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
22 * The GNU C Library now has correct support for ABSOLUTE symbols
23 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
24 relocated incorrectly or in some cases discarded. The GNU linker can
25 make use of the newer semantics, but it must communicate it to the
26 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
27 field) to indicate such support is required.
29 * Unicode 11.0.0 Support: Character encoding, character type info, and
30 transliteration tables are all updated to Unicode 11.0.0, using
31 generator scripts contributed by Mike FABIAN (Red Hat).
33 * <math.h> functions that round their results to a narrower type are added
34 from TS 18661-1:2014 and TS 18661-3:2015:
36 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
39 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
42 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
45 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
48 * Nominative and genitive month names are now supported for the following
49 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Ossetian,
50 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
51 support abbreviated alternative month names: Catalan, Greek, and Kashubian.
53 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
54 include the support of the nominative and genitive month names.
56 * Building and running on GNU/Hurd systems now works without out-of-tree
59 * The renameat2 function has been added, a variant of the renameat function
60 which has a flags argument. If the flags are zero, the renameat2 function
61 acts like renameat. If the flag is not zero and there is no kernel
62 support for renameat2, the function will fail with an errno value of
63 EINVAL. This is different from the existing gnulib function renameatu,
64 which performs a plain rename operation in case of a RENAME_NOREPLACE
65 flags and a non-existing destination (and therefore has a race condition
66 that can clobber the destination inadvertently).
68 * The statx function has been added, a variant of the fstatat64
69 function with an additional flags argument. If there is no direct
70 kernel support for statx, glibc provides basic stat support based on
71 the fstatat64 function.
73 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
74 library if installed. libidn2 version 2.0.5 or later is recommended. If
75 libidn2 is not available, internationalized domain names are not encoded
76 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
77 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
78 with an encoding error.) Flags which used to change the IDN encoding and
79 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
80 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
81 deprecated. They no longer have any effect.
83 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
84 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
85 range of ELF gABI expressions including such constructs as
86 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
87 have been further restricted, and where in the past a dynamic string
88 token sequence may have been interpreted as a literal string it will
89 now cause a load failure. These load failures were always considered
90 unspecified behaviour from the perspective of the dynamic loader, and
91 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
92 results in a load failure now.
94 Deprecated and removed features, and other changes affecting compatibility:
96 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
97 installed. Software that was using either header should be updated to
98 use standard <stdio.h> interfaces instead.
100 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
101 This was never required by the C standard, and the macros just expanded
102 to call alternative names for the same functions. If you hoped getc and
103 putc would provide performance improvements over fgetc and fputc, instead
104 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
105 necessary, flockfile and funlockfile.
107 * All stdio functions now treat end-of-file as a sticky condition. If you
108 read from a file until EOF, and then the file is enlarged by another
109 process, you must call clearerr or another function with the same effect
110 (e.g. fseek, rewind) before you can read the additional data. This
111 corrects a longstanding C99 conformance bug. It is most likely to affect
112 programs that use stdio to read interactive input from a terminal.
115 * The macros 'major', 'minor', and 'makedev' are now only available from
116 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
117 headers that happen to include <sys/types.h>. These macros are rarely
118 used, not part of POSIX nor XSI, and their names frequently collide with
119 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
122 <sys/sysmacros.h> is a GNU extension. Portable programs that require
123 these macros should first include <sys/types.h>, and then include
124 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
126 * The tilegx*-*-linux-gnu configurations are no longer supported.
128 * The obsolete function ustat is no longer available to newly linked
129 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
130 function has been deprecated in favor of fstatfs and statfs.
132 * The obsolete function nfsservctl is no longer available to newly linked
133 binaries. This function was specific to systems using the Linux kernel
134 and could not usefully be used with the GNU C Library on systems with
135 version 3.1 or later of the Linux kernel.
137 * The obsolete function name llseek is no longer available to newly linked
138 binaries. This function was specific to systems using the Linux kernel
139 and was not declared in a header. Programs should use the lseek64 name
140 for this function instead.
142 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
143 getaddrinfo and getnameinfo functions have been deprecated. The behavior
144 previously selected by them is now always enabled.
146 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
147 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
148 restriction (rejecting '_' in host names, among other things) has been
149 removed, for increased compatibility with non-IDN name resolution.
151 * The fcntl function now have a Long File Support variant named fcntl64. It
152 is added to fix some Linux Open File Description (OFD) locks usage on non
153 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
154 fcntl and LFS support is handled transparently. Also for Linux, the OFD
155 locks act as a cancellation entrypoint.
157 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
158 ecb_crypt, and des_setparity are no longer available to newly linked
159 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
160 longer installed. These functions encrypted and decrypted data with the
161 DES block cipher, which is no longer considered secure. Software that
162 still uses these functions should switch to a modern cryptography library,
165 * Reflecting the removal of the encrypt and setkey functions above, the
166 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
167 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
170 * The obsolete function fcrypt is no longer available to newly linked
171 binaries. It was just another name for the standard function crypt,
172 and it has not appeared in any header file in many years.
174 * We have tentative plans to hand off maintenance of the passphrase-hashing
175 library, libcrypt, to a separate development project that will, we hope,
176 keep up better with new passphrase-hashing algorithms. We will continue
177 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
178 'crypt_r' should not need to change at all; however, distributions will
179 need to install <crypt.h> and libcrypt from a separate project.
181 In this release, if the configure option --disable-crypt is used, glibc
182 will not install <crypt.h> or libcrypt, making room for the separate
183 project's versions of these files. The plan is to make this the default
184 behavior in a future release.
186 Changes to build and runtime requirements:
188 GNU make 4.0 or later is now required to build glibc.
190 Security related changes:
192 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
193 been fixed by removing the glibc-internal IDNA implementation and using
194 the system-provided libidn2 library instead. Originally reported by Hanno
195 Böck and Christian Weisgerber.
197 CVE-2017-18269: An SSE2-based memmove implementation for the i386
198 architecture could corrupt memory. Reported by Max Horn.
200 CVE-2018-11236: Very long pathname arguments to realpath function could
201 result in an integer overflow and buffer overflow. Reported by Alexey
204 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
205 architecture could write beyond the target buffer, resulting in a buffer
206 overflow. Reported by Andreas Schwab.
208 The following bugs are resolved with this release:
210 [The release manager will add the list generated by
211 scripts/list-fixed-bugs.py just before the release.]
218 * The GNU C Library can now be compiled with support for building static
219 PIE executables (See --enable-static-pie in INSTALL). These static PIE
220 executables are like static executables but can be loaded at any address
221 and provide additional security hardening benefits at the cost of some
222 memory and performance. When the library is built with --enable-static-pie
223 the resulting libc.a is usable with GCC 8 and above to create static PIE
224 executables using the GCC option '-static-pie'. This feature is currently
225 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
226 aarch64 with binutils 2.30 or later.
228 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
229 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
232 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
234 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
236 * In order to support faster and safer process termination the malloc API
237 family of functions will no longer print a failure address and stack
238 backtrace after detecting heap corruption. The goal is to minimize the
239 amount of work done after corruption is detected and to avoid potential
240 security issues in continued process execution. Reducing shutdown time
241 leads to lower overall process restart latency, so there is benefit both
242 from a security and performance perspective.
244 * The abort function terminates the process immediately, without flushing
245 stdio streams. Previous glibc versions used to flush streams, resulting
246 in deadlocks and further data corruption. This change also affects
247 process aborts as the result of assertion failures.
249 * On platforms where long double has the IEEE binary128 format (aarch64,
250 alpha, mips64, riscv, s390 and sparc), the math library now implements
251 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
252 These are the same interfaces added in version 2.26 for some platforms where
253 this format is supported but is not the format of long double.
255 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
256 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
257 implements interfaces for that type, as defined by ISO/IEC TS
258 18661-3:2015. These are corresponding interfaces to those supported for
261 * The math library now implements interfaces for the _Float32, _Float64 and
262 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
263 corresponding interfaces to those supported for _Float128.
265 * glibc now implements the memfd_create and mlock2 functions on Linux.
267 * Support for memory protection keys was added. The <sys/mman.h> header now
268 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
271 * The copy_file_range function was added.
273 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
275 * The ldconfig utility now processes `include' directives using the C/POSIX
276 collation ordering. Previous glibc versions used locale-specific
277 ordering, the change might break systems that relied on that.
279 * Support for two grammatical forms of month names has been added.
280 In a call to strftime, the "%B" and "%b" format specifiers will now
281 produce the grammatical form required when the month is used as part
282 of a complete date. New "%OB" and "%Ob" specifiers produce the form
283 required when the month is named by itself. For instance, in Greek
284 and in many Slavic and Baltic languages, "%B" will produce the month
285 in genitive case, and "%OB" will produce the month in nominative case.
287 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
288 are all valid and will all accept any known form of month
289 name---standalone or complete, abbreviated or full. In a call to
290 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
291 the strings used by "%B" and "%b", respectively. New query
292 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
293 used by "%OB" and "%Ob", respectively.
295 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
296 define the strings for %OB and %Ob, respectively; these have the
297 same syntax as "mon" and "abmon". These arrays are optional; if they
298 are not provided then they have the same content as "mon" and "abmon",
301 These features are provided for locales which define "alt_mon" and/or
302 "ab_alt_mon" in their locale source data. This release includes such
303 alternative month name data for the following languages: Belarusian,
304 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
306 This feature is currently a GNU extension, but it is expected to
307 be added to the next revision of POSIX, and it is also already
308 available on some BSD-derived operating systems.
310 This feature will cause existing statically compiled applications
311 to fail to load locales and fall back to the builtin C/POSIX locales.
312 See notes below for other changes affecting compatibility.
314 * Support for the RISC-V ISA running on Linux has been added. This port
315 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
316 for the following ISA and ABI pairs:
322 Deprecated and removed features, and other changes affecting compatibility:
324 * Statically compiled applications attempting to load locales compiled for the
325 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
326 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
327 support for two grammatical forms of the month names, also extends the locale
328 data binary format. Static applications needing locale support must be
329 recompiled to match the runtime and data they are deployed with. In some
330 distributions there is an upgrade window where dynamically linked applications
331 may use a new library but the old locale data and also fall back to the
332 builtin C/POSIX locales; restarting the application process is sufficient to
335 * Support for statically linked applications which call dlopen is deprecated
336 and will be removed in a future version of glibc. Applications which call
337 dlopen need to be linked dynamically instead.
339 * Support for old programs which use internal stdio data structures and
340 functions is deprecated. This includes programs which use the C++ streams
341 provided by libstdc++ in GCC 2.95. Programs which use the internal
342 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
343 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
344 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
345 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
346 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
347 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
348 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
349 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
350 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
351 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
352 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
353 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
354 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
355 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
356 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
357 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
358 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
359 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
360 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
361 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
362 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
363 future version of glibc. Unlike other symbol removals, these old
364 applications will not be supported using compatibility symbols.
366 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
367 defined by <sys/ptrace.h>.
369 * libm no longer supports SVID error handling (calling a user-provided
370 matherr function on error) or the _LIB_VERSION variable to control error
371 handling. (SVID error handling and the _LIB_VERSION variable still work
372 for binaries linked against older versions of the GNU C Library.) The
373 libieee.a library is no longer provided. math.h no longer defines struct
374 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
375 TLOSS, PLOSS and HUGE.
377 * The libm functions pow10, pow10f and pow10l are no longer supported for
378 new programs. Programs should use the standard names exp10, exp10f and
379 exp10l for these functions instead.
381 * The mcontext_t type is no longer the same as struct sigcontext. On
382 platforms where it was previously the same, this changes the C++ name
383 mangling for interfaces involving this type.
385 * The add-ons mechanism for building additional packages at the same time as
386 glibc has been removed. The --enable-add-ons configure option is now
389 * The --without-fp configure option is now ignored. Whether hardware
390 floating-point instructions are used is now configured based on whether
391 the compiler used at configure time (without any options implied by a
392 --with-cpu= configure option) uses such instructions.
394 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
395 the specified string can be parsed as a domain name.
397 * In the malloc_info output, the <heap> element may contain another <aspace>
398 element, "subheaps", which contains the number of sub-heaps.
400 * The libresolv function p_secstodate is no longer supported for new
403 * The tilepro-*-linux-gnu configuration is no longer supported.
405 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
406 and will be removed in a future release. Software that is still using
407 either header should be updated to use standard <stdio.h> interfaces
410 libio.h was originally the header for a set of supported GNU extensions,
411 but they have not been maintained as such in many years, they are now
412 standing in the way of improvements to stdio, and we don't think there are
413 any remaining external users. _G_config.h was never intended for public
414 use, but predates the bits convention.
416 Changes to build and runtime requirements:
418 * bison version 2.7 or later is required to generate code in the 'intl'
421 Security related changes:
423 CVE-2009-5064: The ldd script would sometimes run the program under
424 examination directly, without preventing code execution through the
425 dynamic linker. (The glibc project disputes that this is a security
426 vulnerability; only trusted binaries must be examined using the ldd
429 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
430 suffered from a one-byte overflow during ~ operator processing (either
431 on the stack or the heap, depending on the length of the user name).
432 Reported by Tim Rühsen.
434 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
435 would sometimes fail to free memory allocated during ~ operator
436 processing, leading to a memory leak and, potentially, to a denial
439 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
440 without GLOB_NOESCAPE, could write past the end of a buffer while
441 unescaping user names. Reported by Tim Rühsen.
443 CVE-2017-17426: The malloc function, when called with an object size near
444 the value SIZE_MAX, would return a pointer to a buffer which is too small,
445 instead of NULL. This was a regression introduced with the new malloc
446 thread cache in glibc 2.26. Reported by Iain Buclaw.
448 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
449 to the allocation of too much memory. (This is not a security bug per se,
450 it is mentioned here only because of the CVE assignment.) Reported by
453 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
454 of the number of search path components. (This is not a security
455 vulnerability per se because no trust boundary is crossed if the fix for
456 CVE-2017-1000366 has been applied, but it is mentioned here only because
457 of the CVE assignment.) Reported by Qualys.
459 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
460 for AT_SECURE or SUID binaries could be used to load libraries from the
463 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
464 succeeds without returning an absolute path due to unexpected behaviour
465 of the Linux kernel getcwd syscall. Reported by halfdog.
467 CVE-2018-6485: The posix_memalign and memalign functions, when called with
468 an object size near the value of SIZE_MAX, would return a pointer to a
469 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
471 CVE-2018-6551: The malloc function, when called with an object size near
472 the value of SIZE_MAX, would return a pointer to a buffer which is too
473 small, instead of NULL.
475 The following bugs are resolved with this release:
477 [866] glob: glob should match dangling symlinks
478 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
479 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
481 [5997] math: Very slow execution of sinf function
482 [10580] localedata: hr_HR: updated locale
483 [10871] locale: 'mon' array should contain both nominative and genitive
485 [12349] localedata: eu_ES: incorrect thousands separator
486 [13605] localedata: shn_MM: new Shan locale
487 [13805] localedata: ru_RU: currency should use ',' as radix point
488 [13953] localedata: km_KH: locale update
489 [13994] localedata: mjw_IN: new locale
490 [14121] build: make writes .mo files in po directory
491 [14333] libc: Fix the race between atexit() and exit()
492 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
493 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
495 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
496 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
497 width Latin characters
498 [15332] localedata: es_CU: locale update
499 [15436] stdio: Don't close or flush stdio streams on abort
500 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
502 [16148] localedata: ca_ES: incorrect thousands separator
503 [16750] dynamic-link: ldd should not try to execute the binaries
505 [16777] localedata: pl_PL: incorrect thousands separator in locale
506 [16905] localedata: hanzi: new collation
507 [17563] localedata: cmn_TW: add hanzi collation
508 [17750] localedata: wrong collation order of diacritics in most locales
509 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
511 [17956] build: Build fails on missing definitions from header file
512 nss/nss.h when Mozilla NSS is used for cryptography
513 [18203] libc: realpath() does not handle unreachable paths correctly
514 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
515 [18812] localedata: kab_DZ: new Kabyle Algeria locale
516 [18822] libc: Internal functions are called via PLT
517 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
518 [19170] libc: __gmon_start__ defined in hppa in crtn.S
519 [19574] libc: glibc should support building static PIE binaries
520 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
522 [19971] glob: glob: Do not skip entries with zero d_ino values
523 [19982] localedata: fr.po: spelling mistake for error code EXDEV
524 [20008] localedata: km_KH: convert to translit_neutral
525 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
526 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
527 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
528 [20482] localedata: de_CH: abbreviated weekdays should be two letters
529 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
531 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
532 they're set, wrong errors returned by gaih_inet when lookup functions
534 [20756] localedata: [PATCH] Use Unicode wise thousands separator
535 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
537 [20952] localedata: yuw_PG: new locale
538 [21084] localedata: charmaps/IBM858: new codepage
539 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
540 [21242] libc: assert gives pedantic warning in old gcc versions
541 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
542 __regcall calling convention
543 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
544 [21326] libc: C99 functions are not declared for C++11 and later if
545 _GNU_SOURCE is not predefined by g++
546 [21457] libc: sys/ucontext.h namespace
547 [21530] libc: tmpfile() should be implemented using O_TMPFILE
548 [21660] math: GCC fails to compile a formula with tgmath.h
549 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
550 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
551 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
552 [21684] math: tgmath.h handling of complex integers
553 [21685] math: tgmath.h handling of bit-fields
554 [21686] math: tgmath.h handling of __int128
555 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
557 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
559 [21750] localedata: column width of characters incompatible with classical
561 [21754] malloc: malloc: Perform as little work as possible after heap
562 consistency check failures
563 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
564 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
565 [21791] string: Unused XXX_chk_XXX functions in libc.a
566 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
568 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
570 [21845] localedata: Added new Locale bho_NP
571 [21853] localedata: Fix abday Which looks same as day in zh_SG
572 [21854] localedata: Added New Locale en_SC
573 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
574 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
575 _dl_runtime_resolve_avx_slow
576 [21885] network: getaddrinfo: gethosts does not release resolver context
577 on memory allocation failure
578 [21899] libc: XPG4.2 sigaction namespace
579 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
581 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
583 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
584 errno value, causing getaddrinfo to fail
585 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
586 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
587 instead of EAI_NODATA
588 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
589 constant PTRACE_SEIZE_DEVEL
590 [21930] math: C-only gcc builtins used in <math.h> isinf
591 [21932] network: Unpaired __resolv_context_get in generic get*_r
593 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
594 [21944] libc: sigval namespace
595 [21951] localedata: Update hanzi collation by stroke
596 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
597 sysdeps/x86_64/fpu/e_expf.S
598 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
599 allocation in syscall loops)
600 [21959] localedata: Fix Country name for xh_ZA
601 [21960] localedata: Fix abmon for bem_ZM
602 [21966] math: AVX2 mathvec functions use FMA without checking
603 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
604 [21971] localedata: Added New Locale for mfe_MU
605 [21972] libc: assert macro requires operator== (int) for its argument type
606 [21973] math: [sparc] libm missing sqrtl compat symbol
607 [21974] libc: Remove __bb_init_func and __bb_exit_func
608 [21982] string: stratcliff.c: error: assuming signed overflow does not
610 [21986] stdio: __guess_grouping is called incorrectly
611 [21987] math: [sparc32] wrong bits/long-double.h installed
612 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
614 [22022] localedata: Missing country_name for mni_IN
615 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
617 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
619 [22026] locale: iconv_open: heap overflow on gconv_init failure
620 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
621 [22035] math: [m68k] bits/math-inline.h macro namespace
622 [22038] localedata: Fix abbreviated weeks and months for Somali
623 [22044] localedata: Remove redundant data for Limburgish Language
624 [22050] malloc: Linking with -lmcheck does not hook
625 __malloc_initialize_hook correctly
626 [22051] libc: zero terminator in the middle of glibc's .eh_frame
627 [22052] malloc: malloc failed to compile with GCC 7 and -O3
628 [22070] localedata: charmaps/UTF-8: wcwidth for
629 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
630 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
631 Jungseong and Jongseong) should be 0
632 [22078] nss: nss_files performance issue in multi mode
633 [22082] math: bits/math-finite.h exp10 condition
634 [22086] libc: pcprofiledump incorrect cross-endian condition
635 [22093] dynamic-link: ld.so no longer searches in .../x86_64
636 [22095] network: Name server address allocation memory leak in resolv.conf
638 [22096] network: __resolv_conf_attach can incorrectly free passed conf
640 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
641 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
643 [22111] malloc: malloc: per thread cache is not returned when thread exits
644 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
645 [22134] libc: [linux] implement fexecve with execveat
646 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
648 [22145] libc: ttyname() gives up too early in the face of namespaces
649 [22146] math: C++ build issue with float128 on x86_64
650 [22153] nptl: nptl: save error code before process termination
651 [22156] libc: [hppa,ia64,microblaze] Executable stack default
652 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
653 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
654 [22165] libc: [hppa] Text relocations in libc.so
655 [22180] libc: destructor registered via __cxa_atexit is called twice
656 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
657 symlinks") cause "make" segfaults
658 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
660 [22207] libc: FAIL: stdlib/test-atexit-race
661 [22225] math: nearbyint arithmetic moved before feholdexcept
662 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
663 [22235] math: iscanonical in C++ and float128
664 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
665 the svid compat wrapper
666 [22244] math: ynf and yn are wrong without the svid compat wrapper
667 [22273] libc: Improper assert in Linux posix_spawn implementation
668 [22284] libc: -pg -pie doesn't work
669 [22292] locale: localedef exits with error 4 when it should be error 1
670 [22294] locale: Allow "" for int_currency_symbol definition in locales.
671 [22295] locale: Don't warn on non-symbolic characters in locale sources in
673 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
674 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
676 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
677 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
678 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
679 [22322] libc: [mips64] wrong bits/long-double.h installed
680 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
681 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
683 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
684 [22343] malloc: Integer overflow in posix_memalign
685 [22347] libc: getrandom() returns the number of bytes that were copied to
686 the buffer even though the comments say "Return 0 on success and -1 on
688 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
689 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
691 [22370] dynamic-link: Incorrect note padding check
692 [22375] libc: malloc returns pointer from tcache_get when should return
693 NULL (CVE-2017-17426)
694 [22377] math: iseqsig, float128 and C++
695 [22382] localedata: Error in tpi_PG locale
696 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
697 inside the ASCII printable range
698 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
700 [22403] localedata: Slash needs escaping in some locales
701 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
702 [22409] network: res_hnok does not accept some host names used on the
704 [22412] network: res_dnok, res_hnok should perform syntax checks
705 [22413] network: ns_name_pton ignores syntactically invalid trailing
707 [22415] stdio: setvbuf can lead to invalid free/segfault
708 [22432] build: Non-deterministic build
709 [22439] malloc: malloc_info should compute summary statistics for all sub-
711 [22442] network: if_nametoindex could report index for the wrong
713 [22446] build: aliasing violation calling readlink in handle_request
714 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
715 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
716 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
717 __nscd_hash/__nss_hash
718 [22463] network: p_secstodate overflow handling
719 [22469] localedata: pl_PL LC_COLLATE does not use i18n
720 [22478] libc: sigwait can fail with EINTR
721 [22505] libc: ldconfig processes include directive in locale-specific
723 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
724 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
725 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
726 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
727 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
728 [22534] localedata: Collation rules for Serbian and Bosnian should be the
730 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
731 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
732 [22577] libc: missing newline after "cannot allocate TLS data structures
734 [22588] manual: manual/conf.texi: missing underscore in front of
736 [22593] math: nextafter and nexttoward are declared with const attribute
737 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
738 [22603] string: ia64 memchr overflows internal pointer check
739 [22605] libc: SH clone does not set the exit code correctly
740 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
742 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
743 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
744 modified in case of success
745 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
746 [22615] manual: manual: ambiguous wording about errno value in case of
748 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
750 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
751 binaries or if /proc is not mounted (CVE-2017-16997)
752 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
753 [22630] build: $(no-pie-ldflag) is no longer effective
754 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
755 [22635] nptl: pthread_self returns NULL before libpthread is loaded
756 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
757 [22637] nptl: guard size is subtracted from thread stack size instead of
759 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
760 [22657] localedata: hu_HU: Avoid double space in date
761 [22660] math: fmax, fmin sNaN handling on alpha
762 [22664] libc: New warning of GCC8
763 [22665] math: alpha: ceil and floor raise inexact exceptions
764 [22666] math: alpha: trunc raise inexact exceptions
765 [22667] libc: makecontext lacks stack alignment on i386
766 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
767 [22679] libc: getcwd(3) can succeed without returning an absolute path
769 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
771 [22687] math: [powerpc-nofpu] complex long double functions spurious
773 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
774 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
775 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
777 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
778 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
780 [22701] nis: Incomplete removal of libnsl
781 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
782 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
783 [22715] dynamic-link: FAIL: elf/tst-audit10
784 [22719] libc: Backtrace tests fail on hppa
785 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
786 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
787 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
788 before the first call to crypt_r () in crypt/badsalttest.c
795 * A per-thread cache has been added to malloc. Access to the cache requires
796 no locks and therefore significantly accelerates the fast path to allocate
797 and free small amounts of memory. Refilling an empty cache requires locking
798 the underlying arena. Performance measurements show significant gains in a
799 wide variety of user workloads. Workloads were captured using a special
800 instrumented malloc and analyzed with a malloc simulator. Contributed by
801 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
803 * Unicode 10.0.0 Support: Character encoding, character type info, and
804 transliteration tables are all updated to Unicode 10.0.0, using
805 generator scripts contributed by Mike FABIAN (Red Hat).
806 These updates cause user visible changes, especially the changes in
807 wcwidth for many emoji characters cause problems when emoji sequences
808 are rendered with pango, see for example:
809 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
811 * Collation of Hungarian has been overhauled and is now consistent with "The
812 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
815 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
817 - The GNU C Library will now detect when /etc/resolv.conf has been
818 modified and reload the changed configuration. The new resolver option
819 “no-reload” (RES_NORELOAD) disables this behavior.
821 - The GNU C Library now supports an arbitrary number of search domains
822 (configured using the “search” directive in /etc/resolv.conf);
823 previously, there was a hard limit of six domains. For backward
824 compatibility, applications that directly modify the ‘_res’ global
825 object are still limited to six search domains.
827 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
828 Library will now randomly pick a name server from the configuration as a
829 starting point. (Previously, the second name server was always used.)
831 * The tunables feature is now enabled by default. This allows users to tweak
832 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
834 * New function reallocarray, which resizes an allocated block (like realloc)
835 to the product of two sizes, with a guaranteed clean failure upon integer
836 overflow in the multiplication. Originally from OpenBSD, contributed by
837 Dennis Wölfing and Rüdiger Sonderfeld.
839 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
840 These are extended versions of preadv and pwritev, respectively, taking an
841 additional flags argument. The set of supported flags depends on the
842 running kernel; full support currently requires kernel 4.7 or later.
844 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
845 create a new session ID for the spawned process. This feature is
846 scheduled to be added to the next major revision of POSIX; for the time
847 being, it is available under _GNU_SOURCE.
849 * errno.h is now safe to use from C-preprocessed assembly language on all
850 supported operating systems. In this context, it will only define the
851 Exxxx constants, as preprocessor macros expanding to integer literals.
853 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
854 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
855 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
856 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
858 To compile programs that use this feature, the compiler must support
859 128-bit floating point with the type name _Float128 (as defined by TS
860 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
861 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
862 must be defined to make the new interfaces visible.
864 The new functions and macros correspond to those present for other
865 floating-point types (except for a few obsolescent interfaces not
866 supported for the new type), with F128 or f128 suffixes; for example,
867 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
868 printf or scanf formats for the new type; the strfromf128 and strtof128
869 interfaces should be used instead.
871 Deprecated and removed features, and other changes affecting compatibility:
873 * The synchronization that pthread_spin_unlock performs has been changed to
874 now be equivalent to a C11 atomic store with release memory order to the
875 spin lock's memory location. Previously, several (but not all)
876 architectures used stronger synchronization (e.g., containing what is
877 often called a full barrier). This change can improve performance, but
878 may affect odd fringe uses of spin locks that depend on the previous
879 behavior (e.g., using spin locks as atomic variables to try to implement
880 Dekker's mutual exclusion algorithm).
882 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
885 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
886 will only be built and installed when the GNU C Library is configured with
887 --enable-obsolete-rpc. This allows alternative RPC implementations, such
888 as TIRPC or rpcsvc-proto, to be used.
890 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
891 libnss_compat, are deprecated, and will not be built or installed by
894 The NIS(+) support library, libnsl, is also deprecated. By default, a
895 compatibility shared library will be built and installed, but not headers
896 or development libraries. Only a few NIS-related programs require this
897 library. (In particular, the GNU C Library has never required programs
898 that use 'gethostbyname' to be linked with libnsl.)
900 Replacement implementations based on TIRPC, which additionally support
901 IPv6, are available from <https://github.com/thkukuk/>. The configure
902 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
903 name service modules, to be built and installed.
905 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
906 support is enabled, the configured recursive resolver must support EDNS.
907 (Responding to EDNS-enabled queries with responses which are not
908 EDNS-enabled is fine, but FORMERR responses are not.)
910 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
911 servers have not supported this opcode for a long time.
913 * The _res_opcodes variable has been removed from libresolv. It had been
914 exported by accident.
916 * <string.h> no longer includes inline versions of any string functions,
917 as this kind of optimization is better done by the compiler. The macros
918 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
920 * The nonstandard header <xlocale.h> has been removed. Most programs should
921 use <locale.h> instead. If you have a specific need for the definition of
922 locale_t with no other declarations, please contact
923 libc-alpha@sourceware.org and explain.
925 * The obsolete header <sys/ultrasound.h> has been removed.
927 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
929 * The obsolete function cfree has been removed. Applications should use
932 * The stack_t type no longer has the name struct sigaltstack. This changes
933 the C++ name mangling for interfaces involving this type.
935 * The ucontext_t type no longer has the name struct ucontext. This changes
936 the C++ name mangling for interfaces involving this type.
938 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
939 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
940 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
941 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
942 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
943 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
944 fpregset_t type no longer has the name struct fpu. This changes the C++
945 name mangling for interfaces involving those types.
947 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
948 synced with the kernel:
950 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
951 are not supported on this architecture and have been removed.
953 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
954 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
955 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
957 Programs that assume the GET/SETREGS ptrace requests are universally
958 available will now fail to build, instead of malfunctioning at runtime.
960 Changes to build and runtime requirements:
962 * Linux kernel 3.2 or later is required at runtime, on all architectures
963 supported by that kernel. (This is a change from version 2.25 only for
966 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
968 * On most architectures, GCC 4.9 or later is required to build the GNU C
969 Library. On powerpc64le, GCC 6.2 or later is required.
971 Older GCC versions and non-GNU compilers are still supported when
972 compiling programs that use the GNU C Library. (We do not know exactly
973 how old, and some GNU extensions to C may be _de facto_ required. If you
974 are interested in helping us make this statement less vague, please
975 contact libc-alpha@sourceware.org.)
977 Security related changes:
979 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
980 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
982 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
983 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
985 * Avoid printing a backtrace from the __stack_chk_fail function since it is
986 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
989 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
990 fixed (CVE-2017-12133).
992 The following bugs are resolved with this release:
994 [984] network: Respond to changed resolv.conf in gethostbyname
995 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
996 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
998 [12189] libc: __stack_chk_fail should not attempt a backtrace
1000 [14096] time: Race condition on timezone/tst-timezone.out
1001 [14172] localedata: az_IR: new locale
1002 [14995] build: glibc fails to build if gold is the default linker, even if
1004 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1005 [16637] network: inet_pton function is accepting IPv6 with bad format
1006 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1007 x86 and other generic code
1008 [16875] localedata: ko_KR: fix lang_name
1009 [17225] localedata: ar_SY: localized month names for May and June are
1011 [17297] localedata: da_DK: wrong date_fmt string
1012 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1013 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1014 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1015 [19066] localedata: ar_SA abbreviated day and month names are in English
1016 [19569] network: resolv: Support an arbitrary number of search domains
1017 [19570] network: Implement random DNS server selection in the stub
1019 [19838] locale: localedef fails on PA-RISC
1020 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1021 order of 0D36 and 0D37
1022 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1024 [20098] libc: FAIL: debug/backtrace-tst on hppa
1025 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1027 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1029 [20313] localedata: Update locale data to Unicode 9.0
1030 [20424] manual: Document how to provide a malloc replacement
1031 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1033 [20686] locale: Add el_GR@euro to SUPPORTED.
1034 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1035 failures consistently
1036 [21015] dynamic-link: Document and fix --enable-bind-now
1037 [21016] nptl: pthread_cond support is broken on hppa
1038 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1039 frame-pointer on i386
1040 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1042 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1043 [21088] libc: Build fails with --enable-static-nss
1044 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1045 generic c code is used
1046 [21109] libc: Tunables broken on big-endian
1047 [21112] math: powf has large ulp errors with base close to 1 and exponent
1049 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1051 [21120] malloc: glibc malloc is incompatible with GCC 7
1052 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1054 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1055 (-0.0) when linking with -lieee
1056 [21171] math: log10, log2 and lgamma return incorrect results
1057 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1058 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1059 [21207] localedata: ce_RU: update weekdays from CLDR
1060 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1061 [21217] localedata: Update months from CLDR-31
1062 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1063 [21243] libc: support_delete_temp_file should issue warning for failed
1065 [21244] libc: support resolv_test_start() socket fd close should be
1067 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1068 new posix_spawn implementation
1069 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1070 leads to lower CPU frequency
1071 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1072 [21261] libc: [sparc64] bits/setjmp.h namespace
1073 [21267] network: [mips] bits/socket.h IOC* namespace
1074 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1075 [21270] libc: mmap64 silently truncates large offset values
1076 [21275] libc: posix_spawn always crashes on ia64 now
1077 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1078 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1079 [21289] libc: Incorrect declaration for 32-bit platforms with
1080 _FILE_OFFSET_BITS=64 causes build error
1081 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1082 not support gethostbyname4_r
1083 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1085 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1087 [21340] libc: Support POSIX_SPAWN_SETSID
1088 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1089 [21359] network: ns_name_pack needs additional byte in destination buffer
1090 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1091 against fragmentation attacks (CVE-2017-12132)
1092 [21369] network: resolv: Remove EDNS fallback
1093 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1095 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1096 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1097 [21393] stdio: Missing dup3 error check in freopen, freopen64
1098 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1099 [21399] localedata: Bad description for U00EC in
1100 localedata/charmaps/CP1254
1101 [21411] malloc: realloc documentation error
1102 [21426] network: sys/socket.h uio.h namespace
1103 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1104 [21445] libc: signal.h bsd_signal namespace
1105 [21455] network: Network headers stdint.h namespace
1106 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1108 [21475] network: resolv: Overlong search path is truncated mid-label
1109 [21511] libc: sigstack namespace
1110 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1111 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1112 bits/syscall.h: No such file or directory
1113 [21517] libc: struct sigaltstack namespace
1114 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1115 [21533] localedata: Update locale data to Unicode 10.0
1117 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1118 junk at end of line, first unrecognized character is `@'
1119 [21538] libc: SIG_HOLD missing for XPG4
1120 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1121 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1122 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1123 [21543] libc: sigevent namespace
1124 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1125 size for general purpose registers in mcontext_t structure
1126 [21550] libc: sigwait namespace
1127 [21552] libc: XPG4 bsd_signal namespace
1128 [21554] libc: sigpause namespace
1129 [21560] libc: sys/wait.h signal.h namespace
1130 [21561] libc: waitid namespace
1131 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1132 '/usr/include/stdlib.h': Permission denied
1133 [21575] libc: sys/wait.h missing struct rusage definition
1134 [21584] libc: sigaltstack etc namespace
1135 [21597] libc: siginterrupt namespace
1136 [21607] math: hppa: FAIL: math/test-tgmath
1137 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1139 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1140 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1141 stack and heap (CVE-2017-1000366)
1142 [21625] libc: wait3 namespace
1143 [21654] nss: Incorrect pointer alignment in NSS group merge result
1145 [21657] network: Parse interface zone id for node-local multicast
1146 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1148 [21666] libc: .symver is used on common symbol
1149 [21668] network: resolv: res_init cross-thread broadcast introduces race
1151 [21687] math: tgmath.h totalorder, totalordermag return type
1152 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1154 [21696] libc: Incorrect assumption of of __cpu_mask in
1155 posix/sched_cpucount.c
1156 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1157 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1159 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1160 leading to relocation crash
1161 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1162 `resolv_conf_matches (resp, conf)' failed.
1163 [21710] localedata: Added Samoan language locale for Samoa
1164 [21711] localedata: Pashto yesstr/nostr locale are missing
1165 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1166 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1167 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1168 [21723] localedata: yesstr/nostr missing for Chinese language locale
1169 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1171 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1173 [21728] localedata: New Locale for Tongan language
1174 [21729] localedata: incorrect LC_NAME fields for hi_IN
1175 [21733] localedata: yesstr and nostr are missing for zh_HK
1176 [21734] localedata: Missing yesstr and nostr are for kw_GB
1177 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1178 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1179 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1180 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1182 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1184 [21749] localedata: Wrong abbreviated day name (“abday”) for
1186 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1187 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1188 [21759] localedata: missing yesstr and nostr for Tigrinya
1189 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1190 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1191 [21767] localedata: Missing Bislama locales
1192 [21768] localedata: Missing yesstr and nostr for aa_ET
1193 [21770] localedata: Missing Field in li_NL
1194 [21778] nptl: Robust mutex may deadlock
1195 [21779] libc: MicroBlaze segfaults when loading libpthread
1196 [21783] localedata: Fix int_select international_call_prefixes
1197 [21784] localedata: Inconsistency in country_isbn
1198 [21788] localedata: Missing Country Postal Abbreviations
1199 [21794] localedata: Added-country_isbn-for-Italy
1200 [21795] localedata: Add/Fix country_isbn for France
1201 [21796] localedata: Added country_isbn for Republic of Korea
1202 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1203 [21799] localedata: Added int_select international_call_prefixes
1204 [21801] localedata: Added int_select international_call_prefixes
1205 [21804] nptl: Double semicolon in thread-shared-types.h
1206 [21807] localedata: LC_ADDRESS fix for pap_CW
1207 [21808] localedata: Fix LC_ADDRESS for pap_AW
1208 [21821] localedata: Added country_name in mai_IN
1209 [21822] localedata: Fix LC_TIME for mai_IN
1210 [21823] localedata: missing yesstr, nostr for sa_IN
1211 [21825] localedata: Fix name_mrs for mag_IN
1212 [21828] localedata: 2.26 changelog should mention user visible changes
1214 [21835] localedata: Added Maithili language locale for Nepal
1215 [21838] localedata: Removed redundant data for the_NP
1216 [21839] localedata: Fix LC_MONETARY for ta_LK
1217 [21844] localedata: Fix Latin characters and Months Sequence.
1218 [21848] localedata: Fix mai_NP Title Name
1223 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1224 24731-2:2010, is supported to enable declarations of functions from that
1225 TR. Note that not all functions from that TR are supported by the GNU C
1228 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1229 18661-1:2014, is supported to enable declarations of functions and macros
1230 from that TS. Note that not all features from that TS are supported by
1233 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1234 18661-4:2015, is supported to enable declarations of functions and macros
1235 from that TS. Note that most features from that TS are not supported by
1238 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1239 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1240 Since the GNU C Library defaults to a much newer revision of POSIX, this
1241 will only affect programs that specifically request an old conformance
1242 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1243 see a change in the visible declarations, but a program compiled with
1244 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1247 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1248 defined by all multithreaded code, but glibc has not required this for
1251 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1252 means that in a future release, the macros “major”, “minor”, and “makedev”
1253 will only be available from <sys/sysmacros.h>.
1255 These macros are not part of POSIX nor XSI, and their names frequently
1256 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1257 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1258 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1261 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1262 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1263 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1265 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1266 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1267 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1268 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1269 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1270 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1271 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1272 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1273 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1274 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1275 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1278 * New <math.h> features are added from TS 18661-1:2014:
1280 - Signaling NaN macros: SNANF, SNAN, SNANL.
1282 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1283 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1284 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1286 - llogb functions: the llogb, llogbf and llogbl functions, and the
1287 FP_LLOGB0 and FP_LLOGBNAN macros.
1289 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1292 - Comparison macros: iseqsig.
1294 - Classification macros: iscanonical, issubnormal, iszero.
1296 - Total order functions: totalorder, totalorderf, totalorderl,
1297 totalordermag, totalordermagf, totalordermagl.
1299 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1301 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1302 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1304 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1305 are added to libc. They convert a floating-point number into string.
1307 * Most of glibc can now be built with the stack smashing protector enabled.
1308 It is recommended to build glibc with --enable-stack-protector=strong.
1309 Implemented by Nick Alcock (Oracle).
1311 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1312 intended to be used instead of memset() to erase sensitive data after use;
1313 the compiler will not optimize out calls to explicit_bzero even if they
1314 are "unnecessary" (in the sense that no _correct_ program can observe the
1315 effects of the memory clear).
1317 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1318 to float instead of double. This does not affect the ABI of any libraries
1319 that are part of the GNU C Library, but may affect the ABI of other
1320 libraries that use this type in their interfaces.
1322 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1323 float_t and double_t types are now defined to long double instead of float
1324 and double. These options are not the default, and this does not affect
1325 the ABI of any libraries that are part of the GNU C Library, but it may
1326 affect the ABI of other libraries that use this type in their interfaces,
1327 if they are compiled or used with those options.
1329 * The getentropy and getrandom functions, and the <sys/random.h> header file
1332 * The buffer size for byte-oriented stdio streams is now limited to 8192
1333 bytes by default. Previously, on Linux, the default buffer size on most
1334 file systems was 4096 bytes (and thus remains unchanged), except on
1335 network file systems, where the buffer size was unpredictable and could be
1336 as large as several megabytes.
1338 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1339 for the Linux quota interface which predates kernel version 2.4.22 has
1342 * The malloc_get_state and malloc_set_state functions have been removed.
1343 Already-existing binaries that dynamically link to these functions will
1344 get a hidden implementation in which malloc_get_state is a stub. As far
1345 as we know, these functions are used only by GNU Emacs and this change
1346 will not adversely affect already-built Emacs executables. Any undumped
1347 Emacs executables, which normally exist only during an Emacs build, should
1348 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1350 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1351 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1352 “no-ip6-dotint” had already been the default, and support for the
1353 “ip6-dotint” option was removed from the Internet in 2006.
1355 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1356 flag from <resolv.h> have been removed. The option relied on a
1357 backwards-incompatible DNS extension which was never deployed on the
1360 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1361 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1362 They were already unimplemented.
1364 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1365 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1366 removed again from the IETF name lookup interface specification in RFC
1367 2553. Applications should use getaddrinfo instead.
1369 * DNSSEC-related declarations and definitions have been removed from the
1370 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1371 decode the data part of DNSSEC record types. Previous versions of glibc
1372 only implemented minimal support for the previous version of DNSSEC, which
1373 is incompatible with the currently deployed version.
1375 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1376 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1377 <arpa/nameser.h> header file because the distinction between RR types and
1378 meta-RR types is not officially standardized, subject to revision, and
1379 thus not suitable for encoding in a macro.
1381 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1382 and rhook members of the res_state type in <resolv.h> have been removed.
1383 The glibc stub resolver did not support these hooks, but the header file
1384 did not reflect that.
1386 * For multi-arch support it is recommended to use a GCC which has
1387 been built with support for GNU indirect functions. This ensures
1388 that correct debugging information is generated for functions
1389 selected by IFUNC resolvers. This support can either be enabled by
1390 configuring GCC with '--enable-gnu-indirect-function', or by
1391 enabling it by default by setting 'default_gnu_indirect_function'
1392 variable for a particular architecture in the GCC source file
1395 * GDB pretty printers have been added for mutex and condition variable
1396 structures in POSIX Threads. When installed and loaded in gdb these pretty
1397 printers show various pthread variables in human-readable form when read
1398 using the 'print' or 'display' commands in gdb.
1400 * Tunables feature added to allow tweaking of the runtime for an application
1401 program. This feature can be enabled with the '--enable-tunables' configure
1402 flag. The GNU C Library manual has details on usage and README.tunables has
1403 instructions on adding new tunables to the library.
1405 * A new version of condition variables functions have been implemented in
1406 the NPTL implementation of POSIX Threads to provide stronger ordering
1409 * A new version of pthread_rwlock functions have been implemented to use a more
1410 scalable algorithm primarily through not using a critical section anymore to
1413 Security related changes:
1415 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1416 have been created with makecontext could fail to terminate due to a
1417 missing .cantunwind annotation. This has been observed to lead to a hang
1418 (denial of service) in some Go applications compiled with gccgo. Reported
1419 by Andreas Schwab. (CVE-2016-6323)
1421 * The DNS stub resolver functions would crash due to a NULL pointer
1422 dereference when processing a query with a valid DNS question type which
1423 was used internally in the implementation. The stub resolver now uses a
1424 question type which is outside the range of valid question type values.
1427 The following bugs are resolved with this release:
1429 [4099] stdio: Overly agressive caching by stream i/o functions.
1430 [7065] build: Support building glibc with -fstack-protector or -fstack-
1432 [9842] localedata: en_CA: incorrect date format
1433 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1434 before it started waiting
1435 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1436 [15765] nptl: sem_open is wrongly a cancellation point
1437 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1438 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1440 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1441 library linked with pthread
1442 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1443 "FIXME: Ingo" issue)
1444 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1445 [17252] libc: getrandom and getentropy syscall
1446 [17730] malloc: thread-local storage is sometimes improperly free()'d
1447 after being __libc_memalign()'d
1448 [18241] stdio: failed fseek on memstream does not set errno and can fail
1450 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1452 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1453 PTHREAD_PRIO_INHERIT mutex on ARM
1454 [18784] network: res_query and related function crash for special record
1455 type queries (CVE-2015-5180)
1456 [19380] math: strtod does not raise "inexact"
1457 [19387] string: Integer overflow in memchr
1458 [19390] string: Integer overflow in strncat
1459 [19398] build: linknamespace tests fail with massively parallel build
1460 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1462 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1463 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1464 [19514] libc: [PATCH] Fix spelling errors in spelling
1465 "implement"/"implementation" in several places
1466 [19582] network: Deprecate RES_USE_INET6
1467 [19673] manual: clog10 docs appear to be erroneous
1468 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1469 causes a segmentation fault
1470 [19826] libc: invalid pointer returned from __tls_get_addr with static
1472 [20016] network: resolv: Remove hooks support from the API
1473 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1475 [20033] math: [x86_64] vectorized math function don't call the __finite
1477 [20116] nptl: use after free in pthread_create
1478 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1479 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1481 [20311] nptl: please install proc_service.h
1482 [20366] libc: Compilation errors in installed headers in strict-compliance
1484 [20370] malloc: malloc: Arena free list management is still racy
1485 (incorrect fix in bug 19243)
1486 [20386] libc: assert (X = 0) does not result in GCC warning
1487 [20432] malloc: malloc: Minimize interface required for interposition
1488 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1489 in _Unwind_Backtrace (CVE-2016-6323)
1490 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1491 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1493 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1494 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1495 [20477] network: resolv: incorrect double-checked locking related to
1497 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1499 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1501 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1502 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1503 cause transition penalty
1504 [20517] math: sparcv9 missing fdiml compat symbol
1505 [20524] manual: strverscmp is inconsistent
1506 [20525] libc: <sys/quota.h> should be based on kernel headers
1507 [20539] math: GCC 7 -static -lm fails to link at -O3
1508 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1510 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1512 [20591] network: Remove obsolete DNSSEC support
1513 [20592] network: DNS resource record type classification macros in
1514 <arpa/nameser.h> are incorrect
1515 [20593] network: Update DNS RR type definitions
1516 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1517 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1519 [20629] network: libresolv: Remove support for bitstring labels
1521 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1522 assertion: "maxidx >= 2"
1523 [20660] math: [arm] Use VSQRT
1524 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1525 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1526 [20689] libc: Test for FMA should also check for AVX.
1527 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1528 [20715] math: iszero macro breaks existing code
1529 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1530 [20728] libc: powerpc: Missing TOC stub in clone
1531 [20729] build: glibc-2.24 fails to build for i486 with -Os
1532 [20750] build: Build fails with default PIE enabled gcc-6.x
1533 [20768] math: [alpha] sqrt fegetenv namespace
1534 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1536 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1537 __FLT_EVAL_METHOD__ is 2
1538 [20790] network: rpcgen buffer overrun in get_prog_declaration
1539 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1540 [20829] libc: crypt snprintf namespace
1541 [20847] libc: tst-vfork3 failure
1542 [20855] math: Default bits/mathdef.h has inappropriate float_t
1543 [20859] math: [sh4] FP_ILOGB0 invalid
1544 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1545 U+20AC), not same as GBK
1546 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1547 [20916] math: pow handling of sNaN arguments
1548 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1549 [20919] math: Bad pow (qNaN, 0) result with -lieee
1550 [20940] math: hypot sNaN handling
1551 [20947] math: fmax, fmin sNaN handling
1552 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1553 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1555 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1556 [20973] nptl: robust mutexes: Lost wake-ups
1557 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1558 [20978] nis: strlen on null pointer in nss_nisplus
1559 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1561 [21014] string: i686 memchr overflows internal pointer check
1562 [21019] libc: [mips] n32 lseek incorrect on overflow
1563 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1564 [21026] libc: [MIPS] readahead syscall is broken on n64
1565 [21028] math: Fallback fesetexceptflag should always succeed
1566 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1567 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1569 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1570 [21061] librt: [microblaze] librt lost clock_* exports
1571 [21073] libc: tunables: insecure environment variables passed to
1572 subprocesses with AT_SECURE
1573 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1577 * The minimum Linux kernel version that this version of the GNU C Library
1578 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1579 kernel version 2.6.32 or later suffices (on architectures that already
1580 required kernel versions more recent than 3.2, those requirements remain
1581 unchanged). Linux 3.2 or later kernel headers are required on all
1584 * The pap_AN locale has been deleted. This has been deprecated for a long
1585 time. It has been replaced by pap_AW & pap_CW, both of which have long
1586 been included in previous releases.
1588 * The readdir_r and readdir64_r functions have been deprecated. It is
1589 recommended to use readdir and readdir64 instead.
1591 * The type “union wait” has been removed. It was deprecated in the early
1592 1990s and never part of POSIX. Application code should use the int type
1593 instead of “union wait”.
1595 * A new NSS action is added to facilitate large distributed system
1596 administration. The action, MERGE, allows remote user stores like LDAP
1597 to be merged into local user stores like /etc/groups in order to provide
1598 easy to use, updated, and managed sets of merged credentials. The new
1599 action can be used by configuring it in /etc/nsswitch.conf:
1600 group: files [SUCCESS=merge] nis
1601 Implemented by Stephen Gallagher (Red Hat).
1603 * The deprecated __malloc_initialize_hook variable has been removed from the
1606 * The long unused localedef --old-style option has been removed. It hasn't
1607 done anything in over 16 years. Scripts using this option can safely
1610 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1611 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1612 return the next representable value in the direction of positive infinity
1613 and the nextdown functions return the next representable value in the
1614 direction of negative infinity. These are currently enabled as GNU
1617 Security related changes:
1619 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1620 could result in a stack overflow when getnetbyname was called with an
1621 overly long name. (CVE-2016-3075)
1623 * Previously, getaddrinfo copied large amounts of address data to the stack,
1624 even after the fix for CVE-2013-4458 has been applied, potentially
1625 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1626 instead. Reported by Michael Petlan. (CVE-2016-3706)
1628 * The glob function suffered from a stack-based buffer overflow when it was
1629 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1630 Reported by Alexander Cherepanov. (CVE-2016-1234)
1632 * The Sun RPC UDP client could exhaust all available stack space when
1633 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1634 alloca plugin for GCC. (CVE-2016-4429)
1636 * The IPv6 name server management code in libresolv could result in a memory
1637 leak for each thread which is created, performs a failing naming lookup,
1638 and exits. Over time, this could result in a denial of service due to
1639 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1641 The following bugs are resolved with this release:
1643 [1170] localedata: ne_NP: update Nepali locale definition file
1644 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1646 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1647 [6796] math: fdim() does not set errno on overflow
1648 [10354] libc: posix_spawn should use vfork() in more cases than presently
1649 [11213] localedata: localedata: add copyright disclaimer to locale files
1650 [12143] localedata: chr_US: new Cherokee locale
1651 [12450] localedata: sgs_LT: new locale
1652 [12676] localedata: ln_CD: new locale
1653 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1655 [13304] math: fma, fmaf, fmal produce wrong results
1656 [14259] build: --localedir arg to configure is ignored
1657 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1658 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1659 [14934] localedata: es_CL: wrong first weekday chilean locale
1660 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1662 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1664 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1665 [15368] nptl: raise() is not async-signal-safe
1666 [15479] math: ceil, floor, round and trunc raise inexact exception
1667 [15578] localedata: kk_KZ: various updates
1668 [16003] localedata: pap_AN: punt old locale
1669 [16137] localedata: iw_IL: punt old locale
1670 [16190] localedata: eo: new esperanto locale
1671 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1672 [16742] malloc: race condition: pthread_atfork() called before first
1673 malloc() results in unexpected locking behaviour/deadlocks
1674 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1676 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1677 [17565] localedata: pt_PT: wrong (work-)week start
1678 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1679 [17950] build: Build fails with -msse
1680 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1681 [18433] libc: posix_spawn does not return correctly upon failure to
1683 [18453] localedata: charmaps/IBM875: incorrect codes
1684 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1686 [18896] localedata: he_IL: improvements for currency
1687 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1688 Romanian locale data
1689 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1691 [19056] libc: Deprecate readdir_r
1692 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1694 [19198] localedata: nl_NL: small improvements for Dutch locales
1695 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1696 nameservers (CVE-2016-5417)
1697 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1699 [19400] locale: Language missing in "iso-639.def", trivial fix in
1701 [19431] malloc: Deadlock between fflush, getdelim, and fork
1702 [19505] libc: Incorrect file descriptor validity checks in
1703 posix_spawn_file_actions_add{open,close,dup2}
1704 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1705 when using RTLD_NEXT
1706 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1707 `intl/{gettextP,loadinfo}.h'
1708 [19534] libc: execle, execlp may use malloc
1709 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1710 [19573] network: res_nclose and __res_maybe_init disagree about name
1711 server initialization, breaking Hesiod
1712 [19575] localedata: Status of GB18030 tables
1713 [19581] localedata: sr_* date_fmt string contains additional newline
1714 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1716 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1717 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1718 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1719 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1721 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1723 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1725 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1726 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1727 [19633] locale: strfmon_l applies global locale to number formatting
1728 [19642] network: Memory leak in getnameinfo
1729 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1730 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1732 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1733 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1735 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1736 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1738 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1740 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1742 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1743 pointers and lengths in error-case.
1744 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1745 report errors on UTF-16 surrogates values.
1746 [19755] nscd: nscd assertion failure in gc
1747 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1748 [19759] libc: mempcpy shouldn't be inlined
1749 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1750 [19765] libc: s390 needs an optimized mempcpy
1751 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1752 NAME_MAX limit assumption (CVE-2016-1234)
1753 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1754 [19787] network: Missing and incorrect truncation checks in getnameinfo
1755 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1757 [19791] network: Assertion failure in res_query.c with un-connectable name
1759 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1760 [19822] math: libm.so install clobbers old version
1761 [19825] network: resolv: send_vc can return uninitialized data in second
1762 response to getaddrinfo
1763 [19830] network: nss_dns: should check RDATA length against buffer length
1764 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1765 confronted with A/AAAA records of invalid size
1766 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1767 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1768 [19853] stdio: Printing IBM long double in decimal with high precision is
1770 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1771 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1772 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1774 [19865] network: Assertion failure or memory leak in
1775 _nss_dns_getcanonname_r
1776 [19868] network: nss_dns: netent code does not skip over non-PTR records
1777 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1779 [19881] string: Improve x86-64 memset
1780 [19907] string: Incorrect memcpy tests
1781 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1783 [19925] libc: termios.h XCASE namespace
1784 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1785 [19929] libc: limits.h NL_NMAX namespace
1786 [19931] stdio: Memory leak in vfprintf
1787 [19957] libc: clone(CLONE_VM) access invalid parent memory
1788 [19963] localedata: en_IL: New locale
1789 [19989] stdio: stdio.h cuserid namespace
1790 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1792 [19996] locale: langinfo.h nl_langinfo_l namespace
1793 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1794 [20010] network: getaddrinfo: Stack overflow in hostent translation
1796 [20012] stdio: libio: fmemopen append mode failure
1797 [20014] stdio: stdio.h namespace for pre-threads POSIX
1798 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1799 [20023] libc: fcntl.h timespec namespace
1800 [20024] math: [x86_64] vectorized sincos trashes the stack
1801 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1802 [20041] time: sys/time.h timespec namespace
1803 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1804 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1805 [20051] libc: ttyslot in wrong header under wrong conditions
1806 [20054] libc: gethostname not declared for XPG4
1807 [20055] libc: termios.h missing tcgetsid for XPG4
1808 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1810 [20073] libc: sys/stat.h fchmod namespace
1811 [20074] libc: stdlib.h rand_r namespace
1812 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1813 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1815 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1816 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1818 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1819 [20119] libc: Wrong mask for processors level type from CPUID
1820 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1821 AS not supporting AVX512
1822 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1824 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1825 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1827 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1828 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1829 [20178] libc: posix_spawn{p} should not call exit
1830 [20191] stdio: libio: vtables hardening
1831 [20195] string: FMA4 detection requires CPUID execution with register
1833 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1834 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1836 [20212] math: acos (sNaN) returns sNaN
1837 [20213] math: asin (sNaN) returns sNaN
1838 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1839 [20218] math: [i386] asinhl (sNaN) returns sNaN
1840 [20219] math: [i386] atanhl (sNaN) returns sNaN
1841 [20222] stdio: fopencookie: Mangle function pointers
1842 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1843 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1844 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1846 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1847 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1848 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1849 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1850 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1851 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1852 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1853 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1854 [20240] math: modf (sNaN) returns sNaN
1855 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1856 [20250] math: frexp (sNaN) returns sNaN
1857 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1858 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1860 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1861 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1862 subscript is below array bounds [-Werror=array-bounds]
1863 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1864 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1866 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1867 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1869 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1870 "invalid" exceptions
1871 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1872 [20316] localedata: id_ID: Februari instead of Pebruari
1873 [20327] string: POWER8 strcasecmp returns incorrect result
1874 [20347] math: Failure: Test: j0_downward (0xap+0)
1875 [20348] libc: FAIL: misc/tst-preadvwritev64
1876 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1877 [20350] libc: There is no test for p{read,write}64
1878 [20357] math: Incorrect cos result for 1.5174239687223976
1879 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1883 * Unicode 8.0.0 Support: Character encoding, character type info, and
1884 transliteration tables are all updated to Unicode 8.0.0, using new
1885 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1886 These updates cause user visible changes, such as the fixes for bugs
1887 89, 16061, and 18568.
1889 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1890 kernel-internal CPU set size. This means that requests that change the
1891 CPU affinity which failed before (for example, an all-ones CPU mask) will
1892 now succeed. Applications that need to determine the effective CPU
1893 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1894 setting it because the kernel can adjust it (and the previous size check
1895 would not detect this in the majority of cases).
1897 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1898 the following new symbols are used: fts64_children, fts64_close,
1899 fts64_open, fts64_read and fts64_set.
1901 * getaddrinfo now detects certain invalid responses on an internal netlink
1902 socket. If such responses are received, an affected process will
1903 terminate with an error message of "Unexpected error <number> on netlink
1904 descriptor <number>" or "Unexpected netlink response of size <number> on
1905 descriptor <number>". The most likely cause for these errors is a
1906 multi-threaded application which erroneously closes and reuses the netlink
1907 file descriptor while it is used by getaddrinfo.
1909 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1910 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1911 unnecessary serialization of memory allocation requests across threads.
1912 The defect is now corrected. Users should see a substantial increase in
1913 the concurent throughput of allocation requests for applications which
1914 trigger this bug. Affected applications typically create and
1915 destroy threads frequently. (Bug 19048 was reported and analyzed by
1918 * There is now a --disable-timezone-tools configure option for disabling the
1919 building and installing of the timezone related utilities (zic, zdump, and
1920 tzselect). This is useful for people who build the timezone data and code
1921 independent of the GNU C Library.
1923 * The obsolete header <regexp.h> has been removed. Programs that require
1924 this header must be updated to use <regex.h> instead.
1926 * The obsolete functions bdflush, create_module, get_kernel_syms,
1927 query_module and uselib are no longer available to newly linked binaries;
1928 the header <sys/kdaemon.h> has been removed. These functions and header
1929 were specific to systems using the Linux kernel and could not usefully be
1930 used with the GNU C Library on systems with version 2.6 or later of the
1933 * Optimized string, wcsmbs and memory functions for IBM z13.
1934 Implemented by Stefan Liebler.
1936 * Newly linked programs that define a variable called signgam will no longer
1937 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1938 require signgam to be set by those functions must ensure that they use the
1939 variable provided by the GNU C Library and declared in <math.h>, without
1940 defining their own copy.
1942 * The minimum GCC version that can be used to build this version of the GNU
1943 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1944 still be used to compile programs using the GNU C Library.
1946 Security related changes:
1948 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1949 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1951 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1952 the pointer guard feature. It is always enabled. Previously,
1953 LD_POINTER_GUARD could be used to disable security hardening in binaries
1954 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1957 * An integer overflow in hcreate and hcreate_r could lead to an
1958 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1960 * The catopen function no longer has unbounded stack usage. Reported by
1961 Max. (CVE-2015-8779)
1963 * The nan, nanf and nanl functions no longer have unbounded stack usage
1964 depending on the length of the string passed as an argument to the
1965 functions. Reported by Joseph Myers. (CVE-2014-9761)
1967 * A stack-based buffer overflow was found in libresolv when invoked from
1968 libnss_dns, allowing specially crafted DNS responses to seize control
1969 of execution flow in the DNS client. The buffer overflow occurs in
1970 the functions send_dg (send datagram) and send_vc (send TCP) for the
1971 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1972 family. The use of AF_UNSPEC triggers the low-level resolver code to
1973 send out two parallel queries for A and AAAA. A mismanagement of the
1974 buffers used for those queries could result in the response of a query
1975 writing beyond the alloca allocated buffer created by
1976 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1977 the overflow. Thanks to the Google Security Team and Red Hat for
1978 reporting the security impact of this issue, and Robert Holiday of
1979 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1981 The following bugs are resolved with this release:
1983 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1984 [887] math: Math library function "logb" and "nextafter" inconsistent
1985 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1986 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1987 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1988 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1990 [4404] localedata: German translation of "Alarm clock" is misleading
1991 [6799] math: nextafter() and nexttoward() doen't set errno on
1992 overflow/underflow errors
1993 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1995 [10432] nis: _nss_nis_setnetgrent assertion failure
1996 [11460] libc: fts has no LFS support
1997 [12926] network: getaddrinfo()/make_request() may spin forever
1998 [13065] nptl: Race condition in pthread barriers
1999 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2000 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2002 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2003 [14912] libc: Rename non-installed bits/*.h headers
2004 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2005 [15367] math: Let gcc use __builtin_isinf
2006 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2007 [15421] math: lgamma wrongly sets signgam for ISO C
2008 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2009 FE_INVALID with argument out of range
2010 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2012 [15786] dynamic-link: ifunc resolver functions can smash function
2014 [15918] math: Unnecessary check for equality in hypotf()
2015 [16061] localedata: Review / update transliteration data
2016 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2018 [16141] time: strptime %z offset restriction
2019 [16171] math: drem should be alias of remainder
2020 [16296] math: fegetround is pure?
2021 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2022 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2024 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2025 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2027 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2028 "Exception "Inexact" set" on ppc32
2029 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2030 [16517] math: Missing underflow exception from tanf/tan/tanl
2031 [16519] math: Missing underflow exception from sinhf
2032 [16520] math: Missing underflow exception from tanhf
2033 [16521] math: Missing underflow exception from exp2
2034 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2036 [16734] stdio: fopen calls mmap to allocate its buffer
2037 [16961] math: nan function incorrect handling of bad sequences
2038 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2039 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2040 [16985] locale: localedef: confusing error message when opening output
2042 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2043 [17197] locale: Redundant shift character in iconv conversion output at
2045 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2046 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2047 [17250] dynamic-link: static linking breaks nss loading
2048 (getaddrinfo/getpwnam/etc...)
2049 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2051 [17441] math: isnan() should use __builtin_isnan() in GCC
2052 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2053 (related to lock elision)
2054 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2055 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2056 [17887] time: strptime should be able to parse "+01:00" style timezones
2057 [17905] libc: catopen() Multiple unbounded stack allocations
2059 [18084] libc: backtrace (..., 0) dumps core on x86
2060 [18086] libc: nice() sets errno to 0 on success
2061 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2062 is too large (CVE-2015-8778)
2063 [18251] dynamic-link: SONAME missing when audit modules provides path
2064 [18265] libc: add attributes for wchar string and memory functions
2065 [18370] math: csqrt missing underflows
2066 [18421] libc: [hppa] read-only segment has dynamic relocations
2067 [18472] libc: Obsolete syscall wrappers should be compat symbols
2068 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2069 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2070 [18525] localedata: Remove locale timezone information
2071 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2072 [18568] localedata: Update locale data to Unicode 8.0
2073 [18589] locale: sort-test.sh fails at random
2074 [18595] math: ctan, ctanh missing underflows
2075 [18604] libc: assert macro-expands its argument
2076 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2077 contains a vector instruction exception.
2078 [18611] math: j1, jn missing errno setting on underflow
2079 [18618] localedata: sync Chechen locale definitions with other *_RU
2081 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2082 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2083 [18665] network: In send_dg, the recvfrom function is NOT always using the
2084 buffer size of a newly created buffer (CVE-2015-7547)
2085 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2087 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2089 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2090 [18699] math: tilegx cproj() for various complex infinities does not yield
2092 [18724] libc: Harden put*ent functions against data injection
2093 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2094 [18755] build: build errors with -DNDEBUG
2095 [18757] stdio: fmemopen fails to set errno on failure
2096 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2098 [18781] libc: openat64 lacks O_LARGEFILE
2099 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2100 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2101 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2102 [18790] math: [ldbl-128ibm] tanhl inaccurate
2103 [18795] libc: stpncpy fortification misses buffer lengths that are
2104 statically too large
2105 [18796] build: build fails for --disable-mathvec
2106 [18803] math: hypot missing underflows
2107 [18820] stdio: fmemopen may leak memory on failure
2108 [18823] math: csqrt spurious underflows
2109 [18824] math: fma spurious underflows
2110 [18825] math: pow missing underflows
2111 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2112 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2113 [18870] build: sem_open.c fails to compile with missing symbol
2115 [18872] stdio: Fix memory leak in printf_positional
2116 [18873] libc: posix_fallocate overflow check ineffective
2117 [18875] math: Excess precision leads incorrect libm
2118 [18877] libc: arm: mmap offset regression
2119 [18887] libc: memory corruption when using getmntent on blank lines
2120 [18918] localedata: hu_HU: change time to HH:MM:SS format
2121 [18921] libc: Regression: extraneous stat() and fstat() performed by
2123 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2124 binaries (CVE-2015-8777)
2125 [18951] math: tgamma missing underflows
2126 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2128 [18953] localedata: lt_LT: change currency symbol to the euro
2129 [18956] math: powf inaccuracy
2130 [18961] math: [i386] exp missing underflows
2131 [18966] math: [i386] exp10 missing underflows
2132 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2133 [18969] build: multiple string test failures due to missing locale
2135 [18970] libc: Reference of pthread_setcancelstate in libc.a
2136 [18977] math: float / long double Bessel functions not in XSI POSIX
2137 [18980] math: i386 libm functions return with excess range and precision
2138 [18981] math: i386 scalb*, ldexp return with excess range and precision
2139 [18982] stdio: va_list and vprintf
2140 [18985] time: Passing out of range data to strftime() causes a segfault
2142 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2143 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2144 [19012] locale: iconv_open leaks memory on error path
2145 [19016] math: clog, clog10 inaccuracy
2146 [19018] nptl: Mangle function pointers in tls_dtor_list
2147 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2148 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2149 [19048] malloc: malloc: arena free list can become cyclic, increasing
2151 [19049] math: [powerpc] erfc incorrect zero sign
2152 [19050] math: [powerpc] log* incorrect zero sign
2153 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2154 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2155 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2156 [19074] network: Data race in _res_hconf_reorder_addrs
2157 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2158 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2159 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2161 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2163 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2164 [19086] manual: posix_fallocate64 documented argument order is wrong.
2165 [19088] math: lround, llround missing exceptions close to overflow
2167 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2168 [19095] math: dbl-64 lrint incorrect for 64-bit long
2169 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2170 [19124] dynamic-link: ld.so failed to build with older assmebler
2171 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2172 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2173 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2174 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2175 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2176 pthread_setaffinity_np
2177 [19156] math: [ldbl-128] j0l spurious underflows
2178 [19164] nptl: tst-getcpu fails with many possible CPUs
2179 [19168] math: math/test-ildoubl and math/test-ldouble failure
2180 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2181 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2183 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2184 (FE_NOMASK_ENV) do not clear SSE exceptions
2185 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2186 _int_new_arena/reused_arena
2187 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2188 [19201] math: dbl-64 remainder incorrect sign of zero result
2189 [19205] math: bits/math-finite.h conditions do not match math.h and
2191 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2192 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2193 for C99-based standards
2194 [19212] libc: features.h not -Wundef clean
2195 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2197 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2198 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2199 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2201 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2202 "inexact" exceptions
2203 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2205 [19242] libc: strtol incorrect in Turkish locales
2206 [19243] malloc: reused_arena can pick an arena on the free list, leading
2207 to an assertion failure and reference count corruption
2208 [19253] time: tzset() ineffective when temporary TZ did not include DST
2210 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2211 [19270] math: [hppa] Shared libm missing __isnanl
2212 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2213 MAP_HUGETLB and MAP_STACK defines
2214 [19313] nptl: Wrong __cpu_mask for x32
2215 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2216 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2217 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2218 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2219 [19363] time: x32: times() return value wrongly truncates/sign extends
2221 [19367] dynamic-link: Improve branch prediction on Silvermont
2222 [19369] network: Default domain name not reset by res_ninit when "search"
2223 / "domain" entry is removed from resolv.conf
2224 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2226 [19385] localedata: bg_BG: time separator should be colon, not comma
2227 [19408] libc: linux personality syscall wrapper may erroneously return an
2228 error on 32-bit architectures
2229 [19415] libc: dladdr returns wrong names on hppa
2230 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2231 IBM905, IBM907, and IBM909
2232 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2233 [19443] build: build failures with -DDEBUG
2234 [19451] build: Make check fails on test-double-vlen2
2235 [19462] libc: Glibc failed to build with -Os
2236 [19465] math: Wrong code with -Os
2237 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2239 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2241 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2242 [19486] math: S390: Math tests fail with "Exception Inexact set".
2243 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2244 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2246 [19590] math: Fail to build shared objects that use libmvec.so functions.
2250 * The following bugs are resolved with this release:
2252 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2253 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2254 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2255 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2256 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2257 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2258 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2259 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2260 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2261 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2262 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2263 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2264 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2265 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2266 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2267 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2268 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2269 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2270 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2271 18657, 18676, 18694, 18696, 18887.
2273 * Cache information can be queried via sysconf() function on s390 e.g. with
2274 _SC_LEVEL1_ICACHE_SIZE as argument.
2276 * A buffer overflow in gethostbyname_r and related functions performing DNS
2277 requests has been fixed. If the NSS functions were called with a
2278 misaligned buffer, the buffer length change due to pointer alignment was
2279 not taken into account. This could result in application crashes or,
2280 potentially arbitrary code execution, using crafted, but syntactically
2281 valid DNS responses. (CVE-2015-1781)
2283 * The time zone file parser has been made more robust against crafted time
2284 zone files, avoiding heap buffer overflows related to the processing of
2285 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2286 large time zone data files. Overly long time zone specifiers in the TZ
2287 variable no longer result in stack overflows and crashes.
2289 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2290 for LD and GD on x86 and x86-64, has been implemented. You will need
2291 binutils-2.24 or later to enable this optimization.
2293 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2294 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2295 Hat). These updates cause user visible changes, such as the fix for bug
2298 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2299 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2300 condition in some applications.
2302 * Added vector math library named libmvec with the following vectorized x86_64
2303 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2305 The library can be disabled with --disable-mathvec. Use of the functions is
2306 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2307 Shared library libmvec.so is linked in as needed when using -lm (no need to
2308 specify -lmvec explicitly for not static builds).
2309 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2311 * A new fmemopen implementation has been added with the goal of POSIX
2312 compliance. The new implementation fixes the following long-standing
2313 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2314 old implementation is still present for use be by existing binaries.
2316 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2317 releases. It has been fixed to match 2.19 and older, but binaries built
2318 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2320 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2321 Contributed by Roland McGrath (Google).
2323 * The header <regexp.h> is deprecated, and will be removed in a future
2324 release. Use of this header will trigger a deprecation warning.
2325 Application developers should update their code to use <regex.h> instead.
2327 This header was formerly part of SUS, but was deprecated in 1994 and
2328 removed from the standard in 2001. Also, the glibc implementation
2329 leaks memory. See BZ#18681 for more details.
2333 * The following bugs are resolved with this release:
2335 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2336 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2337 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2338 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2339 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2340 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2341 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2342 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2343 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2344 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2347 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2348 too little memory for the to-be-scanned arguments and overflow the
2349 allocated buffer. The implementation now correctly computes the required
2350 buffer size when using malloc, and switches to malloc from alloca as
2353 * A new semaphore algorithm has been implemented in generic C code for all
2354 machines. Previous custom assembly implementations of semaphore were
2355 difficult to reason about or ensure that they were safe. The new version
2356 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2357 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2358 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2360 * Port to Altera Nios II has been contributed by Mentor Graphics.
2362 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2363 implementations for powerpc64/powerpc64le.
2364 Implemented by Adhemerval Zanella (IBM).
2366 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2367 and powerpc64le. This may improve lock scaling of existing programs on
2368 HTM capable systems. The lock elision code is only enabled with
2369 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2370 powerpc will issue a transaction abort on every syscall to avoid side
2371 effects being visible outside transactions.
2373 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2374 AArch64. Contributed by ARM Ltd.
2376 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2378 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2379 under certain input conditions resulting in the execution of a shell for
2380 command substitution when the application did not request it. The
2381 implementation now checks WRDE_NOCMD immediately before executing the
2382 shell and returns the error WRDE_CMDSUB as expected.
2384 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2385 processing format strings with a large number of format specifiers.
2387 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2388 infinite loop if the DNS response contained a PTR record of an unexpected
2391 * The minimum GCC version that can be used to build this version of the GNU
2392 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2393 still be used to compile programs using the GNU C Library.
2395 * The GNU C Library is now built with -Werror by default. This can be
2396 disabled by configuring with --disable-werror.
2398 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2400 * The obsolete sigvec function has been removed. This was the original
2401 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2402 programs have been using instead for about 25 years. Of course, ABI
2403 compatibility for old binaries using sigvec remains intact.
2405 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2406 with newer versions of bison.
2408 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2409 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2410 registers overlay two consecutive single-precision registers. MIPS32R2
2411 introduced a new FPU mode (FR=1) where double-precision registers extend the
2412 corresponding single-precision registers which is incompatible with the
2413 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2414 require the use of FR=1 making a transition necessary. New o32 ABI
2415 extensions enable users to migrate over time from the original o32 ABI
2416 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2417 tracks the ABI of any loaded object and verifies that new objects are
2418 compatible. Mode transitions will also be requested as required and
2419 unsupportable objects will be rejected. The ABI checks include both soft and
2420 hard float ABIs for o32, n32 and n64.
2422 GCC 5 with GNU binutils 2.25 onwards:
2423 It is strongly recommended that all o32 system libraries are built using the
2424 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2425 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2426 compiler using --with-fp-32=xx to set this by default.
2430 * The following bugs are resolved with this release:
2432 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2433 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2434 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2435 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2436 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2437 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2438 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2439 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2440 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2441 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2442 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2443 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2444 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2445 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2446 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2448 * Reverted change of ABI data structures for s390 and s390x:
2449 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2450 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2451 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2452 are preserved pointing straight to the same implementation as the old ones.
2453 Given that, new callers will simply provide a too-big buffer to these
2454 functions. Any applications/libraries out there that embed jmp_buf or
2455 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2456 against 2.19 headers will have to rebuilt again. This is necessary in any
2457 case to revert the breakage in their ABI caused by the glibc change.
2459 * Support for file description locks is added to systems running the
2460 Linux kernel. The standard file locking interfaces are extended to
2461 operate on file descriptions, not file descriptors, via the use of
2462 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2463 are associated with an open file instead of a process.
2465 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2467 * The minimum Linux kernel version that this version of the GNU C Library
2468 can be used with is 2.6.32.
2470 * Running the testsuite no longer terminates as soon as a test fails.
2471 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2472 with PASS or FAIL lines for individual tests. A summary of the results is
2473 printed, including a list of failing lists, and "make check" exits with
2474 error status if there were any unexpected failures. "make check
2475 stop-on-test-failure=y" may be used to keep the old behavior.
2477 * The am33 port, which had not worked for several years, has been removed
2480 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2481 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2482 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2483 conflicted with POSIX (support for which was removed in 2.19), the
2484 interfaces those macros enabled remain available when compiling with
2485 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2486 test macros defined.
2488 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2490 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2491 This may improve lock scaling of existing programs on TX capable systems.
2492 The lock elision code is only built with --enable-lock-elision=yes and
2493 then requires a GCC version supporting the TX builtins. With lock elision
2494 default mutexes are elided via __builtin_tbegin, if the cpu supports
2495 transactions. By default lock elision is not enabled and the elision code
2498 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2499 copy the path argument. This allowed programs to cause posix_spawn to
2500 deference a dangling pointer, or use an unexpected pathname argument if
2501 the string was modified after the posix_spawn_file_actions_addopen
2504 * All supported architectures now use the main glibc sysdeps directory
2505 instead of some being in a separate "ports" directory (which was
2506 distributed separately before glibc 2.17).
2508 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2509 On configurations that support it (all Linux configurations), it's now
2510 used regardless of the --enable-add-ons switch to configure. It is no
2511 longer possible to build such configurations without pthreads support.
2513 * Locale names, including those obtained from environment variables (LANG
2514 and the LC_* variables), are more tightly checked for proper syntax.
2515 setlocale will now fail (with EINVAL) for locale names that are overly
2516 long, contain slashes without starting with a slash, or contain ".." path
2517 components. (CVE-2014-0475) Previously, some valid locale names were
2518 silently replaced with the "C" locale when running in AT_SECURE mode
2519 (e.g., in a SUID program). This is no longer necessary because of the
2522 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2523 with application code using Intel MPX instructions. (With all previous
2524 versions, the MPX register state could be clobbered when making calls
2525 into or out of a shared library.) Note that while the new dynamic
2526 linker is compatible with all known x86 hardware whether or not it
2527 supports Intel MPX, some x86 instruction-set emulators might fail to
2528 handle the new instruction encodings. This is known to affect Valgrind
2529 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2530 release), and might affect other tools that do instruction emulation.
2532 * Support for loadable gconv transliteration modules has been removed.
2533 The support for transliteration modules has been non-functional for
2534 over a decade, and the removal is prompted by security defects. The
2535 normal gconv conversion modules are still supported. Transliteration
2536 with //TRANSLIT is still possible, and the //IGNORE specifier
2537 continues to be supported. (CVE-2014-5119)
2539 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2540 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2541 resulting a denial-of-service security vulnerability in applications which
2542 use functions related to iconv. (CVE-2014-6040)
2546 * The following bugs are resolved with this release:
2548 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2549 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2550 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2551 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2552 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2553 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2554 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2555 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2556 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2557 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2558 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2559 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2560 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2561 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2562 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2563 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2564 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2565 16431, 16453, 16474, 16506, 16510, 16529
2567 * Slovenian translations for glibc messages have been contributed by the
2568 Translation Project's Slovenian team of translators.
2570 * The public headers no longer use __unused nor __block. This change is to
2571 support compiling programs that are derived from BSD sources and use
2572 __unused internally, and to support compiling with Clang's -fblock
2573 extension which uses __block.
2575 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2576 large collation sequences to optimize multiple passes. This cache
2577 computation may overflow for large collation sequences and may cause a
2578 stack or buffer overflow. This is now fixed to use a slower algorithm
2579 which does not use a cache if there is an integer overflow.
2581 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2582 rules for large collation sequences to optimize multiple passes and falls
2583 back to alloca if malloc fails, resulting in a possible stack overflow.
2584 The implementation now falls back to an uncached collation sequence lookup
2587 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2588 initialized for static applications resulting in the security feature
2589 being disabled. The pointer guard is now correctly initialized to a
2590 random value for static applications. Existing static applications need
2591 to be recompiled to take advantage of the fix (bug 15754).
2593 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2594 to the d_name member of struct dirent, or omit the terminating NUL
2595 character. (Bugzilla #14699).
2597 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2598 aligned_alloc functions could allocate too few bytes or corrupt the
2599 heap when passed very large allocation size values (Bugzilla #15855,
2602 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2603 for AF_INET6 has been fixed (Bugzilla #16072).
2605 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2606 pap_CW, quz_PE, the_NP.
2608 * Substantially revised locales: gd_GB, ht_HT
2610 * The LC_ADDRESS field was updated to support country_car for almost all
2613 * ISO 1427 definitions were updated.
2615 * ISO 3166 definitions were updated.
2617 * The localedef utility now supports --big-endian and --little-endian
2618 command-line options to generate locales for a different system from that
2619 for which the C library was built.
2621 * Binary locale files now only depend on the endianness of the system for
2622 which they are generated and not on other properties of that system. As a
2623 consequence, binary files generated with new localedef may be incompatible
2624 with old versions of the GNU C Library, and binary files generated with
2625 old localedef may be incompatible with this version of the GNU C Library,
2626 in the following circumstances:
2628 + Locale files may be incompatible on m68k systems.
2630 + Locale archive files (but not separate files for individual locales) may
2631 be incompatible on systems where plain "char" is signed.
2633 * The configure option --disable-versioning has been removed. Builds with
2634 --disable-versioning had not worked for several years.
2636 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2637 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2639 * SystemTap probes for malloc have been introduced.
2641 * SystemTap probes for slow multiple precision fallback paths of
2642 transcendental functions have been introduced.
2644 * Support for powerpc64le has been added.
2646 * The soft-float powerpc port now supports e500 processors.
2648 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2650 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2651 set of header declarations that are enabled by default, even when other
2652 feature test macros or compiler options such as -std=c99 would otherwise
2653 disable some of those declarations.
2655 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2656 conflict with POSIX. The libbsd-compat library (which was a dummy library
2657 that did nothing) has also been removed.
2659 * Preliminary documentation about Multi-Thread, Async-Signal and
2660 Async-Cancel Safety has been added.
2662 * Change of ABI data structures for s390 and s390x:
2663 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2664 allow for future hardware extensions. All library functions that accept or
2665 return these structures were versioned in order to provide backwards
2666 compatibility with existing code. However, not all cases can be handled
2667 with interface versioning. If an application embeds either structure into
2668 another structure and that structure is passed to another compilation unit
2669 which expects the newer larger-sized object then any access to the new
2670 fields will result in undefined behaviour. Similarly any access to fields
2671 that were moved by the enlarging of the structures will cause undefined
2672 behaviour. To fix the undefined behaviour all objects that are part of the
2673 final in-memory image must agree on the size of structures, and this may
2674 require recompilation.
2678 * The following bugs are resolved with this release:
2680 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2681 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2682 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2683 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2684 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2685 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2686 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2687 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2688 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2689 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2690 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2691 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2692 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2695 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2696 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2697 Distributions can re-enable building and using pt_chown via the new configure
2698 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2699 considerable security risks and should only be used if the distribution
2700 understands and accepts the risks.
2702 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2705 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2706 fixed (Bugzilla #15330).
2708 * Add support for calling C++11 thread_local object destructors on thread
2709 and program exit. This needs compiler support for offloading C++11
2710 destructor calls to glibc.
2712 * Improved worst case performance of libm functions with double inputs and
2715 * Support for priority inherited mutexes in pthread condition variables on
2716 non-x86 architectures.
2718 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2720 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2722 * Optimized string functions for ARM. Implemented by Will Newton and
2725 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2727 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2730 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2731 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2733 * Added a benchmark framework to track performance of functions in glibc.
2735 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2736 It is based on draft TS 18661 and currently enabled as a GNU extension.
2738 * On Linux, the clock function now uses the clock_gettime system call
2739 for improved precision, rather than old times system call.
2741 * Added support for version-3 tz format files. This is needed when using
2742 the tz database release 2013e or later, and affects a few unusual cases --
2743 currently only TZ='America/Godthab' for time stamps after 2037.
2745 * Added new API functions pthread_getattr_default_np and
2746 pthread_setattr_default_np to get and set the default pthread
2747 attributes of a process.
2749 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2750 This may improve lock scaling of existing programs on TSX capable systems.
2751 When the --enable-lock-elision=yes parameter is specified at configure
2752 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2755 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2756 availability). Implemented by Ryan S. Arnold.
2758 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2762 * The following bugs are resolved with this release:
2764 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2765 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2766 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2767 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2768 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2769 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2770 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2771 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2772 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2773 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2774 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2775 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2776 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2777 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2778 14879, 14889, 14893, 14898, 14914.
2780 * Optimization of memcpy for MIPS.
2782 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2783 EMFILE has been fixed (Bugzilla #14889).
2785 * The add-on ports collection is now distributed in the "ports" subdirectory
2786 of the main GNU C Library distribution, rather than separately.
2788 * Port to ARM AArch64 contributed by Linaro.
2790 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2791 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2793 Implemented by Andreas Krebbel.
2795 * The new function secure_getenv allows secure access to the environment,
2796 returning NULL if running in a SUID/SGID process. This function replaces
2797 the internal function __secure_getenv.
2799 * SystemTap static probes have been added into the dynamic linker.
2800 Implemented by Gary Benson.
2802 * Optimizations of string functions strstr, strcasestr and memmem.
2803 Implemented by Maxim Kuvyrkov.
2805 * The minimum Linux kernel version that this version of the GNU C Library
2806 can be used with is 2.6.16.
2808 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2809 powerpc POWER7. Implemented by Will Schmidt.
2811 * New configure option --disable-nscd builds the C library such that it
2812 never attempts to contact the Name Service Caching Daemon (nscd).
2813 New configure option --disable-build-nscd avoids building nscd itself;
2814 this is the default if --disable-nscd is used.
2816 * Improved support for cross-compilation, including cross-testing and
2817 bootstrap builds without a previously built glibc.
2819 * Several testsuite tests are now able to test multiple IFUNC variants of an
2820 interface, rather than just testing the one that would be chooen by
2823 * New configure options --with-bugurl and --with-pkgversion, for
2824 distributors to use to embed their bug-reporting and package version
2825 information in --help and --version output.
2827 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2828 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2829 allows creation of chroots without the procfs mounted on /proc.
2831 * The `crypt' function now fails if passed salt bytes that violate the
2832 specification for those values. On Linux, the `crypt' function will
2833 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2834 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2835 when the mode is enabled.
2837 * The `clock_*' suite of functions (declared in <time.h>) is now available
2838 directly in the main C library. Previously it was necessary to link with
2839 -lrt to use these functions. This change has the effect that a
2840 single-threaded program that uses a function such as `clock_gettime' (and
2841 is not linked with -lrt) will no longer implicitly load the pthreads
2842 library at runtime and so will not suffer the overheads associated with
2843 multi-thread support in other code such as the C++ runtime library.
2845 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2850 * The following bugs are resolved with this release:
2852 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2853 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2854 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2855 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2856 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2857 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2858 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2859 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2860 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2861 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2862 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2863 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2864 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2865 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2866 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2867 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2868 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2869 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2870 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2871 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2872 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2873 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2876 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2877 configuring glibc with:
2878 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2879 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2880 Implemented by H.J. Lu.
2884 + define static_assert
2886 + do not declare gets
2888 + declare at_quick_exit and quick_exit also for ISO C11
2890 + aligned_alloc. NB: The code is deliberately allows the size parameter
2891 to not be a multiple of the alignment. This is a moronic requirement
2892 in the standard but it is only a requirement on the caller, not the
2895 + timespec_get added
2897 + uchar.h support added
2899 + CMPLX, CMPLXF, CMPLXL added
2901 Implemented by Ulrich Drepper.
2903 * Support for the IA-64 has been moved to ports.
2905 * Remove support for anything but ELF binary format
2907 * Checking versions of poll, ppoll added.
2908 Implemented by Ulrich Drepper.
2910 * More generic and 64-bit performance optimizations to math functions.
2911 Implemented by Ulrich Drepper.
2913 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2914 headers and functions available at compile time as they were before
2915 version 2.14. This option will be removed at some time in the future
2916 after the TI-RPC library becomes fully sufficient for the needs of
2917 existing applications.
2919 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2920 Note that glibc is not expected to work with any Linux kernel version
2923 * New header <sys/auxv.h> and function getauxval allowing easy access to
2924 the AT_* key-value pairs passed from the Linux kernel. The header also
2925 defines the HWCAP_* bits associated with the AT_HWCAP key.
2927 * New locales: mag_IN
2929 * New configure option --enable-systemtap builds SystemTap static probes
2930 into libc for setjmp and longjmp and into libpthread for various operations.
2931 So far the setjmp/longjmp probes and some of the libpthread probes are
2932 provided only for i*86 and x86_64.
2933 Implemented by Roland McGrath and Rayson Ho.
2935 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2937 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2940 * More optimized functions for SPARC. Implemented by David S. Miller.
2942 * Improved support for cross-compilation, especially bootstrap builds
2943 without a previously built glibc.
2945 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2946 Chris Metcalf from Tilera.
2948 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2949 now supported for ARM processors.
2951 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2952 as the name of the dynamic linker, to distinguish it from the
2953 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2955 * Support for CRIS has been removed from ports.
2957 * A new class of installed header has been documented for low-level
2958 platform-specific functionality. PowerPC added the first instance with a
2959 function to provide time base register access. Contributed by Tulio
2960 Magno Quites Machado Filho.
2962 * ix86 configurations now install header files that are consistent with
2963 what x86-64 configurations install. These same header files can be used
2964 for -m32, -m64, or -mx32 builds.
2965 Contributed by H.J. Lu.
2967 * Math library bug fixes. A thorough audit of all open math library bugs was
2968 conducted by Joseph Myers. Significant progress was made on many math
2969 library bugs resulting in more accurate exceptions and function results.
2970 Many thanks to all those that contributed including Andreas Jaeger for his
2971 patch review and work on the x87 trigonometric instruction issues.
2973 * Timezone data is no longer installed. Timezone-related binaries and scripts
2974 will continue to be installed. Users should obtain their timezone data from
2975 their distribution provider or from the tzdata package at
2976 <ftp://munnari.oz.au/pub/>.
2980 * The following bugs are resolved with this release:
2982 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2983 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2984 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2985 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2986 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2987 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2988 13484, 13506, 13515, 13523, 13524, 13538, 13540
2990 * New program pldd to list loaded object of a process
2991 Implemented by Ulrich Drepper.
2993 * Add nss_db support back to glibc. No more dependency on Berkeley db
2994 and support for initgroups lookups.
2995 Implemented by Ulrich Drepper.
2997 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2998 Contributed by HJ Lu.
3000 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3001 Contributed by HJ Lu.
3003 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3004 on x86-32 and x86-64.
3005 Contributed by Liubov Dmitrieva.
3007 * Optimized strchr and strrchr for SSE on x86-32.
3008 Contributed by Liubov Dmitrieva.
3010 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3011 for x86-64 and x86-32.
3012 Contributed by Liubov Dmitrieva.
3014 * New interfaces: scandirat, scandirat64
3015 Implemented by Ulrich Drepper.
3017 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3018 Implemented by Ulrich Drepper.
3020 * nscd now also caches the netgroup database.
3021 Implemented by Ulrich Drepper.
3023 * Integrate libm with gcc's -ffinite-math-only option.
3024 Implemented by Ulrich Drepper.
3026 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3027 to math functions. Implemented by Ulrich Drepper.
3029 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3030 Implemented by Ulrich Drepper.
3032 * New Linux interfaces: process_vm_readv, process_vm_writev
3034 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3035 Implemented by Ulrich Drepper.
3037 * Optimized nearbyint and strcasecmp for PPC.
3038 Implemented by Adhemerval Zanella.
3040 * New locales: bho_IN, unm_US, es_CU, ta_LK
3044 * The following bugs are resolved with this release:
3046 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3047 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3048 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3049 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3050 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3051 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3052 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3053 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3054 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3055 12795, 12811, 12813, 12814, 12841
3057 * The RPC implementation in libc is obsoleted. Old programs keep working
3058 but new programs cannot be linked with the routines in libc anymore.
3059 Programs in need of RPC functionality must be linked against TI-RPC.
3060 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3062 Visible changes of this change include (obviously) the inability to link
3063 programs using RPC functions without referencing the TI-RPC library and the
3064 removal of the RPC headers from the glibc headers.
3065 Implemented by Ulrich Drepper.
3067 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3068 syncfs, setns, sendmmsg
3070 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3071 yue_HK, lij_IT, mhr_RU
3073 * New encodings: CP770, CP771, CP772, CP773, CP774
3075 * New program sotruss to trace calls through PLTs
3076 Implemented by Ulrich Drepper.
3078 * The malloc hook implementation is marked deprecated and will be removed
3079 from the default implementation in the next version. The design never
3080 worked ever since the introduction of threads. Even programs which do
3081 not create threads themselves can use multiple threads created internally.
3085 * The following bugs are resolved with this release:
3087 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3088 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3089 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3090 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3093 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3095 * POWER7 optimizations: memset, memcmp, strncmp
3097 * New optimized string functions for x86-64: strnlen (SSE2),
3098 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3099 Implemented by Ulrich Drepper.
3103 * The following bugs are resolved with this release:
3105 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3106 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3107 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3108 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3109 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3110 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3111 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3112 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3114 * New interfaces: pthread_getname_np, pthread_setname_np
3116 * New Linux interface: recvmmsg
3118 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3120 * The dynamic linker now recognizes supported ABI versions from the
3121 EI_ABIVERSION field in the ELF header.
3122 Implemented by Ulrich Drepper.
3124 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3125 will not be used to fill in password fields in the passwd.byname replies.
3126 Instead it is used to synthesize the shadow.byname table, should it be
3127 missing. This is a useful mode in some installations involving Solaris.
3128 Implemented by Ulrich Drepper.
3130 * New locales: kok_IN, sq_MK, cv_RU
3134 * The following bugs are resolved with this release:
3136 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3137 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3138 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3139 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3140 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3141 10789, 10817, 10823, 10840, 10847
3143 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3145 Implemented by Ulrich Drepper.
3147 * Checking version of longjmp added that fails if an uninitialized stack
3148 frame would be created. Implemented by Ulrich Drepper.
3150 * STT_GNU_IFUNC is now supported in static executables.
3151 Implemented by H.J. Lu.
3153 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3155 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3156 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3157 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3158 Contributed by H.J. Lu.
3160 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3161 Implemented by Ulrich Drepper.
3163 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3165 Contributed by H.J. Lu.
3167 * Support for fma instruction in AVX on x86-64.
3168 Implemented by H.J. Lu and Ulrich Drepper.
3170 * AVX support in x86-64 auditing support in ld.so.
3171 Implemented by H.J. Lu.
3173 * STB_GNU_UNIQUE support added.
3174 Implemented by Ulrich Drepper.
3176 * Implement second fallback mode for optimized DNS lookup for even more
3177 broken environments. If two requests from the same port are not
3178 handled correctly close the socket and open a new one before sending
3179 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3180 can be used to select this mode right away, instead of rediscovering the
3181 necessity is every process again.
3182 Implemented by Ulrich Drepper.
3184 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3185 Implemented by Adam Tkac.
3187 * Optimized iconv conversions for S390x.
3188 Implemented by Andreas Krebbel.
3190 * Using condvars with PI mutexes is now more efficient due to kernel
3191 support for requeueing to PI futexes. NPTL support added for x86-64.
3192 Implemented by Ulrich Drepper.
3194 * New locales: ps_AF, my_MM
3198 * The following bugs are resolved with this release:
3200 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3201 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3202 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3203 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3204 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3206 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3207 Implemented by Ulrich Drepper.
3209 * New Linux interfaces: accept4, fallocate, fallocate64.
3210 Implemented by Ulrich Drepper.
3212 * Correct declarations of string function when used in C++ code. This
3213 could lead to compile errors for invalid C++ code.
3215 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3218 * New POSIX 2008 interface: psiginfo
3219 Implemented by Ulrich Drepper.
3221 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3222 Implemented by Ulrich Drepper.
3224 * Support for selecting between multiple function definitions at runtime
3225 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3227 * The libcrypt library can now use the hash function implementations in
3228 NSS. Implemented by Ulrich Drepper.
3230 * The malloc implementation can be compiled to be less memory efficient
3231 but higher performing in multi-threaded programs.
3232 Implemented by Ulrich Drepper.
3234 * New locales: nan_TW@latin, ks_IN
3236 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3237 Implemented by Ulrich Drepper.
3239 * Extended printf hook support. It is possible to use user-defined types
3240 and extend existing format specifiers.
3241 Implemented by Ulrich Drepper.
3243 * Handling for group shadow files has been added.
3244 Implemented by Ulrich Drepper.
3246 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3247 servers (the case, e.g., for some people using the built-in DNS
3248 server in ADSL modems/routers). There is a once-per-process timeout
3249 in case of a broken server. To avoid it, users can run nscd or put
3250 'options single-request' in /etc/resolv.conf.
3251 Implemented by Ulrich Drepper.
3255 * The following bugs are resolved with this release:
3257 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3258 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3259 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3260 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3261 6968, 6974, 6980, 6995, 7008, 7009, 7029
3263 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3264 up at the same time. Implemented by Ulrich Drepper.
3266 * TLS descriptors for LD and GD on x86 and x86-64.
3267 Implemented by Alexandre Oliva.
3269 * getaddrinfo now handles DCCP and UDPlite.
3270 Implemented by Ulrich Drepper.
3272 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3273 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3274 Implemented by Ulrich Drepper.
3276 * New implementation of memmem, strstr, and strcasestr which is O(n).
3277 Implemented by Eric Blake.
3279 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3281 * Implement "e" option for popen to open file descriptor with the
3282 close-on-exec flag set. Implemented by Ulrich Drepper.
3284 * Implement "b" mode for fmemopen. In this mode writes writes don't
3285 implicitly add a NUL byte and seeks from the end of the buffer really
3286 use the buffer end, not the string length as the basis.
3287 Implemented by Ulrich Drepper.
3289 * Many functions, exported and internal, now atomically set the close-on-exec
3290 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3292 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3294 Implemented by Pravin Satpute.
3296 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3300 * The following bugs are resolved with this release:
3302 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3303 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3304 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3305 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3306 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3307 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3308 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3310 * New locales: bo_CN, bo_IN, shs_CA.
3312 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3314 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3315 Implemented by Pravin Satpute.
3317 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3319 * nscd caches now all timeouts for DNS entries
3320 Implemented by Ulrich Drepper.
3322 * nscd is more efficient and wakes up less often.
3323 Implemented by Ulrich Drepper.
3325 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3326 vdprintf, and obstack_vprintf.
3327 Implemented by Jakub Jelinek.
3329 * Faster memset for x86-64.
3330 Implemented by Harsha Jagasia and H.J. Lu.
3332 * Faster memcpy on x86.
3333 Implemented by Ulrich Drepper.
3335 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3336 Implemented by Ulrich Drepper.
3338 * Faster sqrt and sqrtf implemention for some PPC variants.
3339 Implemented by Steven Munroe.
3343 * The following bugs are resolved with this release:
3345 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3346 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3347 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3348 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3349 5071, 5103, 5104, 5112, 5113, 5184, 5186
3351 * More checking functions: fread, fread_unlocked, open*, mq_open.
3352 Implemented by Jakub Jelinek and Ulrich Drepper.
3354 * Extend fortification to C++. Implemented by Jakub Jelinek.
3356 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3357 by not recognizing 'a' as a modifier when those specs are requested.
3358 Implemented by Jakub Jelinek.
3360 * PPC optimizations to math and string functions.
3361 Implemented by Steven Munroe.
3363 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3364 options to be passed. Implemented by Ulrich Drepper.
3366 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3368 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3369 Implemented by Ulrich Drepper.
3371 * Handle private futexes in the NPTL implementation.
3372 Implemented by Jakub Jelinek and Ulrich Drepper.
3374 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3375 Implemented by Roland McGrath and Ulrich Drepper.
3377 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3379 * SHA-256 and SHA-512 based password encryption.
3380 Implemented by Ulrich Drepper.
3382 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3383 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3386 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3387 Implemented by Ulrich Drepper.
3391 * The following bugs are resolved with this release:
3393 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3394 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3395 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3396 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3397 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3398 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3399 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3402 * New Linux interfaces: epoll_pwait, sched_getcpu.
3404 * New generic interfaces: strerror_l.
3406 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3410 * The following bugs are resolved with this release:
3412 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3413 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3414 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3415 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3416 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3417 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3418 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3419 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3420 3225, 3251, 3252, 3253, 3273
3422 * For Linux, the sorting of addresses returned by getaddrinfo now also
3423 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3424 Implemented by Ulrich Drepper.
3426 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3427 Implemented by Ulrich Drepper.
3429 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3431 * New iconv module for MIK. Contributed by Alexander Shopov.
3433 * For sites with broken group and/or passwd database, the auto-propagate
3434 option of nscd can prevent creating ID lookup entries from the results
3435 of a name lookup and vice versa. This usually is no problem but some
3436 site might have problems with the default behavior.
3437 Implemented by Ulrich Drepper.
3439 * Iterating over entire database in NIS can be slow. With the
3440 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3441 to trade time for memory. The entire database will be read at once.
3442 Implemented by Ulrich Drepper.
3444 * The interfaces introduced in RFC 3542 have been implemented by
3447 * Support for the new ELF hash table format was added by Ulrich Drepper.
3449 * Support for priority inheritance mutexes added by Jakub Jelinek and
3452 * Support for priority protected mutexes added by Jakub Jelinek.
3454 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3455 or_IN, csb_PL, fy_NL, sr_ME.
3459 * More overflow detection functions.
3461 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3462 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3464 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3465 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3466 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3467 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3468 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3469 by Masahide Washizawa.
3471 * It is now possible to install audit libraries for the dynamic linker using
3472 LD_AUDIT. Implemented by Ulrich Drepper.
3474 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3475 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3476 For a libc and libpthread that works well on Linux 2.4 kernels, we
3477 recommend using the stable 2.3 branch.
3479 * The new function open_wmemstream acts like open_memstream,
3480 but uses a wchar_t wide character buffer.
3482 * The new function ppoll is an improved version of poll, similar to pselect.
3484 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3485 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3486 renameat, unlinkat, linkat, symlinkat, readlinkat.
3488 * New Linux kernel system calls: unshare,
3489 inotify_init, inotify_add_watch, inotify_rm_watch.
3491 * The euidaccess function is now also known by the alias eaccess,
3492 for compatibility with some other systems.
3494 * Timezone data updated to 2006b version.
3498 * The following bugs are resolved with this release:
3500 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3501 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3502 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3503 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3504 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3505 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3507 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3509 * As of this release, GCC 4 can be used to compile the C Library.
3511 * Timezone data updated to 2005m version.
3515 * The following bugs are resolved with this release:
3517 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3518 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3519 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3520 777, 787, 821, 822, 823, 825
3522 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3526 * Support for RFC 3678. Real implementations exist only for Linux so far.
3527 Implemented by Ulrich Drepper.
3529 * nscd can now cache entries persistently. Expiring entries are reloaded.
3530 For speedups the cache can be shared in memory with client processes.
3531 Implemented by Ulrich Drepper.
3533 * nscd can now perform SELinux checks.
3534 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3536 * getaddrinfo queries are now cached. Canonical name lookup is performed
3538 Implemented by Ulrich Drepper.
3540 * The nothrow function attribute is used when headers are used by gcc when
3541 compiling C code. This can avoid significant amounts of exception
3544 * The malloc functions perform more error checking and are stricter when
3545 it comes to reacting on errors. The default action is to terminate
3546 the process after showing an error message. Implemented by Ulrich Drepper.
3548 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3549 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3551 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3552 namespaces using the new function dlmopen(). This feature is of course,
3553 like most other dynamic loading functionality, not available in statically
3554 linked applications. Implemented by Ulrich Drepper.
3556 * Low-overhead boundary checking variants of string and some stdio functions
3557 were added. These are to be used in conjunction with a gcc patch by
3558 Jakub Jelinek which adds calls to these functions if possible.
3559 Implemented by Jakub Jelinek and Ulrich Drepper.
3561 * Old code for several operating systems and machine architectures that
3562 have not been in working condition in a long time have been removed from
3563 the main source tree maintained by the GNU C Library's maintainers.
3564 These files are now reside in the separate `ports' source module
3565 that is usable as an add-on when building the library.
3569 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3570 interrogate the dynamic linker, compatible with the Solaris interface.
3572 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3573 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3575 * getifaddrs now uses the netlink interface on Linux to get its information.
3576 Implemented by Thorsten Kukuk.
3578 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3579 Implemented by Ulrich Drepper.
3581 * support for non-executable stacks on x86 has been added. Changes mostly
3584 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3587 * getaddrinfo now performs destination address selection according to
3592 * Thread-safe interfaces for many functions that access locale data
3593 were added in version 2.3, but these features were omitted from NEWS.
3594 Many functions have variants with an `_l' suffix that take a `locale_t'
3595 object as a parameter rather than consulting the current locale.
3596 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3597 create and maintain `locale_t' objects. Additionally, the new function
3598 `uselocale' sets "the current locale" (as used by functions not so
3599 parameterized) set for an individual thread. These features were added
3600 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3602 * The functions getresuid, getresgid, setresuid, and setresgid, which
3603 have long been available on Linux, are now declared in <unistd.h>
3604 and are now also available on the Hurd.
3606 * ELF thread-local storage support (TLS) now works on x86-64.
3608 * The new dynamic string token $LIB is expanded in shared library names.
3609 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3611 * Aldy Hernandez contributed complete software floating point support for
3612 PowerPC machines with no FPU.
3614 * fexecve is implemented on Linux.
3616 * The `btowc' function should work at least twice as fast due to
3617 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3619 * With approriate thread add-ons cancelable functions are now implemented
3620 in libc.so as well. No need to call the function in libpthread. This
3621 change allowed to finally disable the incorrect and expensive handling
3622 of weak definition in ld.so.
3624 * Yet more PLT entries in libc.so have been removed. We finally arrived
3625 at the bare minimum. Startup times improved appropriately.
3627 * Support for the new Linux/x86 system call interface was added. The
3628 AT_SYSINFO auxiliary vector entry is recognized and handled.
3632 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3635 * iconv (the program and the interface) now accepts empty names (excluding
3636 options like //TRANSLIT) to mean "use charset of current locale".
3638 * localedef can now transliterate characters in strings which are not in
3639 the provided charmap. The information from the input locale is used.
3641 * Prelinking support was added for ELF targets. This requires additional
3642 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3644 * Read-only stdio streams now use mmap to speed up operation by eliminating
3645 copying and buffer underflows. To use add 'm' to the mode string of
3646 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3648 * The malloc functions were completely rewritten by Wolfram Gloger based
3649 on Doug Lea's malloc-2.7.0.c.
3651 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3652 implementation of regex.
3654 * Bruno Haible upgraded the iconv and locale implementation to support
3657 * Contents of the LC_* and LANG environment variables in the CEN style are
3658 not recognized anymore. It never was used. Change by Ulrich Drepper.
3660 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3661 thread-local storage (TLS) ABI on some platforms.
3662 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3664 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3665 EUC-JISX0213, and TSCII.
3667 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3668 BSD-compatible interface for getting all network interface addresses.
3669 Implementation for IPv4 by Roland McGrath.
3671 * Loading of locale data is faster due to the introduction of a locale
3672 archive. Implemented by Roland McGrath and Ulrich Drepper.
3674 * Startup times are significantly reduced by not using exported functions
3675 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3678 * Steven Munroe contributed a port to PowerPC64/Linux.
3682 * The Hurd now uses the GNU libio implementation of stdio rather than the
3683 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3685 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3686 with the same behavior as the Linux system call of the same name.
3690 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3691 128-bit long double format.
3693 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3694 IBM1161, and IBM1162 charsets.
3696 * Andreas Jaeger contributed a port to x86-64/Linux.
3698 * Peter Bruin contributed a port to PowerPC/Hurd.
3700 * libc functions using I/O streams now can handle wide-oriented streams
3703 * optimizations in the dynamic linker. Binaries created by recent binutils
3704 versions start up quicker due to reduced time spent on relocations.
3706 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3710 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3711 asinh, atanh, j0 for the 128-bit long double format.
3713 * Bruno Haible updated all the code handling Unicode in some form to
3714 support Unicode 3.1.
3716 * Speed of regex for single-byte locales is back to previous levels.
3717 Patch by Isamu Hasegawa.
3719 * Alpha, SPARC, and IA-64 now also using floating stacks.
3721 * Startup time of internationalized applications greatly improved through
3722 iconv cache. Use iconvconfig to generate the cache file.
3723 Contributed by Ulrich Drepper.
3725 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3726 severe performance problems. Patch by David Mosberger.
3728 * The Hurd port got a lot more functionality like AIO, various stdio
3729 extensions, etc. Mainly done by Roland McGrath.
3731 * mtrace can now lookup symbols in shared libraries.
3735 * Intel's IA-64 math library is largely integrated. It provides fast and
3736 accurate implementatations for most basic and standard math functions
3737 in float, double, and long double format.
3739 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3740 and asin for the 96-bit long double format and asin, log, tan for the
3741 128-bit long double format.
3743 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3744 The basic double functions exist today. Contributed by Abraham Ziv
3745 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3746 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3748 * An asynchronous name lookup library was added. The interface is designed
3749 after POSIX AIO. The proposal was circulated beforehand to get comments.
3750 No negative ones came in. Implemented by Ulrich Drepper.
3752 * Port to S390/64bit contributed by Martin Schwidefsky
3753 <schwidefsky@de.ibm.com>.
3755 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3756 of functions for Linux/IA-64.
3758 * The RPC code is now thread safe. Threads can now use the same service
3759 of different services at the same time. Patch by Eric Norum
3760 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3762 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3763 family of functions for Linux/S390.
3765 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3766 of functions for Linux/x86.
3768 * Port to Linux/CRIS contributed by Axis Communications.
3772 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3773 we can now check for namespace violations and missing declarations. The
3774 result is that almost all headers are now Unix-compliant (as defined in
3775 the upcoming XPG6). The negative side is that some programs might need
3776 corrections, too, if they depend on the incorrect form of the headers in
3777 previous versions which defined too many symbols and included too many
3780 * regex now handles multibyte character sets correctly.
3781 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3783 * iconv (the program) does now conform to the upcoming XPG6 and handles
3784 charmaps. Instead of the charset names the path of charmaps can be
3785 provided and the conversion happens based on this data.
3786 Contributed by Ulrich Drepper.
3788 * The locale program now provides detailed information about the installed
3793 only lists the names of the supported locales
3797 provides details such as country, language, and codeset name.
3798 Contributed by Ulrich Drepper.
3802 * The gencat program now parses the input file according to the charset
3803 selected by the LC_CTYPE category. This is important for stateful
3804 character sets. To make generating catalogs easier there is a way
3805 to overwrite the charset selected by the locale: before the first
3806 message or $ quote line the catalog can contain a line like
3808 $ codeset=ISO-8859-2
3810 to select the charset (ISO-8859-2 in this case).
3812 Implemented by Shinya Hanataka and Ulrich Drepper.
3814 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3815 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3816 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3818 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3819 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3821 * The LANGUAGE environment variable is now ignored unless the locale is
3822 changed from the default "C" locale.
3824 * The usual bug fixes.
3828 * Greg McGary added runtime support for bounds checking using gcc's
3829 new -fbounded-pointers option. ix86 target is complete. PowerPC
3832 * Thorsten Kukuk added secure mode support to nscd.
3834 * The Berkeley db libraries have been removed.
3836 Related, the nss_db module is now in a separate package since it
3837 obviously requires a database library being available.
3839 * Wide character I/O streams implemented by Ulrich Drepper.
3841 * Functions from the extended socket API added by Ulrich Drepper.
3843 * Functions feenableexcept and fedisableexcept to control the
3844 behaviour of individual exceptions have been added by Andreas Jaeger.
3846 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3848 * The resolver code has been updated from bind 8.2.3-T5B which supports
3849 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3852 This change could in some situations effect backward compatibility. Since
3853 now `_res' is a thread-local instead of a global variable, modifying it
3854 in one thread does not have any effect in other threads.
3856 The resolver library was also extended to allow IPv6 as the transport
3857 protocol for the requests. This work was done by Stig Venaas.
3859 * Compatibility code for K&R C compilers has been removed from the
3860 header files. A ISO C compiler is needed to use the library
3861 (conforming to either C89 or C99 standard).
3863 * Complete rewrite of the localedef program to support multibyte character
3864 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3865 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3866 Implemented by Ulrich Drepper.
3868 Bruno Haible significantly improved the generation and use of the data
3869 structures for the wide character tables.
3871 * Plural handling in gettext implemented by Ulrich Drepper.
3873 * The utmp daemon has been removed.
3875 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3877 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3880 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3882 * POSIX barriers implemented by Kaz Kylheku.
3884 * POSIX spawn function family implemented by Ulrich Drepper.
3886 * POSIX spinlocks are now available.
3888 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3890 * the configure option --enable-kernel=X.Y.Z allows to strip out
3891 compatibility for kernel versions before X.Y.Z. This is currently only
3892 implemented for Linux.
3894 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3895 field sin6_scope_id. This means that all programs using IPv6 should be
3896 recompiled. Don't expect binary compatibility with previous glibc
3899 * various conversion modules for IBM character sets contributed by
3902 * IA-64 port by Jes Sorensen and HJ Lu.
3916 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3918 * New xdr functions are added; some rpc functions are now 64bit clean.
3920 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3922 * Fixed known incompatibilities with glibc 2.0.
3924 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3926 * Optimized a number of functions (especially the ELF dynamic loader).
3928 * Update timezone data files.
3930 * lots of charmaps corrections
3932 * some new locale definitions and charmaps
3937 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3938 conform to POSIX member type requirements. Symbol versions have been
3939 adjusted as needed within the library, and for direct use by applications,
3940 but there is potential for problems if third-party libraries use
3941 struct timeval as part of their interface. This does not present
3942 a problem for X and other "essential" system libraries.
3944 * An additional locale model to support C++ Standard Library locale
3945 model and probably more was implemented by Ulrich Drepper.
3947 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3950 * Miles Bader provided the `argp' function family to support hierachical
3951 command line argument parsing, layered on top of getopt.
3953 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3955 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3958 * scanf recognizes the %a and %A format for scanning floating point numbers.
3960 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3961 information and interfaces for the available integer types.
3963 * about 130 new math functions were added to implement the ISO C9x math
3966 * the new header <complex.h> contains definitions of the complex math
3967 functions from ISO C 9X.
3969 * the new header <tgmath.h> defines generic macros to use complex or
3970 real valued functions.
3972 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3974 * Andreas Jaeger provided a test suite for the math library.
3976 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3978 * Ulrich Drepper added character set conversion functions (iconv).
3980 * Optimized string functions have been added.
3982 * The localedata addon is now part of glibc.
3984 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3986 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3987 daemon for NSS (nscd).
3989 Missing a better place here are some numbers on improvements. Under
3990 Linux 2.1.125 un-tar-ing the kernel sources takes
3994 using local files 12.19s 6.88s 22.91s
3996 using NIS 13.92s 8.91s 26.34s
3998 using NIS & nscd 10.37s 7.34s 25.30s
4000 using NIS+ 27.57s 30.37s 640.46s
4002 using NIS+ & nscd 10.25s 7.83s 26.51s
4004 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4006 Keep in mind that non-namelookup related operations dominate above times.
4007 It was just a common complain that using NIS+ unpacking the kernel is
4010 [1] The old nscd implementation is not available anymore since it was
4011 distributed with glibc up to version 2.0.98 and thus is now replaced.
4013 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4015 * Mark Kettenis provided a Hesiod NSS module.
4017 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4018 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4020 * rcmd can now handle netgroups (Dick Streefland).
4022 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4025 * Support for the IPv6 protocol has been added to the socket API, as per the
4026 latest draft standards.
4028 * Support for Linux 2.2 has been added.
4030 * Interface changes relative to the latest 2.0.x release:
4031 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4032 addseverity NEW: Unix98
4033 alphasort64 NEW: LFS
4034 argp_err_exit_status NEW: argp, GNU ext
4035 argp_error NEW: argp, GNU ext
4036 argp_failure NEW: argp, GNU ext
4037 argp_help NEW: argp, GNU ext
4038 argp_parse NEW: argp, GNU ext
4039 argp_program_bug_address NEW: argp, GNU ext
4040 argp_program_version NEW: argp, GNU ext
4041 argp_program_version_hook NEW: argp, GNU ext
4042 argp_state_help NEW: argp, GNU ext
4043 argp_usage NEW: argp, GNU ext
4044 authdes_create NEW: Secure RPC
4045 authdes_getucred NEW: Secure RPC
4046 authdes_pk_create NEW: Secure RPC
4047 backtrace NEW: GNU ext.
4048 backtrace_symbols NEW: GNU ext.
4049 backtrace_symbols_fd NEW: GNU ext.
4051 cacosf NEW: ISO C 9x
4052 cacosh NEW: ISO C 9x
4053 cacoshf NEW: ISO C 9x
4054 cacoshl NEW: ISO C 9x
4055 cacosl NEW: ISO C 9x
4062 casinf NEW: ISO C 9x
4063 casinh NEW: ISO C 9x
4064 casinhf NEW: ISO C 9x
4065 casinhl NEW: ISO C 9x
4066 casinl NEW: ISO C 9x
4068 catanf NEW: ISO C 9x
4069 catanh NEW: ISO C 9x
4070 catanhf NEW: ISO C 9x
4071 catanhl NEW: ISO C 9x
4072 catanl NEW: ISO C 9x
4073 cbc_crypt NEW: Secure RPC
4077 ccoshf NEW: ISO C 9x
4078 ccoshl NEW: ISO C 9x
4084 cimagf NEW: ISO C 9x
4085 cimagl NEW: ISO C 9x
4086 clearerr_locked REMOVED
4087 clntunix_create NEW: sunrpc ext
4089 clog10 NEW: ISO C 9x
4090 clog10f NEW: ISO C 9x
4091 clog10l NEW: ISO C 9x
4101 cprojf NEW: ISO C 9x
4102 cprojl NEW: ISO C 9x
4104 crealf NEW: ISO C 9x
4105 creall NEW: ISO C 9x
4110 csinhf NEW: ISO C 9x
4111 csinhl NEW: ISO C 9x
4114 csqrtf NEW: ISO C 9x
4115 csqrtl NEW: ISO C 9x
4119 ctanhf NEW: ISO C 9x
4120 ctanhl NEW: ISO C 9x
4122 des_setparity NEW: Secure RPC
4123 ecb_crypt NEW: Secure RPC
4124 endutxent NEW: Unix98
4126 exp10f NEW: ISO C 9x
4127 exp10l NEW: ISO C 9x
4131 fattach NEW: STREAMS
4132 fdetach NEW: STREAMS
4136 feclearexcept NEW: ISO C 9x
4137 fegetenv NEW: ISO C 9x
4138 fegetexceptflag NEW: ISO C 9x
4139 fegetround NEW: ISO C 9x
4140 feholdexcept NEW: ISO C 9x
4142 feraiseexcept NEW: ISO C 9x
4143 ferror_locked REMOVED
4144 fesetenv NEW: ISO C 9x
4145 fesetexceptflag NEW: ISO C 9x
4146 fesetround NEW: ISO C 9x
4147 fetestexcept NEW: ISO C 9x
4148 feupdateenv NEW: ISO C 9x
4149 fflush_locked REMOVED
4153 fgets_unlocked NEW: GNU ext.
4154 fileno_locked REMOVED
4166 fputc_locked REMOVED
4167 fputs_unlocked NEW: GNU ext.
4168 fread_unlocked NEW: GNU ext.
4173 fstatvfs NEW: Unix98
4177 ftruncate64 NEW: LFS
4179 fwrite_unlocked NEW: GNU ext.
4180 gai_strerror NEW: IPv6
4184 getchar_locked REMOVED
4186 getdate_err NEW: Unix98
4187 getdate_r NEW: GNU ext.
4189 getnameinfo NEW: IPv6
4190 getnetname NEW: Secure RPC
4191 getpmsg NEW: STREAMS
4192 getpt NEW: Unix98 PTY
4193 getrlimit64 NEW: LFS
4194 getutxent NEW: Unix98
4195 getutxid NEW: Unix98
4196 getutxline NEW: Unix98
4197 glob64 NEW: GNU ext.
4198 globfree64 NEW: GNU ext.
4199 gnu_get_libc_release NEW: GNU ext.
4200 gnu_get_libc_version NEW: GNU ext.
4201 grantpt NEW: Unix98 PTY
4202 host2netname NEW: Secure RPC
4204 iconv_close NEW: iconv
4205 iconv_open NEW: iconv
4206 if_freenameindex NEW: IPv6
4207 if_indextoname NEW: IPv6
4208 if_nameindex NEW: IPv6
4209 if_nametoindex NEW: IPv6
4210 in6addr_any NEW: IPv6
4211 in6addr_loopback NEW: IPv6
4212 inet6_isipv4mapped NEW: IPv6
4213 isastream NEW: STREAMS
4214 iswblank NEW: GNU ext.
4215 key_decryptsession NEW: Secure RPC
4216 key_decryptsession_pk NEW: Secure RPC
4217 key_encryptsession NEW: Secure RPC
4218 key_encryptsession_pk NEW: Secure RPC
4219 key_gendes NEW: Secure RPC
4220 key_get_conv NEW: Secure RPC
4221 key_secretkey_is_set NEW: Secure RPC
4222 key_setnet NEW: Secure RPC
4223 key_setsecret NEW: Secure RPC
4224 llrint NEW: ISO C 9x
4225 llrintf NEW: ISO C 9x
4226 llrintl NEW: ISO C 9x
4227 llround NEW: ISO C 9x
4228 llroundf NEW: ISO C 9x
4229 llroundl NEW: ISO C 9x
4234 lrintf NEW: ISO C 9x
4235 lrintl NEW: ISO C 9x
4236 lround NEW: ISO C 9x
4237 lroundf NEW: ISO C 9x
4238 lroundl NEW: ISO C 9x
4240 makecontext NEW: Unix98
4241 mempcpy NEW: GNU ext.
4244 modify_ldt NEW: kernel
4248 nearbyint NEW: ISO C 9x
4249 nearbyintf NEW: ISO C 9x
4250 nearbyintl NEW: ISO C 9x
4251 netname2host NEW: Secure RPC
4252 netname2user NEW: Secure RPC
4253 nexttoward NEW: ISO C 9x
4254 nexttowardf NEW: ISO C 9x
4255 nexttowardl NEW: ISO C 9x
4259 passwd2des NEW: Secure RPC
4261 pow10f NEW: GNU ext.
4262 pow10l NEW: GNU ext.
4265 printf_size NEW: GNU ext.
4266 printf_size_info NEW: GNU ext.
4267 profil_counter REMOVED
4268 pthread_mutexattr_getkind_np REPLACED
4269 pthread_mutexattr_setkind_np REPLACED
4270 ptsname NEW: Unix98 PTY
4271 ptsname_r NEW: Unix98 PTY
4273 putchar_locked REMOVED
4274 putgrent NEW: GNU ext.
4276 putpmsg NEW: STREAMS
4277 pututxline NEW: Unix98
4281 readdir64_r NEW: LFS
4282 remquo NEW: ISO C 9x
4283 remquof NEW: ISO C 9x
4284 remquol NEW: ISO C 9x
4286 roundf NEW: ISO C 9x
4287 roundl NEW: ISO C 9x
4289 scalbln NEW: ISO C 9x
4290 scalblnf NEW: ISO C 9x
4291 scalblnl NEW: ISO C 9x
4293 sendfile NEW: kernel
4294 setcontext NEW: Unix98
4295 setrlimit64 NEW: LFS
4296 setutxent NEW: Unix98
4298 sigignore NEW: Unix98
4299 sigqueue NEW: POSIX.1b
4300 sigrelse NEW: Unix98
4301 sigset NEW: POSIX.1b
4302 sigtimedwait NEW: POSIX.1b
4303 sigwaitinfo NEW: POSIX.1b
4304 sincos NEW: GNU ext.
4305 sincosf NEW: GNU ext.
4306 sincosl NEW: GNU ext.
4310 strcasestr NEW: GNU ext.
4311 strtoimax NEW: ISO C 9x
4312 strtoumax NEW: ISO C 9x
4313 strverscmp NEW: GNU ext.
4314 svcauthdes_stats NEW: Secure RPC
4315 svcunix_create NEW: sunrpc etx
4316 svcunixfd_create NEW: sunrpc ext
4317 swapcontext NEW: Unix98
4318 tcgetsid NEW: Unix98 PTY
4319 tdestroy NEW: GNU ext.
4320 tgamma NEW: ISO C 9x
4321 tgammaf NEW: ISO C 9x
4322 tgammal NEW: ISO C 9x
4326 truncf NEW: ISO C 9x
4327 truncl NEW: ISO C 9x
4329 unlockpt NEW: Unix98 PTY
4330 updwtmpx NEW: Unix98
4331 user2netname NEW: Secure RPC
4332 utmpxname NEW: Unix98
4333 versionsort NEW: GNU ext.
4334 versionsort64 NEW: GNU ext.
4336 wcscasecmp NEW: GNU ext.
4337 wcsncasecmp NEW: GNU ext.
4338 wcsnlen NEW: GNU ext.
4339 wcstoimax NEW: ISO C 9x
4340 wcstoll NEW: ISO C 9x
4341 wcstoull NEW: ISO C 9x
4342 wcstoumax NEW: ISO C 9x
4344 wordexp NEW: POSIX.2
4345 wordfree NEW: POSIX.2
4346 write_profiling REMOVED
4347 xdecrypt NEW: Secure RPC
4348 xdr_authdes_cred NEW: Secure RPC
4349 xdr_authdes_verf NEW: Secure RPC
4350 xdr_cryptkeyarg NEW: Secure RPC
4351 xdr_cryptkeyarg2 NEW: Secure RPC
4352 xdr_cryptkeyres NEW: Secure RPC
4353 xdr_getcredres NEW: Secure RPC
4354 xdr_key_netstarg NEW: Secure RPC
4355 xdr_key_netstres NEW: Secure RPC
4356 xdr_keybuf NEW: Secure RPC
4357 xdr_keystatus NEW: Secure RPC
4358 xdr_netnamestr NEW: Secure RPC
4359 xdr_sizeof NEW: Secure RPC
4360 xdr_unixcred NEW: sunrpc ext
4361 xencrypt NEW: Secure RPC
4362 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4373 * inet_ntoa is thread-safe
4375 * updwtmp is moved from libutil to libc
4377 * rewrite of cbrt function
4379 * update of timezone data
4393 * add atoll function
4395 * fix complex problems in Berkeley DB code
4397 * fix math functions
4401 * fixed lots of header problems (especially Linux/GNU specific)
4403 * dynamic loader preserves all registers
4405 * Roland McGrath provided support for handling of auxiliary objects in
4406 the ELF dynamic loader.
4408 * support for parallel builds is improved
4412 * GNU extensions are no longer declared by default. To enable them you
4413 must define the macro `_GNU_SOURCE' in your program or compile with
4416 * The library has changed from using GNU ld symbol aliases to using weak
4417 symbols where available. The ELF object file format supports weak
4418 symbols; GNU ld also supports weak symbols in the a.out format. (There
4419 is also now support for other GNU ld extensions in ELF. Use the
4420 `--with-elf' option to configure to indicate you have ELF, and
4421 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4422 of many files which contained only symbol aliases, reducing the size of
4423 the source and the compiled library; many other files were renamed to
4424 less cryptic names previously occupied by the symbol alias files.
4425 There is a new header file <elf.h> for programs which operate on
4426 files in the ELF format.
4428 * Converted to Autoconf version 2, so `configure' has more options.
4429 Run `configure --help' to see the details.
4431 * The library can now be configured to build profiling, highly-optimized
4432 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4433 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4434 `configure' enable building these extra libraries. The shared library is
4435 built by default when using both ELF and GNU ld. When shared libraries
4436 are enabled, the new library `-ldl' is available for arbitrary run-time
4437 loading of shared objects; its interface is defined in <dlfcn.h>. The
4438 new header file <link.h> gives access to the internals of the run-time
4439 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4440 application of same name on other systems and it provides information
4441 about dynamically linked binaries.
4443 * The C library now provides the run-time support code for profiling
4444 executables compiled with `-pg'. Programs can control the profiling code
4445 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4446 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4447 the support for this file format was contributed by David Mosberger-Tang.
4449 * The math code has been replaced with a math library based on fdlibm from
4450 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4451 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4452 functions. The math functions now reside in a separate library, so
4453 programs using them will need to use `-lm' their linking commands.
4455 * John C. Bowman contributed optimized ix87 assembler inline functions.
4457 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4458 mechanism similar to that found in Solaris 2. This is now used for the
4459 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4460 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4461 file controls what services are used for each individual database. This
4462 works by loading shared libraries with names specified in `nsswitch.conf',
4463 so service modules can be changed or added at any time without even
4464 relinking any program. Currently there are the file, db, and NIS based
4465 NSS services available.
4467 * The new functions `strtoq' and `strtouq' parse integer values from
4468 strings, like `strtol' and `strtoul', but they return `long long int' and
4469 `unsigned long long int' values, respectively (64-bit quantities).
4471 * The new functions `strtof' and `strtold' parse floating-point values from
4472 strings, like `strtod', but they return `float' and `long double' values,
4473 respectively (on some machines `double' and `long double' are the same).
4475 * Ulrich Drepper has contributed new implementations of the floating-point
4476 printing and reading code used in the `printf' family of functions and
4477 `strtod', `strtof', and `strtold'. These new functions are perfectly
4478 accurate, and much faster than the old ones.
4480 * The implementation of the POSIX locale model was completely rewritten by
4481 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4482 to compile the POSIX locale definition.
4484 * The former dummy implementations of the strcoll and strxfrm function are
4485 now replaced by fully functional code contributed by Ulrich Drepper. The
4486 collation information comes from the POSIX locale definitions.
4488 * The new header <langinfo.h> defines an interface for accessing
4489 various locale-dependent data (using the locale chosen with `setlocale').
4491 * Ulrich Drepper has contributed a new suite of functions for operation on
4492 wide-character and multibyte-character strings, in <wchar.h>;
4493 and classification and case conversion of wide characters, in <wctype.h>.
4494 These new functions are conforming to the ISO C, Amendement 1 specification.
4496 * There is now a second implementation of the standard I/O library available.
4497 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4498 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4500 * You can now use positional parameter specifications in format strings
4501 for the `printf' and `scanf' families of functions. For example,
4502 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4503 ``Number 6, Mr Jones''. This is mainly useful when providing different
4504 format strings for different languages, whose grammars may dictate
4505 different orderings of the values being printed. To support this
4506 feature, the interface for `register_printf_handler' has changed; see
4507 the header file <printf.h> for details.
4509 * The `printf' and `scanf' families of functions now understand a new
4510 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4511 to group numbers as indicated by the locale; for `scanf' and friends, this
4512 says to accept as valid only a number with all the proper grouping
4513 separators in the right places. In the default "C" locale, numbers are
4514 not grouped; but locales for specific countries will define the usual
4515 conventions (i.e. separate thousands with `,' in the US locale).
4517 * The pgrp functions have been regularized, slightly incompatibly but much
4518 less confusingly. The core functions are now `getpgid' and `setpgid',
4519 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4520 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4521 provided for compatibility. There is no longer an incompatible `getpgrp'
4522 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4524 * The new header file <fts.h> and suite of functions simplify programs that
4525 operate on directory trees. This code comes from 4.4 BSD.
4527 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4528 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4529 NSS scheme used in glibc.
4531 * There is a new malloc debugging hook `__memalign_hook'.
4533 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4534 `unsigned int' in <sys/types.h>. These are for compatibility only and
4535 their use is discouraged.
4537 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4538 done differently, so that it works even without GNU ld.
4540 * New function `euidaccess' checks allowed access to a file like `access',
4541 but using the effective IDs instead of the real IDs.
4543 * The time zone data files have been updated for the latest and greatest
4544 local time conventions of the countries of the world.
4546 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4549 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4550 for formatting floating-point numbers. They are provided only for
4551 compatibility; new programs should use `sprintf' instead. There are
4552 also equivalent function for the `long double' floating-point type and
4553 all functions also exist in a reentrant form.
4555 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4556 functions for maintaining the login-record files (primarily of use to
4557 system programs such as `login'), and convenient functions for
4558 allocating and initializing a pseudo-terminal (pty) device.
4560 * Ulrich Drepper has contributed new support for System V style
4561 shared memory and IPC on systems that support it.
4563 * Ulrich Drepper has contributed several miscellaneous new functions found
4564 in System V: The `hsearch' family of functions provide an effective
4565 implementation of hash tables; `a64l' and `l64a' provide a very simple
4566 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4569 * Ulrich Drepper has contributed new reentrant counterparts for the
4570 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4572 * Ulrich Drepper has contributed new, highly-optimized versions of several
4573 string functions for the i486/Pentium family of processors.
4575 * Ulrich Drepper has updated the Linux-specific code, based largely
4576 on work done in Hongjiu Lu's version of GNU libc for Linux.
4577 The GNU library now supports Linux versions 2.0.10 and later,
4578 using the ELF object file format (i[3456]86-*-linux).
4580 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4582 * David Mosberger-Tang and Richard Henderson have ported the C library
4583 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4584 dynamic linking support for ELF/Alpha.
4586 * Richard Henderson contributed several Alpha optimized assembler function
4587 for arithmetic and string handling.
4589 * Ulrich Drepper has contributed a new set of message catalog functions to
4590 support multiple languages using the <libintl.h> interface, for use with
4591 his new package GNU gettext. Translation volunteers have contributed
4592 catalogs of the library's messages in Spanish, German, and Korean.
4594 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4595 program and the `catgets' function for reading the catalog files it
4596 creates. (The <libintl.h> interface is preferred; we include the
4597 <nl_types.h> interface using `catgets' only for source compatibility with
4598 programs already written to use it.)
4600 * New header file <values.h> gives SVID-compatible names for <limits.h>
4603 * Various new macros, declarations, and small header files for compatibility
4606 * New function `group_member' is a convenient way to check if a process has
4607 a given effective group ID.
4609 * When using GCC 2.7 and later, the socket functions are now declared in a
4610 special way so that passing an argument of type `struct sockaddr_in *',
4611 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4612 `struct sockaddr *' type, does not generate a type-clash warning.
4614 * New function `error' declared in header file <error.h> is a convenient
4615 function for printing error messages and optionally exiting; this is the
4616 canonical function used in GNU programs. The new functions `err', `warn',
4617 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4618 doing the same thing.
4620 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4621 POSIX.2 `glob' function to do ~ and {...} expansion.
4623 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4624 `clearenv' which is used in POSIX.9 is also available.
4626 * New function `getsid' returns session ID number on systems that support it.
4628 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4629 files <db.h> and <mpool.h> provide a rich set of functions for several
4630 types of simple databases stored in memory and in files, and <ndbm.h> is
4631 an old `ndbm'-compatible interface using the `db' functions. Link with
4632 `-ldb' to get these functions.
4634 * New macro `strdupa' copies a string like `strdup', but uses local stack
4635 space from `alloca' instead of dynamic heap space from `malloc'.
4637 * New function `strnlen' is like `strlen' but searches only a given maximum
4638 number of characters for the null terminator. `stpncpy', `strndup' and
4639 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4642 * New function `statfs' in header <sys/statfs.h>.
4644 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4645 provide convenient functions for operating on blocks of null-terminated
4648 * A new suite of functions in <utmp.h> handle all the details of reading
4649 and writing the utmp file.
4651 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4654 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4655 completely POSIX compliant and also implemented the extended functionality
4656 to handle alternate digit representation and alternate era date formats.
4658 * Ulrich Drepper provided an implementation of the `strptime' function
4659 defined in XPG4.2 which transforms a string into a `struct tm' value.
4661 * Paul Eggert provided the tzselect shell script as part of the timezone
4662 code. The shell script makes it easy to select the correct timezone
4665 * The implementation of the malloc family of functions is completely replaced
4666 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4667 The implementation uses the mmap function (if available) and it is
4668 optimized for the use in multi threaded programs.
4670 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4671 functions. This new functionality is usable by specifying a special
4672 salt string and it is compatible with implementation on *BSD systems.
4674 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4675 `getsubopt' to handle second level command line options, `bsd_signal'
4676 to access BSD style `signal' functionality, the obsolete `regexp' style
4679 * the `lchown' function is available on system which support this
4682 * The implementation of the shadow password handling function was contributed
4685 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4687 * POSIX.1g support was added. The <sys/select.h> header is available,
4688 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4689 implementation of `getaddrinfo'.
4693 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4695 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4696 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4698 * The new function `daemon' from 4.4 BSD is useful for server programs that
4699 want to put themselves in the background.
4701 * Joel Sherrill has contributed support for several standalone boards that
4702 run without an operating system.
4704 * `printf', `scanf' and friends now accept a `q' type modifier for long
4705 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4707 * All of the code taken from BSD (notably most of the math and networking
4708 routines) has been updated from the BSD 4.4-Lite release.
4710 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4712 * The new functions `getdomainname' and `setdomainname' fetch or change the
4713 YP/NIS domain name. These are system calls which exist on systems which
4716 * The time zone data files have been updated for the latest international
4719 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4720 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4724 * The C library now includes support for Sun RPC, from Sun's free
4725 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4726 are included. (There is still no support for YP.)
4728 * Tom Quinn has contributed a port of the C library to SGI machines running
4729 Irix 4 (mips-sgi-irix4).
4731 * The new `lockf' function is a simplified interface to the locking
4732 facilities of `fcntl', included for compatibility.
4734 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4736 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4738 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4741 * The error message printed by `assert' for a failed assertion now includes
4742 the name of the program (if using GNU ld) and the name of the calling
4743 function (with versions of GCC that support this).
4745 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4747 * The library now includes the <sys/mman.h> header file and memory
4748 management functions `mmap', `munmap', `mprotect', `msync', and
4749 `madvise', on systems that support those facilities.
4751 * The interface for `mcheck' has changed slightly: the function called to
4752 abort the program when an allocation inconsistency is detected now takes
4753 an argument that indicates the type of failure. The new function
4754 `mprobe' lets you request a consistency check for a particular block at
4755 any time (checks are normally done only when you call `free' or `realloc'
4758 * It is now possible to easily cross-compile the C library, building on one
4759 system a library to run on another machine and/or operating system. All
4760 you need to do is set the variable `HOST_CC' in `configparms' to the
4761 native compiler for programs to run on the machine you are building on (a
4762 few generator programs are used on Unix systems); set `CC' to the
4765 * The new function `fexecve' (only implemented on the GNU system) executes
4766 a program file given a file descriptor already open on the file.
4770 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4771 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4773 * You can set the variable `libprefix' in `configparms' to specify a prefix
4774 to be prepended to installed library files; this makes it easy to install
4775 the GNU C library to be linked as `-lgnuc' or whatever.
4777 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4778 copies a limited number of characters from a string, and returns the
4779 address of the last character written.
4781 * You no longer need to check for whether the installed `stddef.h' is
4782 compatible with the GNU C library. configure now checks for you.
4784 * You can now define a per-stream `fileno' function to convert the
4785 stream's cookie into an integral file descriptor.
4787 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4788 allocates zero bytes of storage, and returns a unique pointer which
4789 you can pass to `realloc' or `free'. The behavior is undefined if
4790 you dereference this pointer.
4792 * The C library now runs on Sony NEWS m68k machines running either
4793 NewsOS 3 or NewsOS 4.
4795 * The new `syscall' function is a system-dependent primitive function
4796 for invoking system calls. It has the canonical behavior on Unix
4797 systems, including unreliable return values for some calls (such as
4798 `pipe', `fork' and `getppid').
4800 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4801 to `EAGAIN', which is the preferred name. On systems whose kernels
4802 use two distinct codes, the C library now translates EWOULDBLOCK to
4803 EAGAIN in every system call function.
4807 * The GNU C Library Reference Manual is now distributed with the library.
4808 `make dvi' will produce a DVI file of the printed manual.
4809 `make info' will produce Info files that you can read on line using C-h i
4810 in Emacs or the `info' program.
4811 Please send comments on the manual to bug-glibc-manual@gnu.org.
4813 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4815 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4817 * Jason Merrill has contributed a port to the Sequent Symmetry running
4818 Dynix version 3 (i386-sequent-dynix).
4820 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4821 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4823 * New function `memory_warnings' lets you arrange to get warnings when
4824 malloc is running out of memory to allocate, like Emacs gives you.
4826 * The C library now contains the relocating allocator used in Emacs 19 for
4827 its editing buffers. This allocator (ralloc) minimizes allocation
4828 overhead and fragmentation by moving allocated regions around whenever it
4829 needs to. You always refer to a ralloc'd region with a "handle" (a
4830 pointer to a pointer--an object of type `void **').
4832 * There is a new `printf' format: `%m' gives you the string corresponding
4833 to the error code in `errno'.
4835 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4836 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4837 pass, the `a' modifier says to fill in a `char **' you pass with a
4840 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4841 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4842 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4844 * `mkstemp' is a traditional Unix function to atomically create and open a
4845 uniquely-named temporary file.
4849 * The standard location for the file that says what the local timezone is
4850 has changed again. It is now `/usr/local/etc/localtime' (or more
4851 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4853 * The distribution no longer contains any files with names longer than 14
4856 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4857 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4859 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4863 * The configuration process has changed quite a bit. The `configure'
4864 script is now used just like the configuration scripts for other GNU
4865 packages. The `sysdeps' directory hierarchy is much rearranged.
4866 The file `INSTALL' explains the new scheme in detail.
4868 * The header files no longer need to be processed into ANSI C and
4869 traditional C versions. There is just one set of files to install, and
4870 it will work with ANSI or old C compilers (including `gcc -traditional').
4872 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4873 MIPS DECStation running Ultrix 4.
4875 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4876 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4877 made itself into a shared library.
4879 * Yet further improved support for the i386, running 4.3 BSD-like systems
4880 (such as Mach 3 with the Unix single-server), or System V.
4882 * New function `strncasecmp' to do case-insensitive string comparison
4883 with limited length.
4885 * New function `strsep' is a reentrant alternative to `strtok'.
4887 * New functions `scandir' and `alphasort' for searching directories.
4889 * New function `setenv' is a better interface to `putenv'.
4891 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4892 function for traversing a directory tree.
4894 * The GNU obstack package is now also part of the C library.
4895 The new function `open_obstack_stream' creates a stdio stream that
4896 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4897 formatted output directly to an obstack.
4899 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4900 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4902 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4904 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4905 things to your strings.
4907 * There are some new test programs: test-fseek, testmb, and testrand.
4909 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4910 These ports are not finished, but are a good starting place for really
4911 supporting those systems.
4913 * `/etc/localtime' is now the standard location for the file that says what
4914 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4915 This follows the general principle that `/etc' is the place for all local
4916 configuration files.
4918 * The C library header files now use `extern "C"' when used by the C++
4919 compiler, so the C library should now work with C++ code.
4921 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4922 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4925 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4926 function which uses the `merge sort' algorithm, and is said to be
4927 significantly faster than the old GNU `qsort' function. Merge sort is
4928 now the standard `qsort' function. The new algorithm can require a lot
4929 of temporary storage; so, the old sorting function is called when the
4930 required storage is not available.
4932 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4933 provides the Unix `crypt' function, plus some other entry points.
4935 * The code and header files taken from 4.4 BSD have been updated with the
4936 latest files released from Berkeley.
4938 ----------------------------------------------------------------------
4939 Copyright information:
4941 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4943 Permission is granted to anyone to make or distribute verbatim copies
4944 of this document as received, in any medium, provided that the
4945 copyright notice and this permission notice are preserved,
4946 thus giving the recipient permission to redistribute in turn.
4948 Permission is granted to distribute modified versions
4949 of this document, or of portions of it,
4950 under the above conditions, provided also that they
4951 carry prominent notices stating who last changed them.
4954 version-control: never