ld extensions in ELF. Use the
4423 `--with-elf' option to configure to indicate you have ELF, and
4424 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4425 of many files which contained only symbol aliases, reducing the size of
4426 the source and the compiled library; many other files were renamed to
4427 less cryptic names previously occupied by the symbol alias files.
4428 There is a new header file <elf.h> for programs which operate on
4429 files in the ELF format.
4431 * Converted to Autoconf version 2, so `configure' has more options.
4432 Run `configure --help' to see the details.
4434 * The library can now be configured to build profiling, highly-optimized
4435 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4436 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4437 `configure' enable building these extra libraries. The shared library is
4438 built by default when using both ELF and GNU ld. When shared libraries
4439 are enabled, the new library `-ldl' is available for arbitrary run-time
4440 loading of shared objects; its interface is defined in <dlfcn.h>. The
4441 new header file <link.h> gives access to the internals of the run-time
4442 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4443 application of same name on other systems and it provides information
4444 about dynamically linked binaries.
4446 * The C library now provides the run-time support code for profiling
4447 executables compiled with `-pg'. Programs can control the profiling code
4448 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4449 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4450 the support for this file format was contributed by David Mosberger-Tang.
4452 * The math code has been replaced with a math library based on fdlibm from
4453 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4454 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4455 functions. The math functions now reside in a separate library, so
4456 programs using them will need to use `-lm' their linking commands.
4458 * John C. Bowman contributed optimized ix87 assembler inline functions.
4460 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4461 mechanism similar to that found in Solaris 2. This is now used for the
4462 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4463 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4464 file controls what services are used for each individual database. This
4465 works by loading shared libraries with names specified in `nsswitch.conf',
4466 so service modules can be changed or added at any time without even
4467 relinking any program. Currently there are the file, db, and NIS based
4468 NSS services available.
4470 * The new functions `strtoq' and `strtouq' parse integer values from
4471 strings, like `strtol' and `strtoul', but they return `long long int' and
4472 `unsigned long long int' values, respectively (64-bit quantities).
4474 * The new functions `strtof' and `strtold' parse floating-point values from
4475 strings, like `strtod', but they return `float' and `long double' values,
4476 respectively (on some machines `double' and `long double' are the same).
4478 * Ulrich Drepper has contributed new implementations of the floating-point
4479 printing and reading code used in the `printf' family of functions and
4480 `strtod', `strtof', and `strtold'. These new functions are perfectly
4481 accurate, and much faster than the old ones.
4483 * The implementation of the POSIX locale model was completely rewritten by
4484 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4485 to compile the POSIX locale definition.
4487 * The former dummy implementations of the strcoll and strxfrm function are
4488 now replaced by fully functional code contributed by Ulrich Drepper. The
4489 collation information comes from the POSIX locale definitions.
4491 * The new header <langinfo.h> defines an interface for accessing
4492 various locale-dependent data (using the locale chosen with `setlocale').
4494 * Ulrich Drepper has contributed a new suite of functions for operation on
4495 wide-character and multibyte-character strings, in <wchar.h>;
4496 and classification and case conversion of wide characters, in <wctype.h>.
4497 These new functions are conforming to the ISO C, Amendement 1 specification.
4499 * There is now a second implementation of the standard I/O library available.
4500 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4501 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4503 * You can now use positional parameter specifications in format strings
4504 for the `printf' and `scanf' families of functions. For example,
4505 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4506 ``Number 6, Mr Jones''. This is mainly useful when providing different
4507 format strings for different languages, whose grammars may dictate
4508 different orderings of the values being printed. To support this
4509 feature, the interface for `register_printf_handler' has changed; see
4510 the header file <printf.h> for details.
4512 * The `printf' and `scanf' families of functions now understand a new
4513 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4514 to group numbers as indicated by the locale; for `scanf' and friends, this
4515 says to accept as valid only a number with all the proper grouping
4516 separators in the right places. In the default "C" locale, numbers are
4517 not grouped; but locales for specific countries will define the usual
4518 conventions (i.e. separate thousands with `,' in the US locale).
4520 * The pgrp functions have been regularized, slightly incompatibly but much
4521 less confusingly. The core functions are now `getpgid' and `setpgid',
4522 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4523 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4524 provided for compatibility. There is no longer an incompatible `getpgrp'
4525 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4527 * The new header file <fts.h> and suite of functions simplify programs that
4528 operate on directory trees. This code comes from 4.4 BSD.
4530 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4531 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4532 NSS scheme used in glibc.
4534 * There is a new malloc debugging hook `__memalign_hook'.
4536 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4537 `unsigned int' in <sys/types.h>. These are for compatibility only and
4538 their use is discouraged.
4540 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4541 done differently, so that it works even without GNU ld.
4543 * New function `euidaccess' checks allowed access to a file like `access',
4544 but using the effective IDs instead of the real IDs.
4546 * The time zone data files have been updated for the latest and greatest
4547 local time conventions of the countries of the world.
4549 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4552 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4553 for formatting floating-point numbers. They are provided only for
4554 compatibility; new programs should use `sprintf' instead. There are
4555 also equivalent function for the `long double' floating-point type and
4556 all functions also exist in a reentrant form.
4558 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4559 functions for maintaining the login-record files (primarily of use to
4560 system programs such as `login'), and convenient functions for
4561 allocating and initializing a pseudo-terminal (pty) device.
4563 * Ulrich Drepper has contributed new support for System V style
4564 shared memory and IPC on systems that support it.
4566 * Ulrich Drepper has contributed several miscellaneous new functions found
4567 in System V: The `hsearch' family of functions provide an effective
4568 implementation of hash tables; `a64l' and `l64a' provide a very simple
4569 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4572 * Ulrich Drepper has contributed new reentrant counterparts for the
4573 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4575 * Ulrich Drepper has contributed new, highly-optimized versions of several
4576 string functions for the i486/Pentium family of processors.
4578 * Ulrich Drepper has updated the Linux-specific code, based largely
4579 on work done in Hongjiu Lu's version of GNU libc for Linux.
4580 The GNU library now supports Linux versions 2.0.10 and later,
4581 using the ELF object file format (i[3456]86-*-linux).
4583 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4585 * David Mosberger-Tang and Richard Henderson have ported the C library
4586 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4587 dynamic linking support for ELF/Alpha.
4589 * Richard Henderson contributed several Alpha optimized assembler function
4590 for arithmetic and string handling.
4592 * Ulrich Drepper has contributed a new set of message catalog functions to
4593 support multiple languages using the <libintl.h> interface, for use with
4594 his new package GNU gettext. Translation volunteers have contributed
4595 catalogs of the library's messages in Spanish, German, and Korean.
4597 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4598 program and the `catgets' function for reading the catalog files it
4599 creates. (The <libintl.h> interface is preferred; we include the
4600 <nl_types.h> interface using `catgets' only for source compatibility with
4601 programs already written to use it.)
4603 * New header file <values.h> gives SVID-compatible names for <limits.h>
4606 * Various new macros, declarations, and small header files for compatibility
4609 * New function `group_member' is a convenient way to check if a process has
4610 a given effective group ID.
4612 * When using GCC 2.7 and later, the socket functions are now declared in a
4613 special way so that passing an argument of type `struct sockaddr_in *',
4614 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4615 `struct sockaddr *' type, does not generate a type-clash warning.
4617 * New function `error' declared in header file <error.h> is a convenient
4618 function for printing error messages and optionally exiting; this is the
4619 canonical function used in GNU programs. The new functions `err', `warn',
4620 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4621 doing the same thing.
4623 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4624 POSIX.2 `glob' function to do ~ and {...} expansion.
4626 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4627 `clearenv' which is used in POSIX.9 is also available.
4629 * New function `getsid' returns session ID number on systems that support it.
4631 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4632 files <db.h> and <mpool.h> provide a rich set of functions for several
4633 types of simple databases stored in memory and in files, and <ndbm.h> is
4634 an old `ndbm'-compatible interface using the `db' functions. Link with
4635 `-ldb' to get these functions.
4637 * New macro `strdupa' copies a string like `strdup', but uses local stack
4638 space from `alloca' instead of dynamic heap space from `malloc'.
4640 * New function `strnlen' is like `strlen' but searches only a given maximum
4641 number of characters for the null terminator. `stpncpy', `strndup' and
4642 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4645 * New function `statfs' in header <sys/statfs.h>.
4647 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4648 provide convenient functions for operating on blocks of null-terminated
4651 * A new suite of functions in <utmp.h> handle all the details of reading
4652 and writing the utmp file.
4654 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4657 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4658 completely POSIX compliant and also implemented the extended functionality
4659 to handle alternate digit representation and alternate era date formats.
4661 * Ulrich Drepper provided an implementation of the `strptime' function
4662 defined in XPG4.2 which transforms a string into a `struct tm' value.
4664 * Paul Eggert provided the tzselect shell script as part of the timezone
4665 code. The shell script makes it easy to select the correct timezone
4668 * The implementation of the malloc family of functions is completely replaced
4669 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4670 The implementation uses the mmap function (if available) and it is
4671 optimized for the use in multi threaded programs.
4673 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4674 functions. This new functionality is usable by specifying a special
4675 salt string and it is compatible with implementation on *BSD systems.
4677 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4678 `getsubopt' to handle second level command line options, `bsd_signal'
4679 to access BSD style `signal' functionality, the obsolete `regexp' style
4682 * the `lchown' function is available on system which support this
4685 * The implementation of the shadow password handling function was contributed
4688 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4690 * POSIX.1g support was added. The <sys/select.h> header is available,
4691 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4692 implementation of `getaddrinfo'.
4696 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4698 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4699 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4701 * The new function `daemon' from 4.4 BSD is useful for server programs that
4702 want to put themselves in the background.
4704 * Joel Sherrill has contributed support for several standalone boards that
4705 run without an operating system.
4707 * `printf', `scanf' and friends now accept a `q' type modifier for long
4708 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4710 * All of the code taken from BSD (notably most of the math and networking
4711 routines) has been updated from the BSD 4.4-Lite release.
4713 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4715 * The new functions `getdomainname' and `setdomainname' fetch or change the
4716 YP/NIS domain name. These are system calls which exist on systems which
4719 * The time zone data files have been updated for the latest international
4722 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4723 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4727 * The C library now includes support for Sun RPC, from Sun's free
4728 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4729 are included. (There is still no support for YP.)
4731 * Tom Quinn has contributed a port of the C library to SGI machines running
4732 Irix 4 (mips-sgi-irix4).
4734 * The new `lockf' function is a simplified interface to the locking
4735 facilities of `fcntl', included for compatibility.
4737 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4739 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4741 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4744 * The error message printed by `assert' for a failed assertion now includes
4745 the name of the program (if using GNU ld) and the name of the calling
4746 function (with versions of GCC that support this).
4748 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4750 * The library now includes the <sys/mman.h> header file and memory
4751 management functions `mmap', `munmap', `mprotect', `msync', and
4752 `madvise', on systems that support those facilities.
4754 * The interface for `mcheck' has changed slightly: the function called to
4755 abort the program when an allocation inconsistency is detected now takes
4756 an argument that indicates the type of failure. The new function
4757 `mprobe' lets you request a consistency check for a particular block at
4758 any time (checks are normally done only when you call `free' or `realloc'
4761 * It is now possible to easily cross-compile the C library, building on one
4762 system a library to run on another machine and/or operating system. All
4763 you need to do is set the variable `HOST_CC' in `configparms' to the
4764 native compiler for programs to run on the machine you are building on (a
4765 few generator programs are used on Unix systems); set `CC' to the
4768 * The new function `fexecve' (only implemented on the GNU system) executes
4769 a program file given a file descriptor already open on the file.
4773 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4774 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4776 * You can set the variable `libprefix' in `configparms' to specify a prefix
4777 to be prepended to installed library files; this makes it easy to install
4778 the GNU C library to be linked as `-lgnuc' or whatever.
4780 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4781 copies a limited number of characters from a string, and returns the
4782 address of the last character written.
4784 * You no longer need to check for whether the installed `stddef.h' is
4785 compatible with the GNU C library. configure now checks for you.
4787 * You can now define a per-stream `fileno' function to convert the
4788 stream's cookie into an integral file descriptor.
4790 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4791 allocates zero bytes of storage, and returns a unique pointer which
4792 you can pass to `realloc' or `free'. The behavior is undefined if
4793 you dereference this pointer.
4795 * The C library now runs on Sony NEWS m68k machines running either
4796 NewsOS 3 or NewsOS 4.
4798 * The new `syscall' function is a system-dependent primitive function
4799 for invoking system calls. It has the canonical behavior on Unix
4800 systems, including unreliable return values for some calls (such as
4801 `pipe', `fork' and `getppid').
4803 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4804 to `EAGAIN', which is the preferred name. On systems whose kernels
4805 use two distinct codes, the C library now translates EWOULDBLOCK to
4806 EAGAIN in every system call function.
4810 * The GNU C Library Reference Manual is now distributed with the library.
4811 `make dvi' will produce a DVI file of the printed manual.
4812 `make info' will produce Info files that you can read on line using C-h i
4813 in Emacs or the `info' program.
4814 Please send comments on the manual to bug-glibc-manual@gnu.org.
4816 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4818 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4820 * Jason Merrill has contributed a port to the Sequent Symmetry running
4821 Dynix version 3 (i386-sequent-dynix).
4823 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4824 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4826 * New function `memory_warnings' lets you arrange to get warnings when
4827 malloc is running out of memory to allocate, like Emacs gives you.
4829 * The C library now contains the relocating allocator used in Emacs 19 for
4830 its editing buffers. This allocator (ralloc) minimizes allocation
4831 overhead and fragmentation by moving allocated regions around whenever it
4832 needs to. You always refer to a ralloc'd region with a "handle" (a
4833 pointer to a pointer--an object of type `void **').
4835 * There is a new `printf' format: `%m' gives you the string corresponding
4836 to the error code in `errno'.
4838 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4839 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4840 pass, the `a' modifier says to fill in a `char **' you pass with a
4843 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4844 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4845 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4847 * `mkstemp' is a traditional Unix function to atomically create and open a
4848 uniquely-named temporary file.
4852 * The standard location for the file that says what the local timezone is
4853 has changed again. It is now `/usr/local/etc/localtime' (or more
4854 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4856 * The distribution no longer contains any files with names longer than 14
4859 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4860 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4862 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4866 * The configuration process has changed quite a bit. The `configure'
4867 script is now used just like the configuration scripts for other GNU
4868 packages. The `sysdeps' directory hierarchy is much rearranged.
4869 The file `INSTALL' explains the new scheme in detail.
4871 * The header files no longer need to be processed into ANSI C and
4872 traditional C versions. There is just one set of files to install, and
4873 it will work with ANSI or old C compilers (including `gcc -traditional').
4875 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4876 MIPS DECStation running Ultrix 4.
4878 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4879 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4880 made itself into a shared library.
4882 * Yet further improved support for the i386, running 4.3 BSD-like systems
4883 (such as Mach 3 with the Unix single-server), or System V.
4885 * New function `strncasecmp' to do case-insensitive string comparison
4886 with limited length.
4888 * New function `strsep' is a reentrant alternative to `strtok'.
4890 * New functions `scandir' and `alphasort' for searching directories.
4892 * New function `setenv' is a better interface to `putenv'.
4894 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4895 function for traversing a directory tree.
4897 * The GNU obstack package is now also part of the C library.
4898 The new function `open_obstack_stream' creates a stdio stream that
4899 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4900 formatted output directly to an obstack.
4902 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4903 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4905 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4907 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4908 things to your strings.
4910 * There are some new test programs: test-fseek, testmb, and testrand.
4912 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4913 These ports are not finished, but are a good starting place for really
4914 supporting those systems.
4916 * `/etc/localtime' is now the standard location for the file that says what
4917 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4918 This follows the general principle that `/etc' is the place for all local
4919 configuration files.
4921 * The C library header files now use `extern "C"' when used by the C++
4922 compiler, so the C library should now work with C++ code.
4924 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4925 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4928 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4929 function which uses the `merge sort' algorithm, and is said to be
4930 significantly faster than the old GNU `qsort' function. Merge sort is
4931 now the standard `qsort' function. The new algorithm can require a lot
4932 of temporary storage; so, the old sorting function is called when the
4933 required storage is not available.
4935 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4936 provides the Unix `crypt' function, plus some other entry points.
4938 * The code and header files taken from 4.4 BSD have been updated with the
4939 latest files released from Berkeley.
4941 ----------------------------------------------------------------------
4942 Copyright information:
4944 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4946 Permission is granted to anyone to make or distribute verbatim copies
4947 of this document as received, in any medium, provided that the
4948 copyright notice and this permission notice are preserved,
4949 thus giving the recipient permission to redistribute in turn.
4951 Permission is granted to distribute modified versions
4952 of this document, or of portions of it,
4953 under the above conditions, provided also that they
4954 carry prominent notices stating who last changed them.
4957 version-control: never