1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2019 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Unicode 12.0.0 Support: Character encoding, character type info, and
13 transliteration tables are all updated to Unicode 12.0.0, using
14 generator scripts contributed by Mike FABIAN (Red Hat).
16 * The dynamic linker accepts the --preload argument to preload shared
17 objects, in addition to the LD_PRELOAD environment variable.
19 * On Linux, the gettid function has been added.
21 * Minguo (Republic of China) calendar support has been added as an
22 alternative calendar for the following locales: zh_TW, cmn_TW, hak_TW,
25 * The entry for the new Japanese era has been added for ja_JP locale.
27 * Memory allocation functions malloc, calloc, realloc, reallocarray, valloc,
28 pvalloc, memalign, and posix_memalign fail now with total object size
29 larger than PTRDIFF_MAX. This is to avoid potential undefined behavior with
30 pointer subtraction within the allocated object, where results might
31 overflow the ptrdiff_t type.
33 Deprecated and removed features, and other changes affecting compatibility:
35 * The functions clock_gettime, clock_getres, clock_settime,
36 clock_getcpuclockid, clock_nanosleep were removed from the librt library
37 for new applications (on architectures which had them). Instead, the
38 definitions in libc will be used automatically, which have been available
41 * The obsolete and never-implemented XSI STREAMS header files <stropts.h>
42 and <sys/stropts.h> have been removed.
44 * Support for the "inet6" option in /etc/resolv.conf and the RES_USE_INET6
45 resolver flag (deprecated in glibc 2.25) have been removed.
47 * The obsolete RES_INSECURE1 and RES_INSECURE2 option flags for the DNS stub
48 resolver have been removed from <resolv.h>.
50 Changes to build and runtime requirements:
52 * GCC 6.2 or later is required to build the GNU C Library.
54 Older GCC versions and non-GNU compilers are still supported when
55 compiling programs that use the GNU C Library.
57 Security related changes:
59 CVE-2019-7309: x86-64 memcmp used signed Jcc instructions to check
60 size. For x86-64, memcmp on an object size larger than SSIZE_MAX
61 has undefined behavior. On x32, the size_t argument may be passed
62 in the lower 32 bits of the 64-bit RDX register with non-zero upper
63 32 bits. When it happened with the sign bit of RDX register set,
64 memcmp gave the wrong result since it treated the size argument as
65 zero. Reported by H.J. Lu.
67 CVE-2019-9169: Attempted case-insensitive regular-expression match
68 via proceed_next_node in posix/regexec.c leads to heap-based buffer
69 over-read. Reported by Hongxu Chen.
71 The following bugs are resolved with this release:
73 [The release manager will add the list generated by
74 scripts/list-fixed-bugs.py just before the release.]
81 * The getcpu wrapper function has been added, which returns the currently
82 used CPU and NUMA node. This function is Linux-specific.
84 * A new convenience target has been added for distribution maintainers
85 to build and install all locales as directories with files. The new
86 target is run by issuing the following command in your build tree:
87 'make localedata/install-locale-files', with an optional DESTDIR
88 to set the install root if you wish to install into a non-default
91 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
93 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
94 for _GNU_SOURCE, to match BSD environments.
96 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
97 indicates that it will abort the transaction prior to entering the kernel
98 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
99 suspended, and this caused some undefined side-effects issues by aborting
100 transactions manually. Glibc avoided it by abort transactions manually on
101 each syscall, but it lead to performance issues on newer kernels where the
102 HTM state is saved and restore lazily (the state being saved even when the
103 process actually does not use HTM).
105 * The functions posix_spawn_file_actions_addchdir_np and
106 posix_spawn_file_actions_addfchdir_np have been added, enabling
107 posix_spawn and posix_spawnp to run the new process in a different
108 directory. These functions are GNU extensions. The function
109 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
112 * The popen and system do not run atfork handlers anymore (BZ#17490).
113 Although it is a possible POSIX violation, the POSIX rationale in
114 pthread_atfork documentation regarding atfork handlers is to handle
115 inconsistent mutex state after a fork call in a multi-threaded process.
116 In both popen and system there is no direct access to user-defined mutexes.
118 * Support for the C-SKY ABIV2 running on Linux has been added. This port
119 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
121 - C-SKY ABIV2 soft-float little-endian
122 - C-SKY ABIV2 hard-float little-endian
124 * strftime's default formatting of a locale's alternative year (%Ey)
125 has been changed to zero-pad the year to a minimum of two digits,
126 like "%y". This improves the display of Japanese era years during
127 the first nine years of a new era, and is expected to be harmless
128 for all other locales (only Japanese locales regularly have
129 alternative year numbers less than 10). Zero-padding can be
130 overridden with the '_' or '-' flags (which are GNU extensions).
132 * As a GNU extension, the '_' and '-' flags can now be applied to
133 "%EY" to control how the year number is formatted; they have the
134 same effect that they would on "%Ey".
136 Deprecated and removed features, and other changes affecting compatibility:
138 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
139 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
141 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
142 in <sys/procfs.h>, has been corrected to match the type actually used by
143 the Linux kernel. This affects the size and layout of that structure on
144 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
146 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
147 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
148 defined in <sys/procfs.h>, has been corrected to match the type actually
149 used by the Linux kernel. This affects the size and layout of those
152 * An archaic GNU extension to scanf, under which '%as', '%aS', and '%a[...]'
153 meant to scan a string and allocate space for it with malloc, is now
154 restricted to programs compiled in C89 or C++98 mode with _GNU_SOURCE
155 defined. This extension conflicts with C99's use of '%a' to scan a
156 hexadecimal floating-point number, which is now available to programs
157 compiled as C99 or C++11 or higher, regardless of _GNU_SOURCE.
159 POSIX.1-2008 includes the feature of allocating a buffer for string input
160 with malloc, using the modifier letter 'm' instead. Programs using
161 '%as', '%aS', or '%a[...]' with the old GNU meaning should change to
162 '%ms', '%mS', or '%m[...]' respectively. Programs that wish to use the
163 C99 '%a' no longer need to avoid _GNU_SOURCE.
165 GCC's -Wformat warnings can detect most uses of this extension, as long
166 as all functions that call vscanf, vfscanf, or vsscanf are annotated with
167 __attribute__ ((format (scanf, ...))).
169 Changes to build and runtime requirements:
171 * Python 3.4 or later is required to build the GNU C Library.
173 * On most architectures, GCC 5 or later is required to build the GNU C
174 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
176 Older GCC versions and non-GNU compilers are still supported when
177 compiling programs that use the GNU C Library.
179 Security related changes:
181 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
182 denial of service due to resource exhaustion when processing getaddrinfo
183 calls with crafted host names. Reported by Guido Vranken.
185 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
186 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
187 happened, accessing the 32-bit size_t value as the full 64-bit register
188 in the assembly string/memory functions would cause a buffer overflow.
191 CVE-2016-10739: The getaddrinfo function could successfully parse IPv4
192 addresses with arbitrary trailing characters, potentially leading to data
193 or command injection issues in applications.
195 The following bugs are resolved with this release:
197 [10425] localedata: it_IT/it_CH: LC_TIME format is wrong
198 [10496] localedata: 12h time representation in multiple locales faulty
199 [10797] localedata: it_IT locale numeric does not have a separator for
201 [11319] libc: dprintf doesn't handle errors properly
202 [16346] time: mktime: potentially unsafe use of localtime_offset
203 [17248] build: glibc should not sort CFLAGS (support gcc plugins and
205 [17405] libc: Implement posix_spawn_file_actions_addchdir_np,
206 posix_spawn_file_actions_addfchdir_np
207 [17426] localedata: Indian locales: set the correct date format
208 [17490] stdio: popen should not invoke atfork handlers
209 [17783] libc: TIOCSER_TEMT conditions inconsistent
210 [18040] regex: use-after-free in regexec/get_subexp
211 [18093] libc: Corrupted aux-cache causes ldconfig to segfault
212 [20018] network: getaddrinfo should reject IP addresses with trailing
213 characters (CVE-2016-10739)
214 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
215 [20271] libc: Missing "\n" in __libc_fatal calls
216 [20480] dynamic-link: Patch: ifunc not executable, crashes sudo qemu
217 [20544] libc: RFE: atexit, __cxa_atexit, on_exit should assert function
218 pointer argument is non-NULL
219 [21037] stdio: open_memstream and freopen
220 [21286] libc: bits/siginfo.h is missing enum definition for TRAP_HWBKPT
221 [21716] time: Crash in glibc's mktime in low-memory situations
222 [22834] stdio: Subprocess forked by popen may crash in Linux when
223 multithreads call popen
224 [22927] network: crash in vn_gai_enqueue_request if requests_tail was NULL
225 and pthread_create fails.
226 [23032] hurd: sysdeps/htl/pt-barrier-init.c:39: bad call to memcmp ?
227 [23125] libc: riscv64: endless loop when throwing an exception from a
229 [23275] nptl: Race in pthread_mutex_lock while promoting to
230 PTHREAD_MUTEX_ELISION_NP.
231 [23400] libc: stdlib/test-bz22786.c creates temporary files in glibc
233 [23479] math: [mips] bits/fenv.h should not define some macros for soft-
235 [23490] libc: sysdeps/unix/sysv/linux/x86/tst-cet-property-2.c:49: off by
237 [23497] libc: readdir64@GLIBC_2.1 cannot parse the kernel directory stream
238 [23509] dynamic-link: CET enabled glibc is incompatible with the older
240 [23520] nscd: nscd: Use-after-free in addgetnetgrentX and its callers
241 [23521] nss: get_next_alias nss_files file stream leak
242 [23538] nptl: Hang in pthread_cond_broadcast
243 [23562] libc: Wrong type for si_band in Linux-specific siginfo_t
244 [23578] regex: Invalid memory access if regex pattern contains NUL byte
245 [23579] libc: Errors misreported in preadv2
246 [23597] build: support/test-container.c doesn't work with different
248 [23603] time: mktime signed integer overflow on large timestamps
249 [23606] libc: Missing ENDBR32 in sysdeps/i386/start.S
250 [23614] libc: powerpc: missing CFI register information in __mpn_*
252 [23637] string: Generic strstr/strcasestr fails with huge needles
253 [23640] libc: no way to easily clear FD_CLOEXEC in
254 posix_spawn_file_actions_adddup2()
255 [23649] libc: [microblaze/mips/nios2/riscv] sys/procfs.h pr_uid, pr_gid
257 [23656] libc: [mips n32] sys/procfs.h pr_sigpend, pr_sighold, pr_flag have
259 [23679] libc: gethostid: Missing NULL check for gethostbyname_r result
260 [23689] libc: Bug in documentation for rusage.ru_ixrss in
261 bits/types/struct_rusage.h
262 [23690] dynamic-link: Segfault in _dl_profile_fixup with a high number of
264 [23707] dynamic-link: Missing unwind info in sysdeps/powerpc/powerpc32/dl-
266 [23709] string: glibc 2.25 lacks sse2 optimized strstr()
267 [23716] dynamic-link: _dl_runtime_resolve_shstk isn't selected properly
268 [23717] libc: glibc: stdlib/tst-setcontext9 test suite failure on
270 [23724] localedata: Albanian date formats are incorrect
271 [23735] math: libnldbl_nonshared.a references internal libm symbols
272 [23740] localedata: kl_GL: Month names and date formats need update
273 [23744] regex: regex refactorings to remove BE, avoid duplication
274 [23745] time: mktime fix for Gnulib + coreutils
275 [23758] time: Improve the width of alternate representation for year in
277 [23783] libc: [mips] Missing CMSPAR bits/termios.h
278 [23789] time: mktime does not set errno on failure
279 [23791] localedata: Wrong monetary format for ca_ES locale
280 [23793] locale: c32rtomb and mbrtoc32 should not alias wcrtomb and mbrtowc
281 [23794] locale: c16rtomb does not handle surrogate pairs
282 [23821] libc: si_band in siginfo_t has wrong type long int on sparc64
283 [23822] math: ia64 static libm.a is missing exp2f, log2f and powf symbols
284 [23836] time: time/tst-mktime2 test failure on Arm (32-bit)
285 [23848] libc: [sparc] Some socket syscalls wrongly assumed to be present
286 [23861] nptl: rdlock stalls indefinitely on an unlocked pthread rwlock
287 [23862] libc: [sh] missing kernel-features.h undefines
288 [23864] libc: [riscv] missing kernel-features.h undefines
289 [23867] libc: [arm/microblaze] __ASSUME_MLOCK2 incorrect
290 [23907] malloc: Incorrect double-free malloc tcache check disregards
292 [23913] libc: off-by-one in function maybe_script_execute in
293 sysdeps/posix/spawni.c
294 [23915] libc: [arm] __ASSUME_COPY_FILE_RANGE incorrect
295 [23923] locale: Add --no-hard-links option to localedef
296 [23927] network: Linux if_nametoindex() does not close descriptor
298 [23961] math: powf can overflow to inf without setting errno in non-
299 nearest rounding mode
300 [23967] libc: [2.28 Regression]: New sigaction implementation breaks m68k
301 [23972] libc: __old_getdents64 uses wrong d_off value on overflow
302 [23993] libc: glibc 2.29 doesn't build with gcc 4.9
303 [23995] localedata: Remove execution flags from localedata/locales/bi_VU
304 [24011] localedata: Fixed small type in comment for locale bs_BA
305 [24018] libc: gettext() may return NULL
306 [24022] build: riscv build failure with Linux kernel 4.20-rc7
307 [24023] build: [2.29 Regression] FAIL: elf/check-localplt
308 [24024] string: strerror() might set errno to ENOMEM due to -fno-math-
310 [24027] malloc: glibc: realloc() ncopies 32-bit integer overflow
311 [24034] libc: tst-cancel21-static fails with SIGBUS on pre-ARMv7 when
313 [24046] localedata: en_US locale doesn't define date_fmt
314 [24063] manual: @var{errno} should be @code{errno}
315 [24066] soft-fp: Inconsistent _FP_W_TYPE_SIZE check
316 [24088] libc: VSCR field is not being correctly read in ucontext_t on
318 [24096] time: Specifying '_' or '-' flag for "%EY" does not produce the
320 [24097] string: Can't use 64-bit register for size_t in assembly codes for
322 [24110] hurd: SS_DISABLE never set in stack_t value returned by
324 [24112] network: Do not send DNS queries for non-host names (where all
325 answers will be rejected)
326 [24130] libc: alpha __remqu corrupts $f3 register
333 * The localization data for ISO 14651 is updated to match the 2016
334 Edition 4 release of the standard, this matches data provided by
335 Unicode 9.0.0. This update introduces significant improvements to the
336 collation of Unicode characters. This release deviates slightly from
337 the standard in that the collation element ordering for lowercase and
338 uppercase LATIN script characters is adjusted to ensure that regular
339 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
340 is not matched by [a-z]. With the update many locales have been
341 updated to take advantage of the new collation information. The new
342 collation information has increased the size of the compiled locale
343 archive or binary locales.
345 * The GNU C Library can now be compiled with support for Intel CET, AKA
346 Intel Control-flow Enforcement Technology. When the library is built
347 with --enable-cet, the resulting glibc is protected with indirect
348 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
349 compatible with all existing executables and shared libraries. This
350 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
351 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
352 capable of multi-byte NOPs, like x86-64 processors as well as Intel
353 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
354 x86_64 and x32 on non-CET processors. --enable-cet has been tested
355 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
358 * The GNU C Library now has correct support for ABSOLUTE symbols
359 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
360 relocated incorrectly or in some cases discarded. The GNU linker can
361 make use of the newer semantics, but it must communicate it to the
362 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
363 field) to indicate such support is required.
365 * Unicode 11.0.0 Support: Character encoding, character type info, and
366 transliteration tables are all updated to Unicode 11.0.0, using
367 generator scripts contributed by Mike FABIAN (Red Hat).
369 * <math.h> functions that round their results to a narrower type are added
370 from TS 18661-1:2014 and TS 18661-3:2015:
372 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
375 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
378 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
381 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
384 * Two grammatical forms of month names are now supported for the following
385 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
386 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
387 support two grammatical forms in abbreviated month names: Catalan, Greek,
390 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
391 include the support for two grammatical forms of month names.
393 * Building and running on GNU/Hurd systems now works without out-of-tree
396 * The renameat2 function has been added, a variant of the renameat function
397 which has a flags argument. If the flags are zero, the renameat2 function
398 acts like renameat. If the flag is not zero and there is no kernel
399 support for renameat2, the function will fail with an errno value of
400 EINVAL. This is different from the existing gnulib function renameatu,
401 which performs a plain rename operation in case of a RENAME_NOREPLACE
402 flags and a non-existing destination (and therefore has a race condition
403 that can clobber the destination inadvertently).
405 * The statx function has been added, a variant of the fstatat64
406 function with an additional flags argument. If there is no direct
407 kernel support for statx, glibc provides basic stat support based on
408 the fstatat64 function.
410 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
411 library if installed. libidn2 version 2.0.5 or later is recommended. If
412 libidn2 is not available, internationalized domain names are not encoded
413 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
414 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
415 with an encoding error.) Flags which used to change the IDN encoding and
416 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
417 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
418 deprecated. They no longer have any effect.
420 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
421 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
422 range of ELF gABI expressions including such constructs as
423 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
424 have been further restricted, and where in the past a dynamic string
425 token sequence may have been interpreted as a literal string it will
426 now cause a load failure. These load failures were always considered
427 unspecified behaviour from the perspective of the dynamic loader, and
428 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
429 results in a load failure now.
431 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
432 implementation includes all the standard functions provided by
435 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
436 thrd_detach, thrd_exit, and thrd_join for thread management.
438 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
439 mtx_destroy for mutual exclusion.
441 - call_once for function call synchronization.
443 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
444 cnd_wait for conditional variables.
446 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
448 Application developers must link against libpthread to use ISO C threads.
450 Deprecated and removed features, and other changes affecting compatibility:
452 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
453 installed. Software that was using either header should be updated to
454 use standard <stdio.h> interfaces instead.
456 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
457 This was never required by the C standard, and the macros just expanded
458 to call alternative names for the same functions. If you hoped getc and
459 putc would provide performance improvements over fgetc and fputc, instead
460 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
461 necessary, flockfile and funlockfile.
463 * All stdio functions now treat end-of-file as a sticky condition. If you
464 read from a file until EOF, and then the file is enlarged by another
465 process, you must call clearerr or another function with the same effect
466 (e.g. fseek, rewind) before you can read the additional data. This
467 corrects a longstanding C99 conformance bug. It is most likely to affect
468 programs that use stdio to read interactive input from a terminal.
471 * The macros 'major', 'minor', and 'makedev' are now only available from
472 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
473 headers that happen to include <sys/types.h>. These macros are rarely
474 used, not part of POSIX nor XSI, and their names frequently collide with
475 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
478 <sys/sysmacros.h> is a GNU extension. Portable programs that require
479 these macros should first include <sys/types.h>, and then include
480 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
482 * The tilegx*-*-linux-gnu configurations are no longer supported.
484 * The obsolete function ustat is no longer available to newly linked
485 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
486 function has been deprecated in favor of fstatfs and statfs.
488 * The obsolete function nfsservctl is no longer available to newly linked
489 binaries. This function was specific to systems using the Linux kernel
490 and could not usefully be used with the GNU C Library on systems with
491 version 3.1 or later of the Linux kernel.
493 * The obsolete function name llseek is no longer available to newly linked
494 binaries. This function was specific to systems using the Linux kernel
495 and was not declared in a header. Programs should use the lseek64 name
496 for this function instead.
498 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
499 getaddrinfo and getnameinfo functions have been deprecated. The behavior
500 previously selected by them is now always enabled.
502 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
503 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
504 restriction (rejecting '_' in host names, among other things) has been
505 removed, for increased compatibility with non-IDN name resolution.
507 * The fcntl function now have a Long File Support variant named fcntl64. It
508 is added to fix some Linux Open File Description (OFD) locks usage on non
509 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
510 fcntl and LFS support is handled transparently. Also for Linux, the OFD
511 locks act as a cancellation entrypoint.
513 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
514 ecb_crypt, and des_setparity are no longer available to newly linked
515 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
516 longer installed. These functions encrypted and decrypted data with the
517 DES block cipher, which is no longer considered secure. Software that
518 still uses these functions should switch to a modern cryptography library,
521 * Reflecting the removal of the encrypt and setkey functions above, the
522 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
523 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
526 * The obsolete function fcrypt is no longer available to newly linked
527 binaries. It was just another name for the standard function crypt,
528 and it has not appeared in any header file in many years.
530 * We have tentative plans to hand off maintenance of the passphrase-hashing
531 library, libcrypt, to a separate development project that will, we hope,
532 keep up better with new passphrase-hashing algorithms. We will continue
533 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
534 'crypt_r' should not need to change at all; however, distributions will
535 need to install <crypt.h> and libcrypt from a separate project.
537 In this release, if the configure option --disable-crypt is used, glibc
538 will not install <crypt.h> or libcrypt, making room for the separate
539 project's versions of these files. The plan is to make this the default
540 behavior in a future release.
542 Changes to build and runtime requirements:
544 GNU make 4.0 or later is now required to build glibc.
546 Security related changes:
548 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
549 been fixed by removing the glibc-internal IDNA implementation and using
550 the system-provided libidn2 library instead. Originally reported by Hanno
551 Böck and Christian Weisgerber.
553 CVE-2017-18269: An SSE2-based memmove implementation for the i386
554 architecture could corrupt memory. Reported by Max Horn.
556 CVE-2018-11236: Very long pathname arguments to realpath function could
557 result in an integer overflow and buffer overflow. Reported by Alexey
560 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
561 architecture could write beyond the target buffer, resulting in a buffer
562 overflow. Reported by Andreas Schwab.
564 The following bugs are resolved with this release:
566 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
567 [6889] manual: 'PWD' mentioned but not specified
568 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
569 when __WORDSIZE != 64
570 [13762] regex: re_search etc. should return -2 on memory exhaustion
571 [13888] build: /tmp usage during testing
572 [13932] math: dbl-64 pow unexpectedly slow for some inputs
573 [14092] nptl: Support C11 threads
574 [14095] localedata: Review / update collation data from Unicode / ISO
576 [14508] libc: -Wformat warnings
577 [14553] libc: Namespace pollution loff_t in sys/types.h
578 [14890] libc: Make NT_PRFPREG canonical.
579 [15105] libc: Extra PLT references with -Os
580 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
582 [16335] manual: Feature test macro documentation incomplete and out of
584 [16552] libc: Unify umount implementations in terms of umount2
585 [17082] libc: htons et al.: statement-expressions prevent use on global
586 scope with -O1 and higher
587 [17343] libc: Signed integer overflow in /stdlib/random_r.c
588 [17438] localedata: pt_BR: wrong d_fmt delimiter
589 [17662] libc: please implement binding for the new renameat2 syscall
590 [17721] libc: __restrict defined as /* Ignore */ even in c11
591 [17979] libc: inconsistency between uchar.h and stdint.h
592 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
593 [18023] libc: extend_alloca is broken (questionable pointer comparison,
594 horrible machine code)
595 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
597 [18471] libc: llseek should be a compat symbol
598 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
600 [18991] nss: nss_files skips large entry in database
601 [19239] libc: Including stdlib.h ends up with macros major and minor being
603 [19463] libc: linknamespace failures when compiled with -Os
604 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
605 [19527] locale: Normalized charset name not recognized by setlocale
606 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
607 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
609 [19728] network: out of bounds stack read in libidn function
610 idna_to_ascii_4i (CVE-2016-6261)
611 [19729] network: out of bounds heap read on invalid utf-8 inputs in
612 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
613 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
615 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
616 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
617 [20419] dynamic-link: files with large allocated notes crash in
619 [20530] libc: bswap_16 should use __builtin_bswap16() when available
620 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
621 [20980] manual: CFLAGS environment variable replaces vital options
622 [21163] regex: Assertion failure in pop_fail_stack when executing a
623 malformed regexp (CVE-2015-8985)
624 [21234] manual: use of CFLAGS makes glibc detect no optimization
625 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
626 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
627 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
628 [21508] locale: intl/tst-gettext failure with latest msgfmt
629 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
630 [21812] network: getifaddrs() returns entries with ifa_name == NULL
631 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
633 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
635 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
636 [22247] network: Integer overflow in the decode_digit function in
637 puny_decode.c in libidn (CVE-2017-14062)
638 [22342] nscd: NSCD not properly caching netgroup
639 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
640 [22550] localedata: es_ES locale (and other es_* locales): collation
641 should treat ñ as a primary different character, sync the collation
642 for Spanish with CLDR
643 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
644 by gcc configured with --enable-default-pie
645 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
646 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
647 crossing 2GB threshold (CVE-2017-18269)
648 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
650 [22735] time: Misleading typo in time.h source comment regarding
652 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
653 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
654 `__assert_perror_fail ()' to `__assert_fail_base ()'
655 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
657 [22786] libc: Stack buffer overflow in realpath() if input size is close
658 to SSIZE_MAX (CVE-2018-11236)
659 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
660 through an absolute DT_NEEDED path
661 [22792] build: tcb-offsets.h dependency dropped
662 [22797] libc: pkey_get() uses non-reserved name of argument
663 [22807] libc: PTRACE_* constants missing for powerpc
664 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
665 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
666 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
667 [22848] localedata: ca_ES: update date definitions from CLDR
668 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
669 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
670 [22896] localedata: Update locale data for an_ES
671 [22902] math: float128 test failures with GCC 8
672 [22918] libc: multiple common of `__nss_shadow_database'
673 [22919] libc: sparc32: backtrace yields infinite backtrace with
675 [22926] libc: FTBFS on powerpcspe
676 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
678 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
679 [22947] libc: FAIL: misc/tst-preadvwritev2
680 [22963] localedata: cs_CZ: Add alternative month names
681 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
682 [22996] localedata: change LC_PAPER to en_US in es_BO locale
683 [22998] dynamic-link: execstack tests are disabled when SELinux is
685 [23005] network: Crash in __res_context_send after memory allocation
687 [23007] math: strtod cannot handle -nan
688 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
689 [23036] regex: regex equivalence class regression
690 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
691 [23069] libc: sigaction broken on riscv64-linux-gnu
692 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
693 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
694 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
695 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
696 and libc build with -Os)
697 [23140] localedata: More languages need two forms of month names
698 [23145] libc: _init/_fini aren't marked as hidden
699 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
700 [23171] math: C++ iseqsig for long double converts arguments to double
701 [23178] nscd: sudo will fail when it is run in concurrent with commands
702 that changes /etc/passwd
703 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
705 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
706 [23208] localedata: New locale - Lower Sorbian (dsb)
707 [23233] regex: Memory leak in build_charclass_op function in file
709 [23236] stdio: Harden function pointers in _IO_str_fields
710 [23250] nptl: Offset of __private_ss differs from GCC
711 [23253] math: tgamma test suite failures on i686 with -march=x86-64
712 -mtune=generic -mfpmath=sse
713 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
715 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
716 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
718 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
719 [23277] math: nan function should not have const attribute
720 [23279] math: scanf and strtod wrong for some hex floating-point
721 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
723 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
724 [23303] build: undefined reference to symbol
725 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
726 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
728 [23313] stdio: libio vtables validation and standard file object
730 [23329] libc: The __libc_freeres infrastructure is not properly run across
732 [23349] libc: Various glibc headers no longer compatible with
734 [23351] malloc: Remove unused code related to heap dumps and malloc
736 [23363] stdio: stdio-common/tst-printf.c has non-free license
737 [23396] regex: Regex equivalence regression in single-byte locales
738 [23422] localedata: oc_FR: More updates of locale data
739 [23442] build: New warning with GCC 8
740 [23448] libc: Out of bounds access in IBM-1390 converter
741 [23456] libc: Wrong index_cpu_LZCNT
742 [23458] build: tst-get-cpu-features-static isn't added to tests
743 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
745 [23467] dynamic-link: x86/CET: A property note parser bug
752 * The GNU C Library can now be compiled with support for building static
753 PIE executables (See --enable-static-pie in INSTALL). These static PIE
754 executables are like static executables but can be loaded at any address
755 and provide additional security hardening benefits at the cost of some
756 memory and performance. When the library is built with --enable-static-pie
757 the resulting libc.a is usable with GCC 8 and above to create static PIE
758 executables using the GCC option '-static-pie'. This feature is currently
759 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
760 aarch64 with binutils 2.30 or later.
762 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
763 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
766 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
768 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
770 * In order to support faster and safer process termination the malloc API
771 family of functions will no longer print a failure address and stack
772 backtrace after detecting heap corruption. The goal is to minimize the
773 amount of work done after corruption is detected and to avoid potential
774 security issues in continued process execution. Reducing shutdown time
775 leads to lower overall process restart latency, so there is benefit both
776 from a security and performance perspective.
778 * The abort function terminates the process immediately, without flushing
779 stdio streams. Previous glibc versions used to flush streams, resulting
780 in deadlocks and further data corruption. This change also affects
781 process aborts as the result of assertion failures.
783 * On platforms where long double has the IEEE binary128 format (aarch64,
784 alpha, mips64, riscv, s390 and sparc), the math library now implements
785 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
786 These are the same interfaces added in version 2.26 for some platforms where
787 this format is supported but is not the format of long double.
789 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
790 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
791 implements interfaces for that type, as defined by ISO/IEC TS
792 18661-3:2015. These are corresponding interfaces to those supported for
795 * The math library now implements interfaces for the _Float32, _Float64 and
796 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
797 corresponding interfaces to those supported for _Float128.
799 * glibc now implements the memfd_create and mlock2 functions on Linux.
801 * Support for memory protection keys was added. The <sys/mman.h> header now
802 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
805 * The copy_file_range function was added.
807 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
809 * The ldconfig utility now processes `include' directives using the C/POSIX
810 collation ordering. Previous glibc versions used locale-specific
811 ordering, the change might break systems that relied on that.
813 * Support for two grammatical forms of month names has been added.
814 In a call to strftime, the "%B" and "%b" format specifiers will now
815 produce the grammatical form required when the month is used as part
816 of a complete date. New "%OB" and "%Ob" specifiers produce the form
817 required when the month is named by itself. For instance, in Greek
818 and in many Slavic and Baltic languages, "%B" will produce the month
819 in genitive case, and "%OB" will produce the month in nominative case.
821 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
822 are all valid and will all accept any known form of month
823 name---standalone or complete, abbreviated or full. In a call to
824 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
825 the strings used by "%B" and "%b", respectively. New query
826 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
827 used by "%OB" and "%Ob", respectively.
829 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
830 define the strings for %OB and %Ob, respectively; these have the
831 same syntax as "mon" and "abmon". These arrays are optional; if they
832 are not provided then they have the same content as "mon" and "abmon",
835 These features are provided for locales which define "alt_mon" and/or
836 "ab_alt_mon" in their locale source data. This release includes such
837 alternative month name data for the following languages: Belarusian,
838 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
840 This feature is currently a GNU extension, but it is expected to
841 be added to the next revision of POSIX, and it is also already
842 available on some BSD-derived operating systems.
844 This feature will cause existing statically compiled applications
845 to fail to load locales and fall back to the builtin C/POSIX locales.
846 See notes below for other changes affecting compatibility.
848 * Support for the RISC-V ISA running on Linux has been added. This port
849 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
850 for the following ISA and ABI pairs:
856 Deprecated and removed features, and other changes affecting compatibility:
858 * Statically compiled applications attempting to load locales compiled for the
859 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
860 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
861 support for two grammatical forms of the month names, also extends the locale
862 data binary format. Static applications needing locale support must be
863 recompiled to match the runtime and data they are deployed with. In some
864 distributions there is an upgrade window where dynamically linked applications
865 may use a new library but the old locale data and also fall back to the
866 builtin C/POSIX locales; restarting the application process is sufficient to
869 * Support for statically linked applications which call dlopen is deprecated
870 and will be removed in a future version of glibc. Applications which call
871 dlopen need to be linked dynamically instead.
873 * Support for old programs which use internal stdio data structures and
874 functions is deprecated. This includes programs which use the C++ streams
875 provided by libstdc++ in GCC 2.95. Programs which use the internal
876 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
877 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
878 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
879 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
880 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
881 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
882 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
883 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
884 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
885 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
886 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
887 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
888 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
889 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
890 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
891 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
892 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
893 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
894 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
895 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
896 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
897 future version of glibc. Unlike other symbol removals, these old
898 applications will not be supported using compatibility symbols.
900 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
901 defined by <sys/ptrace.h>.
903 * libm no longer supports SVID error handling (calling a user-provided
904 matherr function on error) or the _LIB_VERSION variable to control error
905 handling. (SVID error handling and the _LIB_VERSION variable still work
906 for binaries linked against older versions of the GNU C Library.) The
907 libieee.a library is no longer provided. math.h no longer defines struct
908 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
909 TLOSS, PLOSS and HUGE.
911 * The libm functions pow10, pow10f and pow10l are no longer supported for
912 new programs. Programs should use the standard names exp10, exp10f and
913 exp10l for these functions instead.
915 * The mcontext_t type is no longer the same as struct sigcontext. On
916 platforms where it was previously the same, this changes the C++ name
917 mangling for interfaces involving this type.
919 * The add-ons mechanism for building additional packages at the same time as
920 glibc has been removed. The --enable-add-ons configure option is now
923 * The --without-fp configure option is now ignored. Whether hardware
924 floating-point instructions are used is now configured based on whether
925 the compiler used at configure time (without any options implied by a
926 --with-cpu= configure option) uses such instructions.
928 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
929 the specified string can be parsed as a domain name.
931 * In the malloc_info output, the <heap> element may contain another <aspace>
932 element, "subheaps", which contains the number of sub-heaps.
934 * The libresolv function p_secstodate is no longer supported for new
937 * The tilepro-*-linux-gnu configuration is no longer supported.
939 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
940 and will be removed in a future release. Software that is still using
941 either header should be updated to use standard <stdio.h> interfaces
944 libio.h was originally the header for a set of supported GNU extensions,
945 but they have not been maintained as such in many years, they are now
946 standing in the way of improvements to stdio, and we don't think there are
947 any remaining external users. _G_config.h was never intended for public
948 use, but predates the bits convention.
950 Changes to build and runtime requirements:
952 * bison version 2.7 or later is required to generate code in the 'intl'
955 Security related changes:
957 CVE-2009-5064: The ldd script would sometimes run the program under
958 examination directly, without preventing code execution through the
959 dynamic linker. (The glibc project disputes that this is a security
960 vulnerability; only trusted binaries must be examined using the ldd
963 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
964 suffered from a one-byte overflow during ~ operator processing (either
965 on the stack or the heap, depending on the length of the user name).
966 Reported by Tim Rühsen.
968 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
969 would sometimes fail to free memory allocated during ~ operator
970 processing, leading to a memory leak and, potentially, to a denial
973 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
974 without GLOB_NOESCAPE, could write past the end of a buffer while
975 unescaping user names. Reported by Tim Rühsen.
977 CVE-2017-17426: The malloc function, when called with an object size near
978 the value SIZE_MAX, would return a pointer to a buffer which is too small,
979 instead of NULL. This was a regression introduced with the new malloc
980 thread cache in glibc 2.26. Reported by Iain Buclaw.
982 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
983 to the allocation of too much memory. (This is not a security bug per se,
984 it is mentioned here only because of the CVE assignment.) Reported by
987 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
988 of the number of search path components. (This is not a security
989 vulnerability per se because no trust boundary is crossed if the fix for
990 CVE-2017-1000366 has been applied, but it is mentioned here only because
991 of the CVE assignment.) Reported by Qualys.
993 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
994 for AT_SECURE or SUID binaries could be used to load libraries from the
997 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
998 succeeds without returning an absolute path due to unexpected behaviour
999 of the Linux kernel getcwd syscall. Reported by halfdog.
1001 CVE-2018-6485: The posix_memalign and memalign functions, when called with
1002 an object size near the value of SIZE_MAX, would return a pointer to a
1003 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
1005 CVE-2018-6551: The malloc function, when called with an object size near
1006 the value of SIZE_MAX, would return a pointer to a buffer which is too
1007 small, instead of NULL.
1009 The following bugs are resolved with this release:
1011 [866] glob: glob should match dangling symlinks
1012 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
1013 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
1015 [5997] math: Very slow execution of sinf function
1016 [10580] localedata: hr_HR: updated locale
1017 [10871] locale: 'mon' array should contain both nominative and genitive
1019 [12349] localedata: eu_ES: incorrect thousands separator
1020 [13605] localedata: shn_MM: new Shan locale
1021 [13805] localedata: ru_RU: currency should use ',' as radix point
1022 [13953] localedata: km_KH: locale update
1023 [13994] localedata: mjw_IN: new locale
1024 [14121] build: make writes .mo files in po directory
1025 [14333] libc: Fix the race between atexit() and exit()
1026 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
1027 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
1029 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
1030 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
1031 width Latin characters
1032 [15332] localedata: es_CU: locale update
1033 [15436] stdio: Don't close or flush stdio streams on abort
1034 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
1036 [16148] localedata: ca_ES: incorrect thousands separator
1037 [16750] dynamic-link: ldd should not try to execute the binaries
1039 [16777] localedata: pl_PL: incorrect thousands separator in locale
1040 [16905] localedata: hanzi: new collation
1041 [17563] localedata: cmn_TW: add hanzi collation
1042 [17750] localedata: wrong collation order of diacritics in most locales
1043 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
1045 [17956] build: Build fails on missing definitions from header file
1046 nss/nss.h when Mozilla NSS is used for cryptography
1047 [18203] libc: realpath() does not handle unreachable paths correctly
1048 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
1049 [18812] localedata: kab_DZ: new Kabyle Algeria locale
1050 [18822] libc: Internal functions are called via PLT
1051 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
1052 [19170] libc: __gmon_start__ defined in hppa in crtn.S
1053 [19574] libc: glibc should support building static PIE binaries
1054 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
1056 [19971] glob: glob: Do not skip entries with zero d_ino values
1057 [19982] localedata: fr.po: spelling mistake for error code EXDEV
1058 [20008] localedata: km_KH: convert to translit_neutral
1059 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
1060 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
1061 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
1062 [20482] localedata: de_CH: abbreviated weekdays should be two letters
1063 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
1065 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
1066 they're set, wrong errors returned by gaih_inet when lookup functions
1068 [20756] localedata: [PATCH] Use Unicode wise thousands separator
1069 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
1071 [20952] localedata: yuw_PG: new locale
1072 [21084] localedata: charmaps/IBM858: new codepage
1073 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
1074 [21242] libc: assert gives pedantic warning in old gcc versions
1075 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
1076 __regcall calling convention
1077 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
1078 [21326] libc: C99 functions are not declared for C++11 and later if
1079 _GNU_SOURCE is not predefined by g++
1080 [21457] libc: sys/ucontext.h namespace
1081 [21530] libc: tmpfile() should be implemented using O_TMPFILE
1082 [21660] math: GCC fails to compile a formula with tgmath.h
1083 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
1084 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
1085 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
1086 [21684] math: tgmath.h handling of complex integers
1087 [21685] math: tgmath.h handling of bit-fields
1088 [21686] math: tgmath.h handling of __int128
1089 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1091 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
1093 [21750] localedata: column width of characters incompatible with classical
1095 [21754] malloc: malloc: Perform as little work as possible after heap
1096 consistency check failures
1097 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
1098 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
1099 [21791] string: Unused XXX_chk_XXX functions in libc.a
1100 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
1102 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
1104 [21845] localedata: Added new Locale bho_NP
1105 [21853] localedata: Fix abday Which looks same as day in zh_SG
1106 [21854] localedata: Added New Locale en_SC
1107 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
1108 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
1109 _dl_runtime_resolve_avx_slow
1110 [21885] network: getaddrinfo: gethosts does not release resolver context
1111 on memory allocation failure
1112 [21899] libc: XPG4.2 sigaction namespace
1113 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
1115 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
1117 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
1118 errno value, causing getaddrinfo to fail
1119 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
1120 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
1121 instead of EAI_NODATA
1122 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
1123 constant PTRACE_SEIZE_DEVEL
1124 [21930] math: C-only gcc builtins used in <math.h> isinf
1125 [21932] network: Unpaired __resolv_context_get in generic get*_r
1127 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
1128 [21944] libc: sigval namespace
1129 [21951] localedata: Update hanzi collation by stroke
1130 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
1131 sysdeps/x86_64/fpu/e_expf.S
1132 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
1133 allocation in syscall loops)
1134 [21959] localedata: Fix Country name for xh_ZA
1135 [21960] localedata: Fix abmon for bem_ZM
1136 [21966] math: AVX2 mathvec functions use FMA without checking
1137 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
1138 [21971] localedata: Added New Locale for mfe_MU
1139 [21972] libc: assert macro requires operator== (int) for its argument type
1140 [21973] math: [sparc] libm missing sqrtl compat symbol
1141 [21974] libc: Remove __bb_init_func and __bb_exit_func
1142 [21982] string: stratcliff.c: error: assuming signed overflow does not
1144 [21986] stdio: __guess_grouping is called incorrectly
1145 [21987] math: [sparc32] wrong bits/long-double.h installed
1146 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
1148 [22022] localedata: Missing country_name for mni_IN
1149 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
1151 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
1153 [22026] locale: iconv_open: heap overflow on gconv_init failure
1154 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
1155 [22035] math: [m68k] bits/math-inline.h macro namespace
1156 [22038] localedata: Fix abbreviated weeks and months for Somali
1157 [22044] localedata: Remove redundant data for Limburgish Language
1158 [22050] malloc: Linking with -lmcheck does not hook
1159 __malloc_initialize_hook correctly
1160 [22051] libc: zero terminator in the middle of glibc's .eh_frame
1161 [22052] malloc: malloc failed to compile with GCC 7 and -O3
1162 [22070] localedata: charmaps/UTF-8: wcwidth for
1163 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
1164 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
1165 Jungseong and Jongseong) should be 0
1166 [22078] nss: nss_files performance issue in multi mode
1167 [22082] math: bits/math-finite.h exp10 condition
1168 [22086] libc: pcprofiledump incorrect cross-endian condition
1169 [22093] dynamic-link: ld.so no longer searches in .../x86_64
1170 [22095] network: Name server address allocation memory leak in resolv.conf
1172 [22096] network: __resolv_conf_attach can incorrectly free passed conf
1174 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
1175 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
1177 [22111] malloc: malloc: per thread cache is not returned when thread exits
1178 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
1179 [22134] libc: [linux] implement fexecve with execveat
1180 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
1182 [22145] libc: ttyname() gives up too early in the face of namespaces
1183 [22146] math: C++ build issue with float128 on x86_64
1184 [22153] nptl: nptl: save error code before process termination
1185 [22156] libc: [hppa,ia64,microblaze] Executable stack default
1186 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
1187 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
1188 [22165] libc: [hppa] Text relocations in libc.so
1189 [22180] libc: destructor registered via __cxa_atexit is called twice
1190 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
1191 symlinks") cause "make" segfaults
1192 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
1194 [22207] libc: FAIL: stdlib/test-atexit-race
1195 [22225] math: nearbyint arithmetic moved before feholdexcept
1196 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
1197 [22235] math: iscanonical in C++ and float128
1198 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
1199 the svid compat wrapper
1200 [22244] math: ynf and yn are wrong without the svid compat wrapper
1201 [22273] libc: Improper assert in Linux posix_spawn implementation
1202 [22284] libc: -pg -pie doesn't work
1203 [22292] locale: localedef exits with error 4 when it should be error 1
1204 [22294] locale: Allow "" for int_currency_symbol definition in locales.
1205 [22295] locale: Don't warn on non-symbolic characters in locale sources in
1207 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
1208 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
1210 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
1211 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
1212 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
1213 [22322] libc: [mips64] wrong bits/long-double.h installed
1214 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
1215 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
1217 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
1218 [22343] malloc: Integer overflow in posix_memalign
1219 [22347] libc: getrandom() returns the number of bytes that were copied to
1220 the buffer even though the comments say "Return 0 on success and -1 on
1222 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1223 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1225 [22370] dynamic-link: Incorrect note padding check
1226 [22375] libc: malloc returns pointer from tcache_get when should return
1227 NULL (CVE-2017-17426)
1228 [22377] math: iseqsig, float128 and C++
1229 [22382] localedata: Error in tpi_PG locale
1230 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1231 inside the ASCII printable range
1232 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1234 [22403] localedata: Slash needs escaping in some locales
1235 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1236 [22409] network: res_hnok does not accept some host names used on the
1238 [22412] network: res_dnok, res_hnok should perform syntax checks
1239 [22413] network: ns_name_pton ignores syntactically invalid trailing
1241 [22415] stdio: setvbuf can lead to invalid free/segfault
1242 [22432] build: Non-deterministic build
1243 [22439] malloc: malloc_info should compute summary statistics for all sub-
1245 [22442] network: if_nametoindex could report index for the wrong
1246 networking interface
1247 [22446] build: aliasing violation calling readlink in handle_request
1248 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1249 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1250 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1251 __nscd_hash/__nss_hash
1252 [22463] network: p_secstodate overflow handling
1253 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1254 [22478] libc: sigwait can fail with EINTR
1255 [22505] libc: ldconfig processes include directive in locale-specific
1257 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1258 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1259 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1260 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1261 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1262 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1263 same as for Croatian
1264 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1265 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1266 [22577] libc: missing newline after "cannot allocate TLS data structures
1268 [22588] manual: manual/conf.texi: missing underscore in front of
1270 [22593] math: nextafter and nexttoward are declared with const attribute
1271 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1272 [22603] string: ia64 memchr overflows internal pointer check
1273 [22605] libc: SH clone does not set the exit code correctly
1274 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1276 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1277 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1278 modified in case of success
1279 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1280 [22615] manual: manual: ambiguous wording about errno value in case of
1282 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1284 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1285 binaries or if /proc is not mounted (CVE-2017-16997)
1286 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1287 [22630] build: $(no-pie-ldflag) is no longer effective
1288 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1289 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1290 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1291 [22637] nptl: guard size is subtracted from thread stack size instead of
1293 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1294 [22657] localedata: hu_HU: Avoid double space in date
1295 [22660] math: fmax, fmin sNaN handling on alpha
1296 [22664] libc: New warning of GCC8
1297 [22665] math: alpha: ceil and floor raise inexact exceptions
1298 [22666] math: alpha: trunc raise inexact exceptions
1299 [22667] libc: makecontext lacks stack alignment on i386
1300 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1301 [22679] libc: getcwd(3) can succeed without returning an absolute path
1303 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1304 elision and tunables
1305 [22687] math: [powerpc-nofpu] complex long double functions spurious
1307 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1308 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1309 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1311 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1312 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1314 [22701] nis: Incomplete removal of libnsl
1315 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1316 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1317 [22715] dynamic-link: FAIL: elf/tst-audit10
1318 [22719] libc: Backtrace tests fail on hppa
1319 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1320 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1321 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1322 before the first call to crypt_r () in crypt/badsalttest.c
1329 * A per-thread cache has been added to malloc. Access to the cache requires
1330 no locks and therefore significantly accelerates the fast path to allocate
1331 and free small amounts of memory. Refilling an empty cache requires locking
1332 the underlying arena. Performance measurements show significant gains in a
1333 wide variety of user workloads. Workloads were captured using a special
1334 instrumented malloc and analyzed with a malloc simulator. Contributed by
1335 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1337 * Unicode 10.0.0 Support: Character encoding, character type info, and
1338 transliteration tables are all updated to Unicode 10.0.0, using
1339 generator scripts contributed by Mike FABIAN (Red Hat).
1340 These updates cause user visible changes, especially the changes in
1341 wcwidth for many emoji characters cause problems when emoji sequences
1342 are rendered with pango, see for example:
1343 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1345 * Collation of Hungarian has been overhauled and is now consistent with "The
1346 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1349 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1351 - The GNU C Library will now detect when /etc/resolv.conf has been
1352 modified and reload the changed configuration. The new resolver option
1353 “no-reload” (RES_NORELOAD) disables this behavior.
1355 - The GNU C Library now supports an arbitrary number of search domains
1356 (configured using the “search” directive in /etc/resolv.conf);
1357 previously, there was a hard limit of six domains. For backward
1358 compatibility, applications that directly modify the ‘_res’ global
1359 object are still limited to six search domains.
1361 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1362 Library will now randomly pick a name server from the configuration as a
1363 starting point. (Previously, the second name server was always used.)
1365 * The tunables feature is now enabled by default. This allows users to tweak
1366 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1368 * New function reallocarray, which resizes an allocated block (like realloc)
1369 to the product of two sizes, with a guaranteed clean failure upon integer
1370 overflow in the multiplication. Originally from OpenBSD, contributed by
1371 Dennis Wölfing and Rüdiger Sonderfeld.
1373 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1374 These are extended versions of preadv and pwritev, respectively, taking an
1375 additional flags argument. The set of supported flags depends on the
1376 running kernel; full support currently requires kernel 4.7 or later.
1378 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1379 create a new session ID for the spawned process. This feature is
1380 scheduled to be added to the next major revision of POSIX; for the time
1381 being, it is available under _GNU_SOURCE.
1383 * errno.h is now safe to use from C-preprocessed assembly language on all
1384 supported operating systems. In this context, it will only define the
1385 Exxxx constants, as preprocessor macros expanding to integer literals.
1387 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1388 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1389 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1390 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1392 To compile programs that use this feature, the compiler must support
1393 128-bit floating point with the type name _Float128 (as defined by TS
1394 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1395 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1396 must be defined to make the new interfaces visible.
1398 The new functions and macros correspond to those present for other
1399 floating-point types (except for a few obsolescent interfaces not
1400 supported for the new type), with F128 or f128 suffixes; for example,
1401 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1402 printf or scanf formats for the new type; the strfromf128 and strtof128
1403 interfaces should be used instead.
1405 Deprecated and removed features, and other changes affecting compatibility:
1407 * The synchronization that pthread_spin_unlock performs has been changed to
1408 now be equivalent to a C11 atomic store with release memory order to the
1409 spin lock's memory location. Previously, several (but not all)
1410 architectures used stronger synchronization (e.g., containing what is
1411 often called a full barrier). This change can improve performance, but
1412 may affect odd fringe uses of spin locks that depend on the previous
1413 behavior (e.g., using spin locks as atomic variables to try to implement
1414 Dekker's mutual exclusion algorithm).
1416 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1419 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1420 will only be built and installed when the GNU C Library is configured with
1421 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1422 as TIRPC or rpcsvc-proto, to be used.
1424 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1425 libnss_compat, are deprecated, and will not be built or installed by
1428 The NIS(+) support library, libnsl, is also deprecated. By default, a
1429 compatibility shared library will be built and installed, but not headers
1430 or development libraries. Only a few NIS-related programs require this
1431 library. (In particular, the GNU C Library has never required programs
1432 that use 'gethostbyname' to be linked with libnsl.)
1434 Replacement implementations based on TIRPC, which additionally support
1435 IPv6, are available from <https://github.com/thkukuk/>. The configure
1436 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1437 name service modules, to be built and installed.
1439 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1440 support is enabled, the configured recursive resolver must support EDNS.
1441 (Responding to EDNS-enabled queries with responses which are not
1442 EDNS-enabled is fine, but FORMERR responses are not.)
1444 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1445 servers have not supported this opcode for a long time.
1447 * The _res_opcodes variable has been removed from libresolv. It had been
1448 exported by accident.
1450 * <string.h> no longer includes inline versions of any string functions,
1451 as this kind of optimization is better done by the compiler. The macros
1452 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1454 * The nonstandard header <xlocale.h> has been removed. Most programs should
1455 use <locale.h> instead. If you have a specific need for the definition of
1456 locale_t with no other declarations, please contact
1457 libc-alpha@sourceware.org and explain.
1459 * The obsolete header <sys/ultrasound.h> has been removed.
1461 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1463 * The obsolete function cfree has been removed. Applications should use
1466 * The stack_t type no longer has the name struct sigaltstack. This changes
1467 the C++ name mangling for interfaces involving this type.
1469 * The ucontext_t type no longer has the name struct ucontext. This changes
1470 the C++ name mangling for interfaces involving this type.
1472 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1473 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1474 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1475 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1476 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1477 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1478 fpregset_t type no longer has the name struct fpu. This changes the C++
1479 name mangling for interfaces involving those types.
1481 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1482 synced with the kernel:
1484 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1485 are not supported on this architecture and have been removed.
1487 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1488 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1489 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1491 Programs that assume the GET/SETREGS ptrace requests are universally
1492 available will now fail to build, instead of malfunctioning at runtime.
1494 Changes to build and runtime requirements:
1496 * Linux kernel 3.2 or later is required at runtime, on all architectures
1497 supported by that kernel. (This is a change from version 2.25 only for
1500 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1502 * On most architectures, GCC 4.9 or later is required to build the GNU C
1503 Library. On powerpc64le, GCC 6.2 or later is required.
1505 Older GCC versions and non-GNU compilers are still supported when
1506 compiling programs that use the GNU C Library. (We do not know exactly
1507 how old, and some GNU extensions to C may be _de facto_ required. If you
1508 are interested in helping us make this statement less vague, please
1509 contact libc-alpha@sourceware.org.)
1511 Security related changes:
1513 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1514 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1516 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1517 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1519 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1520 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1523 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1524 fixed (CVE-2017-12133).
1526 The following bugs are resolved with this release:
1528 [984] network: Respond to changed resolv.conf in gethostbyname
1529 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1530 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1532 [12189] libc: __stack_chk_fail should not attempt a backtrace
1534 [14096] time: Race condition on timezone/tst-timezone.out
1535 [14172] localedata: az_IR: new locale
1536 [14995] build: glibc fails to build if gold is the default linker, even if
1538 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1539 [16637] network: inet_pton function is accepting IPv6 with bad format
1540 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1541 x86 and other generic code
1542 [16875] localedata: ko_KR: fix lang_name
1543 [17225] localedata: ar_SY: localized month names for May and June are
1545 [17297] localedata: da_DK: wrong date_fmt string
1546 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1547 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1548 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1549 [19066] localedata: ar_SA abbreviated day and month names are in English
1550 [19569] network: resolv: Support an arbitrary number of search domains
1551 [19570] network: Implement random DNS server selection in the stub
1553 [19838] locale: localedef fails on PA-RISC
1554 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1555 order of 0D36 and 0D37
1556 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1558 [20098] libc: FAIL: debug/backtrace-tst on hppa
1559 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1561 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1563 [20313] localedata: Update locale data to Unicode 9.0
1564 [20424] manual: Document how to provide a malloc replacement
1565 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1567 [20686] locale: Add el_GR@euro to SUPPORTED.
1568 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1569 failures consistently
1570 [21015] dynamic-link: Document and fix --enable-bind-now
1571 [21016] nptl: pthread_cond support is broken on hppa
1572 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1573 frame-pointer on i386
1574 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1576 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1577 [21088] libc: Build fails with --enable-static-nss
1578 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1579 generic c code is used
1580 [21109] libc: Tunables broken on big-endian
1581 [21112] math: powf has large ulp errors with base close to 1 and exponent
1583 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1585 [21120] malloc: glibc malloc is incompatible with GCC 7
1586 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1588 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1589 (-0.0) when linking with -lieee
1590 [21171] math: log10, log2 and lgamma return incorrect results
1591 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1592 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1593 [21207] localedata: ce_RU: update weekdays from CLDR
1594 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1595 [21217] localedata: Update months from CLDR-31
1596 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1597 [21243] libc: support_delete_temp_file should issue warning for failed
1599 [21244] libc: support resolv_test_start() socket fd close should be
1601 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1602 new posix_spawn implementation
1603 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1604 leads to lower CPU frequency
1605 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1606 [21261] libc: [sparc64] bits/setjmp.h namespace
1607 [21267] network: [mips] bits/socket.h IOC* namespace
1608 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1609 [21270] libc: mmap64 silently truncates large offset values
1610 [21275] libc: posix_spawn always crashes on ia64 now
1611 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1612 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1613 [21289] libc: Incorrect declaration for 32-bit platforms with
1614 _FILE_OFFSET_BITS=64 causes build error
1615 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1616 not support gethostbyname4_r
1617 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1619 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1621 [21340] libc: Support POSIX_SPAWN_SETSID
1622 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1623 [21359] network: ns_name_pack needs additional byte in destination buffer
1624 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1625 against fragmentation attacks (CVE-2017-12132)
1626 [21369] network: resolv: Remove EDNS fallback
1627 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1629 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1630 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1631 [21393] stdio: Missing dup3 error check in freopen, freopen64
1632 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1633 [21399] localedata: Bad description for U00EC in
1634 localedata/charmaps/CP1254
1635 [21411] malloc: realloc documentation error
1636 [21426] network: sys/socket.h uio.h namespace
1637 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1638 [21445] libc: signal.h bsd_signal namespace
1639 [21455] network: Network headers stdint.h namespace
1640 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1642 [21475] network: resolv: Overlong search path is truncated mid-label
1643 [21511] libc: sigstack namespace
1644 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1645 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1646 bits/syscall.h: No such file or directory
1647 [21517] libc: struct sigaltstack namespace
1648 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1649 [21533] localedata: Update locale data to Unicode 10.0
1651 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1652 junk at end of line, first unrecognized character is `@'
1653 [21538] libc: SIG_HOLD missing for XPG4
1654 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1655 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1656 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1657 [21543] libc: sigevent namespace
1658 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1659 size for general purpose registers in mcontext_t structure
1660 [21550] libc: sigwait namespace
1661 [21552] libc: XPG4 bsd_signal namespace
1662 [21554] libc: sigpause namespace
1663 [21560] libc: sys/wait.h signal.h namespace
1664 [21561] libc: waitid namespace
1665 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1666 '/usr/include/stdlib.h': Permission denied
1667 [21575] libc: sys/wait.h missing struct rusage definition
1668 [21584] libc: sigaltstack etc namespace
1669 [21597] libc: siginterrupt namespace
1670 [21607] math: hppa: FAIL: math/test-tgmath
1671 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1673 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1674 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1675 stack and heap (CVE-2017-1000366)
1676 [21625] libc: wait3 namespace
1677 [21654] nss: Incorrect pointer alignment in NSS group merge result
1679 [21657] network: Parse interface zone id for node-local multicast
1680 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1682 [21666] libc: .symver is used on common symbol
1683 [21668] network: resolv: res_init cross-thread broadcast introduces race
1685 [21687] math: tgmath.h totalorder, totalordermag return type
1686 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1688 [21696] libc: Incorrect assumption of of __cpu_mask in
1689 posix/sched_cpucount.c
1690 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1691 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1693 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1694 leading to relocation crash
1695 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1696 `resolv_conf_matches (resp, conf)' failed.
1697 [21710] localedata: Added Samoan language locale for Samoa
1698 [21711] localedata: Pashto yesstr/nostr locale are missing
1699 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1700 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1701 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1702 [21723] localedata: yesstr/nostr missing for Chinese language locale
1703 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1705 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1707 [21728] localedata: New Locale for Tongan language
1708 [21729] localedata: incorrect LC_NAME fields for hi_IN
1709 [21733] localedata: yesstr and nostr are missing for zh_HK
1710 [21734] localedata: Missing yesstr and nostr are for kw_GB
1711 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1712 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1713 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1714 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1716 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1718 [21749] localedata: Wrong abbreviated day name (“abday”) for
1720 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1721 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1722 [21759] localedata: missing yesstr and nostr for Tigrinya
1723 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1724 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1725 [21767] localedata: Missing Bislama locales
1726 [21768] localedata: Missing yesstr and nostr for aa_ET
1727 [21770] localedata: Missing Field in li_NL
1728 [21778] nptl: Robust mutex may deadlock
1729 [21779] libc: MicroBlaze segfaults when loading libpthread
1730 [21783] localedata: Fix int_select international_call_prefixes
1731 [21784] localedata: Inconsistency in country_isbn
1732 [21788] localedata: Missing Country Postal Abbreviations
1733 [21794] localedata: Added-country_isbn-for-Italy
1734 [21795] localedata: Add/Fix country_isbn for France
1735 [21796] localedata: Added country_isbn for Republic of Korea
1736 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1737 [21799] localedata: Added int_select international_call_prefixes
1738 [21801] localedata: Added int_select international_call_prefixes
1739 [21804] nptl: Double semicolon in thread-shared-types.h
1740 [21807] localedata: LC_ADDRESS fix for pap_CW
1741 [21808] localedata: Fix LC_ADDRESS for pap_AW
1742 [21821] localedata: Added country_name in mai_IN
1743 [21822] localedata: Fix LC_TIME for mai_IN
1744 [21823] localedata: missing yesstr, nostr for sa_IN
1745 [21825] localedata: Fix name_mrs for mag_IN
1746 [21828] localedata: 2.26 changelog should mention user visible changes
1748 [21835] localedata: Added Maithili language locale for Nepal
1749 [21838] localedata: Removed redundant data for the_NP
1750 [21839] localedata: Fix LC_MONETARY for ta_LK
1751 [21844] localedata: Fix Latin characters and Months Sequence.
1752 [21848] localedata: Fix mai_NP Title Name
1757 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1758 24731-2:2010, is supported to enable declarations of functions from that
1759 TR. Note that not all functions from that TR are supported by the GNU C
1762 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1763 18661-1:2014, is supported to enable declarations of functions and macros
1764 from that TS. Note that not all features from that TS are supported by
1767 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1768 18661-4:2015, is supported to enable declarations of functions and macros
1769 from that TS. Note that most features from that TS are not supported by
1772 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1773 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1774 Since the GNU C Library defaults to a much newer revision of POSIX, this
1775 will only affect programs that specifically request an old conformance
1776 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1777 see a change in the visible declarations, but a program compiled with
1778 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1781 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1782 defined by all multithreaded code, but glibc has not required this for
1785 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1786 means that in a future release, the macros “major”, “minor”, and “makedev”
1787 will only be available from <sys/sysmacros.h>.
1789 These macros are not part of POSIX nor XSI, and their names frequently
1790 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1791 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1792 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1795 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1796 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1797 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1799 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1800 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1801 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1802 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1803 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1804 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1805 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1806 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1807 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1808 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1809 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1812 * New <math.h> features are added from TS 18661-1:2014:
1814 - Signaling NaN macros: SNANF, SNAN, SNANL.
1816 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1817 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1818 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1820 - llogb functions: the llogb, llogbf and llogbl functions, and the
1821 FP_LLOGB0 and FP_LLOGBNAN macros.
1823 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1826 - Comparison macros: iseqsig.
1828 - Classification macros: iscanonical, issubnormal, iszero.
1830 - Total order functions: totalorder, totalorderf, totalorderl,
1831 totalordermag, totalordermagf, totalordermagl.
1833 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1835 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1836 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1838 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1839 are added to libc. They convert a floating-point number into string.
1841 * Most of glibc can now be built with the stack smashing protector enabled.
1842 It is recommended to build glibc with --enable-stack-protector=strong.
1843 Implemented by Nick Alcock (Oracle).
1845 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1846 intended to be used instead of memset() to erase sensitive data after use;
1847 the compiler will not optimize out calls to explicit_bzero even if they
1848 are "unnecessary" (in the sense that no _correct_ program can observe the
1849 effects of the memory clear).
1851 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1852 to float instead of double. This does not affect the ABI of any libraries
1853 that are part of the GNU C Library, but may affect the ABI of other
1854 libraries that use this type in their interfaces.
1856 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1857 float_t and double_t types are now defined to long double instead of float
1858 and double. These options are not the default, and this does not affect
1859 the ABI of any libraries that are part of the GNU C Library, but it may
1860 affect the ABI of other libraries that use this type in their interfaces,
1861 if they are compiled or used with those options.
1863 * The getentropy and getrandom functions, and the <sys/random.h> header file
1866 * The buffer size for byte-oriented stdio streams is now limited to 8192
1867 bytes by default. Previously, on Linux, the default buffer size on most
1868 file systems was 4096 bytes (and thus remains unchanged), except on
1869 network file systems, where the buffer size was unpredictable and could be
1870 as large as several megabytes.
1872 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1873 for the Linux quota interface which predates kernel version 2.4.22 has
1876 * The malloc_get_state and malloc_set_state functions have been removed.
1877 Already-existing binaries that dynamically link to these functions will
1878 get a hidden implementation in which malloc_get_state is a stub. As far
1879 as we know, these functions are used only by GNU Emacs and this change
1880 will not adversely affect already-built Emacs executables. Any undumped
1881 Emacs executables, which normally exist only during an Emacs build, should
1882 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1884 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1885 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1886 “no-ip6-dotint” had already been the default, and support for the
1887 “ip6-dotint” option was removed from the Internet in 2006.
1889 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1890 flag from <resolv.h> have been removed. The option relied on a
1891 backwards-incompatible DNS extension which was never deployed on the
1894 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1895 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1896 They were already unimplemented.
1898 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1899 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1900 removed again from the IETF name lookup interface specification in RFC
1901 2553. Applications should use getaddrinfo instead.
1903 * DNSSEC-related declarations and definitions have been removed from the
1904 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1905 decode the data part of DNSSEC record types. Previous versions of glibc
1906 only implemented minimal support for the previous version of DNSSEC, which
1907 is incompatible with the currently deployed version.
1909 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1910 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1911 <arpa/nameser.h> header file because the distinction between RR types and
1912 meta-RR types is not officially standardized, subject to revision, and
1913 thus not suitable for encoding in a macro.
1915 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1916 and rhook members of the res_state type in <resolv.h> have been removed.
1917 The glibc stub resolver did not support these hooks, but the header file
1918 did not reflect that.
1920 * For multi-arch support it is recommended to use a GCC which has
1921 been built with support for GNU indirect functions. This ensures
1922 that correct debugging information is generated for functions
1923 selected by IFUNC resolvers. This support can either be enabled by
1924 configuring GCC with '--enable-gnu-indirect-function', or by
1925 enabling it by default by setting 'default_gnu_indirect_function'
1926 variable for a particular architecture in the GCC source file
1929 * GDB pretty printers have been added for mutex and condition variable
1930 structures in POSIX Threads. When installed and loaded in gdb these pretty
1931 printers show various pthread variables in human-readable form when read
1932 using the 'print' or 'display' commands in gdb.
1934 * Tunables feature added to allow tweaking of the runtime for an application
1935 program. This feature can be enabled with the '--enable-tunables' configure
1936 flag. The GNU C Library manual has details on usage and README.tunables has
1937 instructions on adding new tunables to the library.
1939 * A new version of condition variables functions have been implemented in
1940 the NPTL implementation of POSIX Threads to provide stronger ordering
1943 * A new version of pthread_rwlock functions have been implemented to use a more
1944 scalable algorithm primarily through not using a critical section anymore to
1947 Security related changes:
1949 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1950 have been created with makecontext could fail to terminate due to a
1951 missing .cantunwind annotation. This has been observed to lead to a hang
1952 (denial of service) in some Go applications compiled with gccgo. Reported
1953 by Andreas Schwab. (CVE-2016-6323)
1955 * The DNS stub resolver functions would crash due to a NULL pointer
1956 dereference when processing a query with a valid DNS question type which
1957 was used internally in the implementation. The stub resolver now uses a
1958 question type which is outside the range of valid question type values.
1961 The following bugs are resolved with this release:
1963 [4099] stdio: Overly agressive caching by stream i/o functions.
1964 [7065] build: Support building glibc with -fstack-protector or -fstack-
1966 [9842] localedata: en_CA: incorrect date format
1967 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1968 before it started waiting
1969 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1970 [15765] nptl: sem_open is wrongly a cancellation point
1971 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1972 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1974 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1975 library linked with pthread
1976 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1977 "FIXME: Ingo" issue)
1978 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1979 [17252] libc: getrandom and getentropy syscall
1980 [17730] malloc: thread-local storage is sometimes improperly free()'d
1981 after being __libc_memalign()'d
1982 [18241] stdio: failed fseek on memstream does not set errno and can fail
1984 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1986 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1987 PTHREAD_PRIO_INHERIT mutex on ARM
1988 [18784] network: res_query and related function crash for special record
1989 type queries (CVE-2015-5180)
1990 [19380] math: strtod does not raise "inexact"
1991 [19387] string: Integer overflow in memchr
1992 [19390] string: Integer overflow in strncat
1993 [19398] build: linknamespace tests fail with massively parallel build
1994 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1996 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1997 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1998 [19514] libc: [PATCH] Fix spelling errors in spelling
1999 "implement"/"implementation" in several places
2000 [19582] network: Deprecate RES_USE_INET6
2001 [19673] manual: clog10 docs appear to be erroneous
2002 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
2003 causes a segmentation fault
2004 [19826] libc: invalid pointer returned from __tls_get_addr with static
2006 [20016] network: resolv: Remove hooks support from the API
2007 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
2009 [20033] math: [x86_64] vectorized math function don't call the __finite
2011 [20116] nptl: use after free in pthread_create
2012 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
2013 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
2015 [20311] nptl: please install proc_service.h
2016 [20366] libc: Compilation errors in installed headers in strict-compliance
2018 [20370] malloc: malloc: Arena free list management is still racy
2019 (incorrect fix in bug 19243)
2020 [20386] libc: assert (X = 0) does not result in GCC warning
2021 [20432] malloc: malloc: Minimize interface required for interposition
2022 [20435] libc: Missing unwind info in __startcontext causes infinite loop
2023 in _Unwind_Backtrace (CVE-2016-6323)
2024 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
2025 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
2027 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
2028 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
2029 [20477] network: resolv: incorrect double-checked locking related to
2031 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
2033 [20495] math: x86_64 performance degradation due to AVX/SSE transition
2035 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
2036 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
2037 cause transition penalty
2038 [20517] math: sparcv9 missing fdiml compat symbol
2039 [20524] manual: strverscmp is inconsistent
2040 [20525] libc: <sys/quota.h> should be based on kernel headers
2041 [20539] math: GCC 7 -static -lm fails to link at -O3
2042 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
2044 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
2046 [20591] network: Remove obsolete DNSSEC support
2047 [20592] network: DNS resource record type classification macros in
2048 <arpa/nameser.h> are incorrect
2049 [20593] network: Update DNS RR type definitions
2050 [20611] network: getaddrinfo accepts invalid numeric scope IDs
2051 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
2053 [20629] network: libresolv: Remove support for bitstring labels
2055 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
2056 assertion: "maxidx >= 2"
2057 [20660] math: [arm] Use VSQRT
2058 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
2059 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
2060 [20689] libc: Test for FMA should also check for AVX.
2061 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
2062 [20715] math: iszero macro breaks existing code
2063 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
2064 [20728] libc: powerpc: Missing TOC stub in clone
2065 [20729] build: glibc-2.24 fails to build for i486 with -Os
2066 [20750] build: Build fails with default PIE enabled gcc-6.x
2067 [20768] math: [alpha] sqrt fegetenv namespace
2068 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
2070 [20787] math: float_t is defined as float incorrectly on x86_64 even if
2071 __FLT_EVAL_METHOD__ is 2
2072 [20790] network: rpcgen buffer overrun in get_prog_declaration
2073 [20822] nptl: race condition in __lll_unlock_elision on powerpc
2074 [20829] libc: crypt snprintf namespace
2075 [20847] libc: tst-vfork3 failure
2076 [20855] math: Default bits/mathdef.h has inappropriate float_t
2077 [20859] math: [sh4] FP_ILOGB0 invalid
2078 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
2079 U+20AC), not same as GBK
2080 [20915] dynamic-link: global-dynamic TLS broken on aarch64
2081 [20916] math: pow handling of sNaN arguments
2082 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
2083 [20919] math: Bad pow (qNaN, 0) result with -lieee
2084 [20940] math: hypot sNaN handling
2085 [20947] math: fmax, fmin sNaN handling
2086 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
2087 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
2089 [20971] string: powerpc64/power7 memchr overflows internal pointer check
2090 [20973] nptl: robust mutexes: Lost wake-ups
2091 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
2092 [20978] nis: strlen on null pointer in nss_nisplus
2093 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
2095 [21014] string: i686 memchr overflows internal pointer check
2096 [21019] libc: [mips] n32 lseek incorrect on overflow
2097 [21022] libc: [microblaze] __backtrace get_frame_size namespace
2098 [21026] libc: [MIPS] readahead syscall is broken on n64
2099 [21028] math: Fallback fesetexceptflag should always succeed
2100 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
2101 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
2103 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
2104 [21061] librt: [microblaze] librt lost clock_* exports
2105 [21073] libc: tunables: insecure environment variables passed to
2106 subprocesses with AT_SECURE
2107 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
2111 * The minimum Linux kernel version that this version of the GNU C Library
2112 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
2113 kernel version 2.6.32 or later suffices (on architectures that already
2114 required kernel versions more recent than 3.2, those requirements remain
2115 unchanged). Linux 3.2 or later kernel headers are required on all
2118 * The pap_AN locale has been deleted. This has been deprecated for a long
2119 time. It has been replaced by pap_AW & pap_CW, both of which have long
2120 been included in previous releases.
2122 * The readdir_r and readdir64_r functions have been deprecated. It is
2123 recommended to use readdir and readdir64 instead.
2125 * The type “union wait” has been removed. It was deprecated in the early
2126 1990s and never part of POSIX. Application code should use the int type
2127 instead of “union wait”.
2129 * A new NSS action is added to facilitate large distributed system
2130 administration. The action, MERGE, allows remote user stores like LDAP
2131 to be merged into local user stores like /etc/groups in order to provide
2132 easy to use, updated, and managed sets of merged credentials. The new
2133 action can be used by configuring it in /etc/nsswitch.conf:
2134 group: files [SUCCESS=merge] nis
2135 Implemented by Stephen Gallagher (Red Hat).
2137 * The deprecated __malloc_initialize_hook variable has been removed from the
2140 * The long unused localedef --old-style option has been removed. It hasn't
2141 done anything in over 16 years. Scripts using this option can safely
2144 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
2145 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
2146 return the next representable value in the direction of positive infinity
2147 and the nextdown functions return the next representable value in the
2148 direction of negative infinity. These are currently enabled as GNU
2151 Security related changes:
2153 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
2154 could result in a stack overflow when getnetbyname was called with an
2155 overly long name. (CVE-2016-3075)
2157 * Previously, getaddrinfo copied large amounts of address data to the stack,
2158 even after the fix for CVE-2013-4458 has been applied, potentially
2159 resulting in a stack overflow. getaddrinfo now uses a heap allocation
2160 instead. Reported by Michael Petlan. (CVE-2016-3706)
2162 * The glob function suffered from a stack-based buffer overflow when it was
2163 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
2164 Reported by Alexander Cherepanov. (CVE-2016-1234)
2166 * The Sun RPC UDP client could exhaust all available stack space when
2167 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
2168 alloca plugin for GCC. (CVE-2016-4429)
2170 * The IPv6 name server management code in libresolv could result in a memory
2171 leak for each thread which is created, performs a failing naming lookup,
2172 and exits. Over time, this could result in a denial of service due to
2173 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
2175 The following bugs are resolved with this release:
2177 [1170] localedata: ne_NP: update Nepali locale definition file
2178 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
2180 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
2181 [6796] math: fdim() does not set errno on overflow
2182 [10354] libc: posix_spawn should use vfork() in more cases than presently
2183 [11213] localedata: localedata: add copyright disclaimer to locale files
2184 [12143] localedata: chr_US: new Cherokee locale
2185 [12450] localedata: sgs_LT: new locale
2186 [12676] localedata: ln_CD: new locale
2187 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
2189 [13304] math: fma, fmaf, fmal produce wrong results
2190 [14259] build: --localedir arg to configure is ignored
2191 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
2192 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
2193 [14934] localedata: es_CL: wrong first weekday chilean locale
2194 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
2196 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
2198 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
2199 [15368] nptl: raise() is not async-signal-safe
2200 [15479] math: ceil, floor, round and trunc raise inexact exception
2201 [15578] localedata: kk_KZ: various updates
2202 [16003] localedata: pap_AN: punt old locale
2203 [16137] localedata: iw_IL: punt old locale
2204 [16190] localedata: eo: new esperanto locale
2205 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
2206 [16742] malloc: race condition: pthread_atfork() called before first
2207 malloc() results in unexpected locking behaviour/deadlocks
2208 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
2210 [16983] localedata: postal_fmt does not allow %l and %n modifiers
2211 [17565] localedata: pt_PT: wrong (work-)week start
2212 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
2213 [17950] build: Build fails with -msse
2214 [18205] localedata: be_BY*: wrong first_weekday and first_workday
2215 [18433] libc: posix_spawn does not return correctly upon failure to
2217 [18453] localedata: charmaps/IBM875: incorrect codes
2218 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2220 [18896] localedata: he_IL: improvements for currency
2221 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2222 Romanian locale data
2223 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2225 [19056] libc: Deprecate readdir_r
2226 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2228 [19198] localedata: nl_NL: small improvements for Dutch locales
2229 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2230 nameservers (CVE-2016-5417)
2231 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2233 [19400] locale: Language missing in "iso-639.def", trivial fix in
2235 [19431] malloc: Deadlock between fflush, getdelim, and fork
2236 [19505] libc: Incorrect file descriptor validity checks in
2237 posix_spawn_file_actions_add{open,close,dup2}
2238 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2239 when using RTLD_NEXT
2240 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2241 `intl/{gettextP,loadinfo}.h'
2242 [19534] libc: execle, execlp may use malloc
2243 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2244 [19573] network: res_nclose and __res_maybe_init disagree about name
2245 server initialization, breaking Hesiod
2246 [19575] localedata: Status of GB18030 tables
2247 [19581] localedata: sr_* date_fmt string contains additional newline
2248 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2250 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2251 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2252 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2253 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2255 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2257 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2259 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2260 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2261 [19633] locale: strfmon_l applies global locale to number formatting
2262 [19642] network: Memory leak in getnameinfo
2263 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2264 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2266 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2267 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2269 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2270 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2272 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2274 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2276 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2277 pointers and lengths in error-case.
2278 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2279 report errors on UTF-16 surrogates values.
2280 [19755] nscd: nscd assertion failure in gc
2281 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2282 [19759] libc: mempcpy shouldn't be inlined
2283 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2284 [19765] libc: s390 needs an optimized mempcpy
2285 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2286 NAME_MAX limit assumption (CVE-2016-1234)
2287 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2288 [19787] network: Missing and incorrect truncation checks in getnameinfo
2289 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2291 [19791] network: Assertion failure in res_query.c with un-connectable name
2293 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2294 [19822] math: libm.so install clobbers old version
2295 [19825] network: resolv: send_vc can return uninitialized data in second
2296 response to getaddrinfo
2297 [19830] network: nss_dns: should check RDATA length against buffer length
2298 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2299 confronted with A/AAAA records of invalid size
2300 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2301 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2302 [19853] stdio: Printing IBM long double in decimal with high precision is
2304 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2305 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2306 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2308 [19865] network: Assertion failure or memory leak in
2309 _nss_dns_getcanonname_r
2310 [19868] network: nss_dns: netent code does not skip over non-PTR records
2311 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2313 [19881] string: Improve x86-64 memset
2314 [19907] string: Incorrect memcpy tests
2315 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2317 [19925] libc: termios.h XCASE namespace
2318 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2319 [19929] libc: limits.h NL_NMAX namespace
2320 [19931] stdio: Memory leak in vfprintf
2321 [19957] libc: clone(CLONE_VM) access invalid parent memory
2322 [19963] localedata: en_IL: New locale
2323 [19989] stdio: stdio.h cuserid namespace
2324 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2326 [19996] locale: langinfo.h nl_langinfo_l namespace
2327 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2328 [20010] network: getaddrinfo: Stack overflow in hostent translation
2330 [20012] stdio: libio: fmemopen append mode failure
2331 [20014] stdio: stdio.h namespace for pre-threads POSIX
2332 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2333 [20023] libc: fcntl.h timespec namespace
2334 [20024] math: [x86_64] vectorized sincos trashes the stack
2335 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2336 [20041] time: sys/time.h timespec namespace
2337 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2338 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2339 [20051] libc: ttyslot in wrong header under wrong conditions
2340 [20054] libc: gethostname not declared for XPG4
2341 [20055] libc: termios.h missing tcgetsid for XPG4
2342 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2344 [20073] libc: sys/stat.h fchmod namespace
2345 [20074] libc: stdlib.h rand_r namespace
2346 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2347 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2349 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2350 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2352 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2353 [20119] libc: Wrong mask for processors level type from CPUID
2354 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2355 AS not supporting AVX512
2356 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2358 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2359 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2361 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2362 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2363 [20178] libc: posix_spawn{p} should not call exit
2364 [20191] stdio: libio: vtables hardening
2365 [20195] string: FMA4 detection requires CPUID execution with register
2367 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2368 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2370 [20212] math: acos (sNaN) returns sNaN
2371 [20213] math: asin (sNaN) returns sNaN
2372 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2373 [20218] math: [i386] asinhl (sNaN) returns sNaN
2374 [20219] math: [i386] atanhl (sNaN) returns sNaN
2375 [20222] stdio: fopencookie: Mangle function pointers
2376 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2377 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2378 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2380 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2381 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2382 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2383 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2384 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2385 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2386 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2387 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2388 [20240] math: modf (sNaN) returns sNaN
2389 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2390 [20250] math: frexp (sNaN) returns sNaN
2391 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2392 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2394 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2395 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2396 subscript is below array bounds [-Werror=array-bounds]
2397 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2398 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2400 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2401 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2403 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2404 "invalid" exceptions
2405 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2406 [20316] localedata: id_ID: Februari instead of Pebruari
2407 [20327] string: POWER8 strcasecmp returns incorrect result
2408 [20347] math: Failure: Test: j0_downward (0xap+0)
2409 [20348] libc: FAIL: misc/tst-preadvwritev64
2410 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2411 [20350] libc: There is no test for p{read,write}64
2412 [20357] math: Incorrect cos result for 1.5174239687223976
2413 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2417 * Unicode 8.0.0 Support: Character encoding, character type info, and
2418 transliteration tables are all updated to Unicode 8.0.0, using new
2419 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2420 These updates cause user visible changes, such as the fixes for bugs
2421 89, 16061, and 18568.
2423 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2424 kernel-internal CPU set size. This means that requests that change the
2425 CPU affinity which failed before (for example, an all-ones CPU mask) will
2426 now succeed. Applications that need to determine the effective CPU
2427 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2428 setting it because the kernel can adjust it (and the previous size check
2429 would not detect this in the majority of cases).
2431 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2432 the following new symbols are used: fts64_children, fts64_close,
2433 fts64_open, fts64_read and fts64_set.
2435 * getaddrinfo now detects certain invalid responses on an internal netlink
2436 socket. If such responses are received, an affected process will
2437 terminate with an error message of "Unexpected error <number> on netlink
2438 descriptor <number>" or "Unexpected netlink response of size <number> on
2439 descriptor <number>". The most likely cause for these errors is a
2440 multi-threaded application which erroneously closes and reuses the netlink
2441 file descriptor while it is used by getaddrinfo.
2443 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2444 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2445 unnecessary serialization of memory allocation requests across threads.
2446 The defect is now corrected. Users should see a substantial increase in
2447 the concurent throughput of allocation requests for applications which
2448 trigger this bug. Affected applications typically create and
2449 destroy threads frequently. (Bug 19048 was reported and analyzed by
2452 * There is now a --disable-timezone-tools configure option for disabling the
2453 building and installing of the timezone related utilities (zic, zdump, and
2454 tzselect). This is useful for people who build the timezone data and code
2455 independent of the GNU C Library.
2457 * The obsolete header <regexp.h> has been removed. Programs that require
2458 this header must be updated to use <regex.h> instead.
2460 * The obsolete functions bdflush, create_module, get_kernel_syms,
2461 query_module and uselib are no longer available to newly linked binaries;
2462 the header <sys/kdaemon.h> has been removed. These functions and header
2463 were specific to systems using the Linux kernel and could not usefully be
2464 used with the GNU C Library on systems with version 2.6 or later of the
2467 * Optimized string, wcsmbs and memory functions for IBM z13.
2468 Implemented by Stefan Liebler.
2470 * Newly linked programs that define a variable called signgam will no longer
2471 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2472 require signgam to be set by those functions must ensure that they use the
2473 variable provided by the GNU C Library and declared in <math.h>, without
2474 defining their own copy.
2476 * The minimum GCC version that can be used to build this version of the GNU
2477 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2478 still be used to compile programs using the GNU C Library.
2480 Security related changes:
2482 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2483 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2485 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2486 the pointer guard feature. It is always enabled. Previously,
2487 LD_POINTER_GUARD could be used to disable security hardening in binaries
2488 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2491 * An integer overflow in hcreate and hcreate_r could lead to an
2492 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2494 * The catopen function no longer has unbounded stack usage. Reported by
2495 Max. (CVE-2015-8779)
2497 * The nan, nanf and nanl functions no longer have unbounded stack usage
2498 depending on the length of the string passed as an argument to the
2499 functions. Reported by Joseph Myers. (CVE-2014-9761)
2501 * A stack-based buffer overflow was found in libresolv when invoked from
2502 libnss_dns, allowing specially crafted DNS responses to seize control
2503 of execution flow in the DNS client. The buffer overflow occurs in
2504 the functions send_dg (send datagram) and send_vc (send TCP) for the
2505 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2506 family. The use of AF_UNSPEC triggers the low-level resolver code to
2507 send out two parallel queries for A and AAAA. A mismanagement of the
2508 buffers used for those queries could result in the response of a query
2509 writing beyond the alloca allocated buffer created by
2510 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2511 the overflow. Thanks to the Google Security Team and Red Hat for
2512 reporting the security impact of this issue, and Robert Holiday of
2513 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2515 The following bugs are resolved with this release:
2517 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2518 [887] math: Math library function "logb" and "nextafter" inconsistent
2519 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2520 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2521 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2522 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2524 [4404] localedata: German translation of "Alarm clock" is misleading
2525 [6799] math: nextafter() and nexttoward() doen't set errno on
2526 overflow/underflow errors
2527 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2529 [10432] nis: _nss_nis_setnetgrent assertion failure
2530 [11460] libc: fts has no LFS support
2531 [12926] network: getaddrinfo()/make_request() may spin forever
2532 [13065] nptl: Race condition in pthread barriers
2533 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2534 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2536 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2537 [14912] libc: Rename non-installed bits/*.h headers
2538 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2539 [15367] math: Let gcc use __builtin_isinf
2540 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2541 [15421] math: lgamma wrongly sets signgam for ISO C
2542 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2543 FE_INVALID with argument out of range
2544 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2546 [15786] dynamic-link: ifunc resolver functions can smash function
2548 [15918] math: Unnecessary check for equality in hypotf()
2549 [16061] localedata: Review / update transliteration data
2550 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2552 [16141] time: strptime %z offset restriction
2553 [16171] math: drem should be alias of remainder
2554 [16296] math: fegetround is pure?
2555 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2556 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2558 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2559 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2561 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2562 "Exception "Inexact" set" on ppc32
2563 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2564 [16517] math: Missing underflow exception from tanf/tan/tanl
2565 [16519] math: Missing underflow exception from sinhf
2566 [16520] math: Missing underflow exception from tanhf
2567 [16521] math: Missing underflow exception from exp2
2568 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2570 [16734] stdio: fopen calls mmap to allocate its buffer
2571 [16961] math: nan function incorrect handling of bad sequences
2572 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2573 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2574 [16985] locale: localedef: confusing error message when opening output
2576 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2577 [17197] locale: Redundant shift character in iconv conversion output at
2579 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2580 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2581 [17250] dynamic-link: static linking breaks nss loading
2582 (getaddrinfo/getpwnam/etc...)
2583 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2585 [17441] math: isnan() should use __builtin_isnan() in GCC
2586 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2587 (related to lock elision)
2588 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2589 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2590 [17887] time: strptime should be able to parse "+01:00" style timezones
2591 [17905] libc: catopen() Multiple unbounded stack allocations
2593 [18084] libc: backtrace (..., 0) dumps core on x86
2594 [18086] libc: nice() sets errno to 0 on success
2595 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2596 is too large (CVE-2015-8778)
2597 [18251] dynamic-link: SONAME missing when audit modules provides path
2598 [18265] libc: add attributes for wchar string and memory functions
2599 [18370] math: csqrt missing underflows
2600 [18421] libc: [hppa] read-only segment has dynamic relocations
2601 [18472] libc: Obsolete syscall wrappers should be compat symbols
2602 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2603 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2604 [18525] localedata: Remove locale timezone information
2605 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2606 [18568] localedata: Update locale data to Unicode 8.0
2607 [18589] locale: sort-test.sh fails at random
2608 [18595] math: ctan, ctanh missing underflows
2609 [18604] libc: assert macro-expands its argument
2610 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2611 contains a vector instruction exception.
2612 [18611] math: j1, jn missing errno setting on underflow
2613 [18618] localedata: sync Chechen locale definitions with other *_RU
2615 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2616 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2617 [18665] network: In send_dg, the recvfrom function is NOT always using the
2618 buffer size of a newly created buffer (CVE-2015-7547)
2619 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2621 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2623 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2624 [18699] math: tilegx cproj() for various complex infinities does not yield
2626 [18724] libc: Harden put*ent functions against data injection
2627 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2628 [18755] build: build errors with -DNDEBUG
2629 [18757] stdio: fmemopen fails to set errno on failure
2630 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2632 [18781] libc: openat64 lacks O_LARGEFILE
2633 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2634 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2635 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2636 [18790] math: [ldbl-128ibm] tanhl inaccurate
2637 [18795] libc: stpncpy fortification misses buffer lengths that are
2638 statically too large
2639 [18796] build: build fails for --disable-mathvec
2640 [18803] math: hypot missing underflows
2641 [18820] stdio: fmemopen may leak memory on failure
2642 [18823] math: csqrt spurious underflows
2643 [18824] math: fma spurious underflows
2644 [18825] math: pow missing underflows
2645 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2646 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2647 [18870] build: sem_open.c fails to compile with missing symbol
2649 [18872] stdio: Fix memory leak in printf_positional
2650 [18873] libc: posix_fallocate overflow check ineffective
2651 [18875] math: Excess precision leads incorrect libm
2652 [18877] libc: arm: mmap offset regression
2653 [18887] libc: memory corruption when using getmntent on blank lines
2654 [18918] localedata: hu_HU: change time to HH:MM:SS format
2655 [18921] libc: Regression: extraneous stat() and fstat() performed by
2657 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2658 binaries (CVE-2015-8777)
2659 [18951] math: tgamma missing underflows
2660 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2662 [18953] localedata: lt_LT: change currency symbol to the euro
2663 [18956] math: powf inaccuracy
2664 [18961] math: [i386] exp missing underflows
2665 [18966] math: [i386] exp10 missing underflows
2666 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2667 [18969] build: multiple string test failures due to missing locale
2669 [18970] libc: Reference of pthread_setcancelstate in libc.a
2670 [18977] math: float / long double Bessel functions not in XSI POSIX
2671 [18980] math: i386 libm functions return with excess range and precision
2672 [18981] math: i386 scalb*, ldexp return with excess range and precision
2673 [18982] stdio: va_list and vprintf
2674 [18985] time: Passing out of range data to strftime() causes a segfault
2676 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2677 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2678 [19012] locale: iconv_open leaks memory on error path
2679 [19016] math: clog, clog10 inaccuracy
2680 [19018] nptl: Mangle function pointers in tls_dtor_list
2681 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2682 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2683 [19048] malloc: malloc: arena free list can become cyclic, increasing
2685 [19049] math: [powerpc] erfc incorrect zero sign
2686 [19050] math: [powerpc] log* incorrect zero sign
2687 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2688 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2689 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2690 [19074] network: Data race in _res_hconf_reorder_addrs
2691 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2692 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2693 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2695 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2697 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2698 [19086] manual: posix_fallocate64 documented argument order is wrong.
2699 [19088] math: lround, llround missing exceptions close to overflow
2701 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2702 [19095] math: dbl-64 lrint incorrect for 64-bit long
2703 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2704 [19124] dynamic-link: ld.so failed to build with older assmebler
2705 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2706 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2707 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2708 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2709 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2710 pthread_setaffinity_np
2711 [19156] math: [ldbl-128] j0l spurious underflows
2712 [19164] nptl: tst-getcpu fails with many possible CPUs
2713 [19168] math: math/test-ildoubl and math/test-ldouble failure
2714 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2715 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2717 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2718 (FE_NOMASK_ENV) do not clear SSE exceptions
2719 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2720 _int_new_arena/reused_arena
2721 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2722 [19201] math: dbl-64 remainder incorrect sign of zero result
2723 [19205] math: bits/math-finite.h conditions do not match math.h and
2725 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2726 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2727 for C99-based standards
2728 [19212] libc: features.h not -Wundef clean
2729 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2731 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2732 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2733 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2735 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2736 "inexact" exceptions
2737 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2739 [19242] libc: strtol incorrect in Turkish locales
2740 [19243] malloc: reused_arena can pick an arena on the free list, leading
2741 to an assertion failure and reference count corruption
2742 [19253] time: tzset() ineffective when temporary TZ did not include DST
2744 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2745 [19270] math: [hppa] Shared libm missing __isnanl
2746 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2747 MAP_HUGETLB and MAP_STACK defines
2748 [19313] nptl: Wrong __cpu_mask for x32
2749 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2750 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2751 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2752 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2753 [19363] time: x32: times() return value wrongly truncates/sign extends
2755 [19367] dynamic-link: Improve branch prediction on Silvermont
2756 [19369] network: Default domain name not reset by res_ninit when "search"
2757 / "domain" entry is removed from resolv.conf
2758 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2760 [19385] localedata: bg_BG: time separator should be colon, not comma
2761 [19408] libc: linux personality syscall wrapper may erroneously return an
2762 error on 32-bit architectures
2763 [19415] libc: dladdr returns wrong names on hppa
2764 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2765 IBM905, IBM907, and IBM909
2766 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2767 [19443] build: build failures with -DDEBUG
2768 [19451] build: Make check fails on test-double-vlen2
2769 [19462] libc: Glibc failed to build with -Os
2770 [19465] math: Wrong code with -Os
2771 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2773 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2775 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2776 [19486] math: S390: Math tests fail with "Exception Inexact set".
2777 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2778 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2780 [19590] math: Fail to build shared objects that use libmvec.so functions.
2784 * The following bugs are resolved with this release:
2786 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2787 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2788 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2789 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2790 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2791 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2792 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2793 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2794 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2795 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2796 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2797 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2798 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2799 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2800 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2801 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2802 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2803 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2804 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2805 18657, 18676, 18694, 18696, 18887.
2807 * Cache information can be queried via sysconf() function on s390 e.g. with
2808 _SC_LEVEL1_ICACHE_SIZE as argument.
2810 * A buffer overflow in gethostbyname_r and related functions performing DNS
2811 requests has been fixed. If the NSS functions were called with a
2812 misaligned buffer, the buffer length change due to pointer alignment was
2813 not taken into account. This could result in application crashes or,
2814 potentially arbitrary code execution, using crafted, but syntactically
2815 valid DNS responses. (CVE-2015-1781)
2817 * The time zone file parser has been made more robust against crafted time
2818 zone files, avoiding heap buffer overflows related to the processing of
2819 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2820 large time zone data files. Overly long time zone specifiers in the TZ
2821 variable no longer result in stack overflows and crashes.
2823 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2824 for LD and GD on x86 and x86-64, has been implemented. You will need
2825 binutils-2.24 or later to enable this optimization.
2827 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2828 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2829 Hat). These updates cause user visible changes, such as the fix for bug
2832 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2833 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2834 condition in some applications.
2836 * Added vector math library named libmvec with the following vectorized x86_64
2837 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2839 The library can be disabled with --disable-mathvec. Use of the functions is
2840 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2841 Shared library libmvec.so is linked in as needed when using -lm (no need to
2842 specify -lmvec explicitly for not static builds).
2843 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2845 * A new fmemopen implementation has been added with the goal of POSIX
2846 compliance. The new implementation fixes the following long-standing
2847 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2848 old implementation is still present for use be by existing binaries.
2850 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2851 releases. It has been fixed to match 2.19 and older, but binaries built
2852 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2854 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2855 Contributed by Roland McGrath (Google).
2857 * The header <regexp.h> is deprecated, and will be removed in a future
2858 release. Use of this header will trigger a deprecation warning.
2859 Application developers should update their code to use <regex.h> instead.
2861 This header was formerly part of SUS, but was deprecated in 1994 and
2862 removed from the standard in 2001. Also, the glibc implementation
2863 leaks memory. See BZ#18681 for more details.
2867 * The following bugs are resolved with this release:
2869 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2870 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2871 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2872 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2873 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2874 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2875 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2876 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2877 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2878 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2881 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2882 too little memory for the to-be-scanned arguments and overflow the
2883 allocated buffer. The implementation now correctly computes the required
2884 buffer size when using malloc, and switches to malloc from alloca as
2887 * A new semaphore algorithm has been implemented in generic C code for all
2888 machines. Previous custom assembly implementations of semaphore were
2889 difficult to reason about or ensure that they were safe. The new version
2890 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2891 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2892 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2894 * Port to Altera Nios II has been contributed by Mentor Graphics.
2896 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2897 implementations for powerpc64/powerpc64le.
2898 Implemented by Adhemerval Zanella (IBM).
2900 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2901 and powerpc64le. This may improve lock scaling of existing programs on
2902 HTM capable systems. The lock elision code is only enabled with
2903 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2904 powerpc will issue a transaction abort on every syscall to avoid side
2905 effects being visible outside transactions.
2907 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2908 AArch64. Contributed by ARM Ltd.
2910 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2912 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2913 under certain input conditions resulting in the execution of a shell for
2914 command substitution when the application did not request it. The
2915 implementation now checks WRDE_NOCMD immediately before executing the
2916 shell and returns the error WRDE_CMDSUB as expected.
2918 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2919 processing format strings with a large number of format specifiers.
2921 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2922 infinite loop if the DNS response contained a PTR record of an unexpected
2925 * The minimum GCC version that can be used to build this version of the GNU
2926 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2927 still be used to compile programs using the GNU C Library.
2929 * The GNU C Library is now built with -Werror by default. This can be
2930 disabled by configuring with --disable-werror.
2932 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2934 * The obsolete sigvec function has been removed. This was the original
2935 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2936 programs have been using instead for about 25 years. Of course, ABI
2937 compatibility for old binaries using sigvec remains intact.
2939 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2940 with newer versions of bison.
2942 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2943 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2944 registers overlay two consecutive single-precision registers. MIPS32R2
2945 introduced a new FPU mode (FR=1) where double-precision registers extend the
2946 corresponding single-precision registers which is incompatible with the
2947 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2948 require the use of FR=1 making a transition necessary. New o32 ABI
2949 extensions enable users to migrate over time from the original o32 ABI
2950 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2951 tracks the ABI of any loaded object and verifies that new objects are
2952 compatible. Mode transitions will also be requested as required and
2953 unsupportable objects will be rejected. The ABI checks include both soft and
2954 hard float ABIs for o32, n32 and n64.
2956 GCC 5 with GNU binutils 2.25 onwards:
2957 It is strongly recommended that all o32 system libraries are built using the
2958 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2959 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2960 compiler using --with-fp-32=xx to set this by default.
2964 * The following bugs are resolved with this release:
2966 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2967 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2968 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2969 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2970 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2971 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2972 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2973 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2974 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2975 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2976 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2977 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2978 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2979 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2980 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2982 * Reverted change of ABI data structures for s390 and s390x:
2983 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2984 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2985 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2986 are preserved pointing straight to the same implementation as the old ones.
2987 Given that, new callers will simply provide a too-big buffer to these
2988 functions. Any applications/libraries out there that embed jmp_buf or
2989 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2990 against 2.19 headers will have to rebuilt again. This is necessary in any
2991 case to revert the breakage in their ABI caused by the glibc change.
2993 * Support for file description locks is added to systems running the
2994 Linux kernel. The standard file locking interfaces are extended to
2995 operate on file descriptions, not file descriptors, via the use of
2996 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2997 are associated with an open file instead of a process.
2999 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
3001 * The minimum Linux kernel version that this version of the GNU C Library
3002 can be used with is 2.6.32.
3004 * Running the testsuite no longer terminates as soon as a test fails.
3005 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
3006 with PASS or FAIL lines for individual tests. A summary of the results is
3007 printed, including a list of failing lists, and "make check" exits with
3008 error status if there were any unexpected failures. "make check
3009 stop-on-test-failure=y" may be used to keep the old behavior.
3011 * The am33 port, which had not worked for several years, has been removed
3014 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
3015 supported; they now act the same as _DEFAULT_SOURCE (but generate a
3016 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
3017 conflicted with POSIX (support for which was removed in 2.19), the
3018 interfaces those macros enabled remain available when compiling with
3019 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
3020 test macros defined.
3022 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
3024 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
3025 This may improve lock scaling of existing programs on TX capable systems.
3026 The lock elision code is only built with --enable-lock-elision=yes and
3027 then requires a GCC version supporting the TX builtins. With lock elision
3028 default mutexes are elided via __builtin_tbegin, if the cpu supports
3029 transactions. By default lock elision is not enabled and the elision code
3032 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
3033 copy the path argument. This allowed programs to cause posix_spawn to
3034 deference a dangling pointer, or use an unexpected pathname argument if
3035 the string was modified after the posix_spawn_file_actions_addopen
3038 * All supported architectures now use the main glibc sysdeps directory
3039 instead of some being in a separate "ports" directory (which was
3040 distributed separately before glibc 2.17).
3042 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
3043 On configurations that support it (all Linux configurations), it's now
3044 used regardless of the --enable-add-ons switch to configure. It is no
3045 longer possible to build such configurations without pthreads support.
3047 * Locale names, including those obtained from environment variables (LANG
3048 and the LC_* variables), are more tightly checked for proper syntax.
3049 setlocale will now fail (with EINVAL) for locale names that are overly
3050 long, contain slashes without starting with a slash, or contain ".." path
3051 components. (CVE-2014-0475) Previously, some valid locale names were
3052 silently replaced with the "C" locale when running in AT_SECURE mode
3053 (e.g., in a SUID program). This is no longer necessary because of the
3056 * On x86-64, the dynamic linker's lazy-binding support is now compatible
3057 with application code using Intel MPX instructions. (With all previous
3058 versions, the MPX register state could be clobbered when making calls
3059 into or out of a shared library.) Note that while the new dynamic
3060 linker is compatible with all known x86 hardware whether or not it
3061 supports Intel MPX, some x86 instruction-set emulators might fail to
3062 handle the new instruction encodings. This is known to affect Valgrind
3063 versions up through 3.9 (but will be fixed in the forthcoming 3.10
3064 release), and might affect other tools that do instruction emulation.
3066 * Support for loadable gconv transliteration modules has been removed.
3067 The support for transliteration modules has been non-functional for
3068 over a decade, and the removal is prompted by security defects. The
3069 normal gconv conversion modules are still supported. Transliteration
3070 with //TRANSLIT is still possible, and the //IGNORE specifier
3071 continues to be supported. (CVE-2014-5119)
3073 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
3074 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
3075 resulting a denial-of-service security vulnerability in applications which
3076 use functions related to iconv. (CVE-2014-6040)
3080 * The following bugs are resolved with this release:
3082 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
3083 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
3084 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
3085 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
3086 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
3087 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
3088 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
3089 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
3090 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
3091 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
3092 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
3093 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
3094 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
3095 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
3096 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
3097 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
3098 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
3099 16431, 16453, 16474, 16506, 16510, 16529
3101 * Slovenian translations for glibc messages have been contributed by the
3102 Translation Project's Slovenian team of translators.
3104 * The public headers no longer use __unused nor __block. This change is to
3105 support compiling programs that are derived from BSD sources and use
3106 __unused internally, and to support compiling with Clang's -fblock
3107 extension which uses __block.
3109 * CVE-2012-4412 The strcoll implementation caches indices and rules for
3110 large collation sequences to optimize multiple passes. This cache
3111 computation may overflow for large collation sequences and may cause a
3112 stack or buffer overflow. This is now fixed to use a slower algorithm
3113 which does not use a cache if there is an integer overflow.
3115 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
3116 rules for large collation sequences to optimize multiple passes and falls
3117 back to alloca if malloc fails, resulting in a possible stack overflow.
3118 The implementation now falls back to an uncached collation sequence lookup
3121 * CVE-2013-4788 The pointer guard used for pointer mangling was not
3122 initialized for static applications resulting in the security feature
3123 being disabled. The pointer guard is now correctly initialized to a
3124 random value for static applications. Existing static applications need
3125 to be recompiled to take advantage of the fix (bug 15754).
3127 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
3128 to the d_name member of struct dirent, or omit the terminating NUL
3129 character. (Bugzilla #14699).
3131 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
3132 aligned_alloc functions could allocate too few bytes or corrupt the
3133 heap when passed very large allocation size values (Bugzilla #15855,
3136 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
3137 for AF_INET6 has been fixed (Bugzilla #16072).
3139 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
3140 pap_CW, quz_PE, the_NP.
3142 * Substantially revised locales: gd_GB, ht_HT
3144 * The LC_ADDRESS field was updated to support country_car for almost all
3147 * ISO 1427 definitions were updated.
3149 * ISO 3166 definitions were updated.
3151 * The localedef utility now supports --big-endian and --little-endian
3152 command-line options to generate locales for a different system from that
3153 for which the C library was built.
3155 * Binary locale files now only depend on the endianness of the system for
3156 which they are generated and not on other properties of that system. As a
3157 consequence, binary files generated with new localedef may be incompatible
3158 with old versions of the GNU C Library, and binary files generated with
3159 old localedef may be incompatible with this version of the GNU C Library,
3160 in the following circumstances:
3162 + Locale files may be incompatible on m68k systems.
3164 + Locale archive files (but not separate files for individual locales) may
3165 be incompatible on systems where plain "char" is signed.
3167 * The configure option --disable-versioning has been removed. Builds with
3168 --disable-versioning had not worked for several years.
3170 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
3171 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
3173 * SystemTap probes for malloc have been introduced.
3175 * SystemTap probes for slow multiple precision fallback paths of
3176 transcendental functions have been introduced.
3178 * Support for powerpc64le has been added.
3180 * The soft-float powerpc port now supports e500 processors.
3182 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
3184 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
3185 set of header declarations that are enabled by default, even when other
3186 feature test macros or compiler options such as -std=c99 would otherwise
3187 disable some of those declarations.
3189 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
3190 conflict with POSIX. The libbsd-compat library (which was a dummy library
3191 that did nothing) has also been removed.
3193 * Preliminary documentation about Multi-Thread, Async-Signal and
3194 Async-Cancel Safety has been added.
3196 * Change of ABI data structures for s390 and s390x:
3197 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
3198 allow for future hardware extensions. All library functions that accept or
3199 return these structures were versioned in order to provide backwards
3200 compatibility with existing code. However, not all cases can be handled
3201 with interface versioning. If an application embeds either structure into
3202 another structure and that structure is passed to another compilation unit
3203 which expects the newer larger-sized object then any access to the new
3204 fields will result in undefined behaviour. Similarly any access to fields
3205 that were moved by the enlarging of the structures will cause undefined
3206 behaviour. To fix the undefined behaviour all objects that are part of the
3207 final in-memory image must agree on the size of structures, and this may
3208 require recompilation.
3212 * The following bugs are resolved with this release:
3214 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
3215 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
3216 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
3217 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
3218 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3219 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3220 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3221 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3222 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3223 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3224 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3225 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3226 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3229 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3230 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3231 Distributions can re-enable building and using pt_chown via the new configure
3232 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3233 considerable security risks and should only be used if the distribution
3234 understands and accepts the risks.
3236 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3239 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3240 fixed (Bugzilla #15330).
3242 * Add support for calling C++11 thread_local object destructors on thread
3243 and program exit. This needs compiler support for offloading C++11
3244 destructor calls to glibc.
3246 * Improved worst case performance of libm functions with double inputs and
3249 * Support for priority inherited mutexes in pthread condition variables on
3250 non-x86 architectures.
3252 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3254 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3256 * Optimized string functions for ARM. Implemented by Will Newton and
3259 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3261 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3264 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3265 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3267 * Added a benchmark framework to track performance of functions in glibc.
3269 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3270 It is based on draft TS 18661 and currently enabled as a GNU extension.
3272 * On Linux, the clock function now uses the clock_gettime system call
3273 for improved precision, rather than old times system call.
3275 * Added support for version-3 tz format files. This is needed when using
3276 the tz database release 2013e or later, and affects a few unusual cases --
3277 currently only TZ='America/Godthab' for time stamps after 2037.
3279 * Added new API functions pthread_getattr_default_np and
3280 pthread_setattr_default_np to get and set the default pthread
3281 attributes of a process.
3283 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3284 This may improve lock scaling of existing programs on TSX capable systems.
3285 When the --enable-lock-elision=yes parameter is specified at configure
3286 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3289 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3290 availability). Implemented by Ryan S. Arnold.
3292 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3296 * The following bugs are resolved with this release:
3298 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3299 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3300 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3301 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3302 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3303 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3304 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3305 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3306 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3307 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3308 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3309 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3310 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3311 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3312 14879, 14889, 14893, 14898, 14914.
3314 * Optimization of memcpy for MIPS.
3316 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3317 EMFILE has been fixed (Bugzilla #14889).
3319 * The add-on ports collection is now distributed in the "ports" subdirectory
3320 of the main GNU C Library distribution, rather than separately.
3322 * Port to ARM AArch64 contributed by Linaro.
3324 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3325 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3327 Implemented by Andreas Krebbel.
3329 * The new function secure_getenv allows secure access to the environment,
3330 returning NULL if running in a SUID/SGID process. This function replaces
3331 the internal function __secure_getenv.
3333 * SystemTap static probes have been added into the dynamic linker.
3334 Implemented by Gary Benson.
3336 * Optimizations of string functions strstr, strcasestr and memmem.
3337 Implemented by Maxim Kuvyrkov.
3339 * The minimum Linux kernel version that this version of the GNU C Library
3340 can be used with is 2.6.16.
3342 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3343 powerpc POWER7. Implemented by Will Schmidt.
3345 * New configure option --disable-nscd builds the C library such that it
3346 never attempts to contact the Name Service Caching Daemon (nscd).
3347 New configure option --disable-build-nscd avoids building nscd itself;
3348 this is the default if --disable-nscd is used.
3350 * Improved support for cross-compilation, including cross-testing and
3351 bootstrap builds without a previously built glibc.
3353 * Several testsuite tests are now able to test multiple IFUNC variants of an
3354 interface, rather than just testing the one that would be chooen by
3357 * New configure options --with-bugurl and --with-pkgversion, for
3358 distributors to use to embed their bug-reporting and package version
3359 information in --help and --version output.
3361 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3362 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3363 allows creation of chroots without the procfs mounted on /proc.
3365 * The `crypt' function now fails if passed salt bytes that violate the
3366 specification for those values. On Linux, the `crypt' function will
3367 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3368 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3369 when the mode is enabled.
3371 * The `clock_*' suite of functions (declared in <time.h>) is now available
3372 directly in the main C library. Previously it was necessary to link with
3373 -lrt to use these functions. This change has the effect that a
3374 single-threaded program that uses a function such as `clock_gettime' (and
3375 is not linked with -lrt) will no longer implicitly load the pthreads
3376 library at runtime and so will not suffer the overheads associated with
3377 multi-thread support in other code such as the C++ runtime library.
3379 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3384 * The following bugs are resolved with this release:
3386 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3387 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3388 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3389 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3390 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3391 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3392 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3393 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3394 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3395 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3396 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3397 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3398 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3399 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3400 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3401 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3402 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3403 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3404 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3405 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3406 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3407 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3410 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3411 configuring glibc with:
3412 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3413 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3414 Implemented by H.J. Lu.
3418 + define static_assert
3420 + do not declare gets
3422 + declare at_quick_exit and quick_exit also for ISO C11
3424 + aligned_alloc. NB: The code is deliberately allows the size parameter
3425 to not be a multiple of the alignment. This is a moronic requirement
3426 in the standard but it is only a requirement on the caller, not the
3429 + timespec_get added
3431 + uchar.h support added
3433 + CMPLX, CMPLXF, CMPLXL added
3435 Implemented by Ulrich Drepper.
3437 * Support for the IA-64 has been moved to ports.
3439 * Remove support for anything but ELF binary format
3441 * Checking versions of poll, ppoll added.
3442 Implemented by Ulrich Drepper.
3444 * More generic and 64-bit performance optimizations to math functions.
3445 Implemented by Ulrich Drepper.
3447 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3448 headers and functions available at compile time as they were before
3449 version 2.14. This option will be removed at some time in the future
3450 after the TI-RPC library becomes fully sufficient for the needs of
3451 existing applications.
3453 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3454 Note that glibc is not expected to work with any Linux kernel version
3457 * New header <sys/auxv.h> and function getauxval allowing easy access to
3458 the AT_* key-value pairs passed from the Linux kernel. The header also
3459 defines the HWCAP_* bits associated with the AT_HWCAP key.
3461 * New locales: mag_IN
3463 * New configure option --enable-systemtap builds SystemTap static probes
3464 into libc for setjmp and longjmp and into libpthread for various operations.
3465 So far the setjmp/longjmp probes and some of the libpthread probes are
3466 provided only for i*86 and x86_64.
3467 Implemented by Roland McGrath and Rayson Ho.
3469 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3471 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3474 * More optimized functions for SPARC. Implemented by David S. Miller.
3476 * Improved support for cross-compilation, especially bootstrap builds
3477 without a previously built glibc.
3479 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3480 Chris Metcalf from Tilera.
3482 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3483 now supported for ARM processors.
3485 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3486 as the name of the dynamic linker, to distinguish it from the
3487 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3489 * Support for CRIS has been removed from ports.
3491 * A new class of installed header has been documented for low-level
3492 platform-specific functionality. PowerPC added the first instance with a
3493 function to provide time base register access. Contributed by Tulio
3494 Magno Quites Machado Filho.
3496 * ix86 configurations now install header files that are consistent with
3497 what x86-64 configurations install. These same header files can be used
3498 for -m32, -m64, or -mx32 builds.
3499 Contributed by H.J. Lu.
3501 * Math library bug fixes. A thorough audit of all open math library bugs was
3502 conducted by Joseph Myers. Significant progress was made on many math
3503 library bugs resulting in more accurate exceptions and function results.
3504 Many thanks to all those that contributed including Andreas Jaeger for his
3505 patch review and work on the x87 trigonometric instruction issues.
3507 * Timezone data is no longer installed. Timezone-related binaries and scripts
3508 will continue to be installed. Users should obtain their timezone data from
3509 their distribution provider or from the tzdata package at
3510 <ftp://munnari.oz.au/pub/>.
3514 * The following bugs are resolved with this release:
3516 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3517 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3518 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3519 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3520 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3521 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3522 13484, 13506, 13515, 13523, 13524, 13538, 13540
3524 * New program pldd to list loaded object of a process
3525 Implemented by Ulrich Drepper.
3527 * Add nss_db support back to glibc. No more dependency on Berkeley db
3528 and support for initgroups lookups.
3529 Implemented by Ulrich Drepper.
3531 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3532 Contributed by HJ Lu.
3534 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3535 Contributed by HJ Lu.
3537 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3538 on x86-32 and x86-64.
3539 Contributed by Liubov Dmitrieva.
3541 * Optimized strchr and strrchr for SSE on x86-32.
3542 Contributed by Liubov Dmitrieva.
3544 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3545 for x86-64 and x86-32.
3546 Contributed by Liubov Dmitrieva.
3548 * New interfaces: scandirat, scandirat64
3549 Implemented by Ulrich Drepper.
3551 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3552 Implemented by Ulrich Drepper.
3554 * nscd now also caches the netgroup database.
3555 Implemented by Ulrich Drepper.
3557 * Integrate libm with gcc's -ffinite-math-only option.
3558 Implemented by Ulrich Drepper.
3560 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3561 to math functions. Implemented by Ulrich Drepper.
3563 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3564 Implemented by Ulrich Drepper.
3566 * New Linux interfaces: process_vm_readv, process_vm_writev
3568 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3569 Implemented by Ulrich Drepper.
3571 * Optimized nearbyint and strcasecmp for PPC.
3572 Implemented by Adhemerval Zanella.
3574 * New locales: bho_IN, unm_US, es_CU, ta_LK
3578 * The following bugs are resolved with this release:
3580 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3581 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3582 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3583 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3584 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3585 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3586 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3587 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3588 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3589 12795, 12811, 12813, 12814, 12841
3591 * The RPC implementation in libc is obsoleted. Old programs keep working
3592 but new programs cannot be linked with the routines in libc anymore.
3593 Programs in need of RPC functionality must be linked against TI-RPC.
3594 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3596 Visible changes of this change include (obviously) the inability to link
3597 programs using RPC functions without referencing the TI-RPC library and the
3598 removal of the RPC headers from the glibc headers.
3599 Implemented by Ulrich Drepper.
3601 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3602 syncfs, setns, sendmmsg
3604 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3605 yue_HK, lij_IT, mhr_RU
3607 * New encodings: CP770, CP771, CP772, CP773, CP774
3609 * New program sotruss to trace calls through PLTs
3610 Implemented by Ulrich Drepper.
3612 * The malloc hook implementation is marked deprecated and will be removed
3613 from the default implementation in the next version. The design never
3614 worked ever since the introduction of threads. Even programs which do
3615 not create threads themselves can use multiple threads created internally.
3619 * The following bugs are resolved with this release:
3621 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3622 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3623 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3624 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3627 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3629 * POWER7 optimizations: memset, memcmp, strncmp
3631 * New optimized string functions for x86-64: strnlen (SSE2),
3632 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3633 Implemented by Ulrich Drepper.
3637 * The following bugs are resolved with this release:
3639 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3640 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3641 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3642 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3643 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3644 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3645 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3646 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3648 * New interfaces: pthread_getname_np, pthread_setname_np
3650 * New Linux interface: recvmmsg
3652 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3654 * The dynamic linker now recognizes supported ABI versions from the
3655 EI_ABIVERSION field in the ELF header.
3656 Implemented by Ulrich Drepper.
3658 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3659 will not be used to fill in password fields in the passwd.byname replies.
3660 Instead it is used to synthesize the shadow.byname table, should it be
3661 missing. This is a useful mode in some installations involving Solaris.
3662 Implemented by Ulrich Drepper.
3664 * New locales: kok_IN, sq_MK, cv_RU
3668 * The following bugs are resolved with this release:
3670 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3671 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3672 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3673 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3674 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3675 10789, 10817, 10823, 10840, 10847
3677 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3679 Implemented by Ulrich Drepper.
3681 * Checking version of longjmp added that fails if an uninitialized stack
3682 frame would be created. Implemented by Ulrich Drepper.
3684 * STT_GNU_IFUNC is now supported in static executables.
3685 Implemented by H.J. Lu.
3687 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3689 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3690 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3691 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3692 Contributed by H.J. Lu.
3694 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3695 Implemented by Ulrich Drepper.
3697 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3699 Contributed by H.J. Lu.
3701 * Support for fma instruction in AVX on x86-64.
3702 Implemented by H.J. Lu and Ulrich Drepper.
3704 * AVX support in x86-64 auditing support in ld.so.
3705 Implemented by H.J. Lu.
3707 * STB_GNU_UNIQUE support added.
3708 Implemented by Ulrich Drepper.
3710 * Implement second fallback mode for optimized DNS lookup for even more
3711 broken environments. If two requests from the same port are not
3712 handled correctly close the socket and open a new one before sending
3713 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3714 can be used to select this mode right away, instead of rediscovering the
3715 necessity is every process again.
3716 Implemented by Ulrich Drepper.
3718 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3719 Implemented by Adam Tkac.
3721 * Optimized iconv conversions for S390x.
3722 Implemented by Andreas Krebbel.
3724 * Using condvars with PI mutexes is now more efficient due to kernel
3725 support for requeueing to PI futexes. NPTL support added for x86-64.
3726 Implemented by Ulrich Drepper.
3728 * New locales: ps_AF, my_MM
3732 * The following bugs are resolved with this release:
3734 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3735 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3736 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3737 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3738 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3740 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3741 Implemented by Ulrich Drepper.
3743 * New Linux interfaces: accept4, fallocate, fallocate64.
3744 Implemented by Ulrich Drepper.
3746 * Correct declarations of string function when used in C++ code. This
3747 could lead to compile errors for invalid C++ code.
3749 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3752 * New POSIX 2008 interface: psiginfo
3753 Implemented by Ulrich Drepper.
3755 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3756 Implemented by Ulrich Drepper.
3758 * Support for selecting between multiple function definitions at runtime
3759 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3761 * The libcrypt library can now use the hash function implementations in
3762 NSS. Implemented by Ulrich Drepper.
3764 * The malloc implementation can be compiled to be less memory efficient
3765 but higher performing in multi-threaded programs.
3766 Implemented by Ulrich Drepper.
3768 * New locales: nan_TW@latin, ks_IN
3770 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3771 Implemented by Ulrich Drepper.
3773 * Extended printf hook support. It is possible to use user-defined types
3774 and extend existing format specifiers.
3775 Implemented by Ulrich Drepper.
3777 * Handling for group shadow files has been added.
3778 Implemented by Ulrich Drepper.
3780 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3781 servers (the case, e.g., for some people using the built-in DNS
3782 server in ADSL modems/routers). There is a once-per-process timeout
3783 in case of a broken server. To avoid it, users can run nscd or put
3784 'options single-request' in /etc/resolv.conf.
3785 Implemented by Ulrich Drepper.
3789 * The following bugs are resolved with this release:
3791 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3792 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3793 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3794 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3795 6968, 6974, 6980, 6995, 7008, 7009, 7029
3797 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3798 up at the same time. Implemented by Ulrich Drepper.
3800 * TLS descriptors for LD and GD on x86 and x86-64.
3801 Implemented by Alexandre Oliva.
3803 * getaddrinfo now handles DCCP and UDPlite.
3804 Implemented by Ulrich Drepper.
3806 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3807 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3808 Implemented by Ulrich Drepper.
3810 * New implementation of memmem, strstr, and strcasestr which is O(n).
3811 Implemented by Eric Blake.
3813 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3815 * Implement "e" option for popen to open file descriptor with the
3816 close-on-exec flag set. Implemented by Ulrich Drepper.
3818 * Implement "b" mode for fmemopen. In this mode writes writes don't
3819 implicitly add a NUL byte and seeks from the end of the buffer really
3820 use the buffer end, not the string length as the basis.
3821 Implemented by Ulrich Drepper.
3823 * Many functions, exported and internal, now atomically set the close-on-exec
3824 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3826 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3828 Implemented by Pravin Satpute.
3830 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3834 * The following bugs are resolved with this release:
3836 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3837 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3838 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3839 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3840 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3841 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3842 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3844 * New locales: bo_CN, bo_IN, shs_CA.
3846 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3848 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3849 Implemented by Pravin Satpute.
3851 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3853 * nscd caches now all timeouts for DNS entries
3854 Implemented by Ulrich Drepper.
3856 * nscd is more efficient and wakes up less often.
3857 Implemented by Ulrich Drepper.
3859 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3860 vdprintf, and obstack_vprintf.
3861 Implemented by Jakub Jelinek.
3863 * Faster memset for x86-64.
3864 Implemented by Harsha Jagasia and H.J. Lu.
3866 * Faster memcpy on x86.
3867 Implemented by Ulrich Drepper.
3869 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3870 Implemented by Ulrich Drepper.
3872 * Faster sqrt and sqrtf implemention for some PPC variants.
3873 Implemented by Steven Munroe.
3877 * The following bugs are resolved with this release:
3879 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3880 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3881 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3882 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3883 5071, 5103, 5104, 5112, 5113, 5184, 5186
3885 * More checking functions: fread, fread_unlocked, open*, mq_open.
3886 Implemented by Jakub Jelinek and Ulrich Drepper.
3888 * Extend fortification to C++. Implemented by Jakub Jelinek.
3890 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3891 by not recognizing 'a' as a modifier when those specs are requested.
3892 Implemented by Jakub Jelinek.
3894 * PPC optimizations to math and string functions.
3895 Implemented by Steven Munroe.
3897 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3898 options to be passed. Implemented by Ulrich Drepper.
3900 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3902 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3903 Implemented by Ulrich Drepper.
3905 * Handle private futexes in the NPTL implementation.
3906 Implemented by Jakub Jelinek and Ulrich Drepper.
3908 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3909 Implemented by Roland McGrath and Ulrich Drepper.
3911 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3913 * SHA-256 and SHA-512 based password encryption.
3914 Implemented by Ulrich Drepper.
3916 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3917 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3920 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3921 Implemented by Ulrich Drepper.
3925 * The following bugs are resolved with this release:
3927 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3928 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3929 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3930 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3931 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3932 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3933 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3936 * New Linux interfaces: epoll_pwait, sched_getcpu.
3938 * New generic interfaces: strerror_l.
3940 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3944 * The following bugs are resolved with this release:
3946 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3947 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3948 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3949 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3950 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3951 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3952 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3953 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3954 3225, 3251, 3252, 3253, 3273
3956 * For Linux, the sorting of addresses returned by getaddrinfo now also
3957 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3958 Implemented by Ulrich Drepper.
3960 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3961 Implemented by Ulrich Drepper.
3963 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3965 * New iconv module for MIK. Contributed by Alexander Shopov.
3967 * For sites with broken group and/or passwd database, the auto-propagate
3968 option of nscd can prevent creating ID lookup entries from the results
3969 of a name lookup and vice versa. This usually is no problem but some
3970 site might have problems with the default behavior.
3971 Implemented by Ulrich Drepper.
3973 * Iterating over entire database in NIS can be slow. With the
3974 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3975 to trade time for memory. The entire database will be read at once.
3976 Implemented by Ulrich Drepper.
3978 * The interfaces introduced in RFC 3542 have been implemented by
3981 * Support for the new ELF hash table format was added by Ulrich Drepper.
3983 * Support for priority inheritance mutexes added by Jakub Jelinek and
3986 * Support for priority protected mutexes added by Jakub Jelinek.
3988 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3989 or_IN, csb_PL, fy_NL, sr_ME.
3993 * More overflow detection functions.
3995 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3996 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3998 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3999 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
4000 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
4001 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
4002 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
4003 by Masahide Washizawa.
4005 * It is now possible to install audit libraries for the dynamic linker using
4006 LD_AUDIT. Implemented by Ulrich Drepper.
4008 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
4009 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
4010 For a libc and libpthread that works well on Linux 2.4 kernels, we
4011 recommend using the stable 2.3 branch.
4013 * The new function open_wmemstream acts like open_memstream,
4014 but uses a wchar_t wide character buffer.
4016 * The new function ppoll is an improved version of poll, similar to pselect.
4018 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
4019 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
4020 renameat, unlinkat, linkat, symlinkat, readlinkat.
4022 * New Linux kernel system calls: unshare,
4023 inotify_init, inotify_add_watch, inotify_rm_watch.
4025 * The euidaccess function is now also known by the alias eaccess,
4026 for compatibility with some other systems.
4028 * Timezone data updated to 2006b version.
4032 * The following bugs are resolved with this release:
4034 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
4035 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
4036 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
4037 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
4038 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
4039 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
4041 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4043 * As of this release, GCC 4 can be used to compile the C Library.
4045 * Timezone data updated to 2005m version.
4049 * The following bugs are resolved with this release:
4051 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
4052 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
4053 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
4054 777, 787, 821, 822, 823, 825
4056 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4060 * Support for RFC 3678. Real implementations exist only for Linux so far.
4061 Implemented by Ulrich Drepper.
4063 * nscd can now cache entries persistently. Expiring entries are reloaded.
4064 For speedups the cache can be shared in memory with client processes.
4065 Implemented by Ulrich Drepper.
4067 * nscd can now perform SELinux checks.
4068 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
4070 * getaddrinfo queries are now cached. Canonical name lookup is performed
4072 Implemented by Ulrich Drepper.
4074 * The nothrow function attribute is used when headers are used by gcc when
4075 compiling C code. This can avoid significant amounts of exception
4078 * The malloc functions perform more error checking and are stricter when
4079 it comes to reacting on errors. The default action is to terminate
4080 the process after showing an error message. Implemented by Ulrich Drepper.
4082 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
4083 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
4085 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
4086 namespaces using the new function dlmopen(). This feature is of course,
4087 like most other dynamic loading functionality, not available in statically
4088 linked applications. Implemented by Ulrich Drepper.
4090 * Low-overhead boundary checking variants of string and some stdio functions
4091 were added. These are to be used in conjunction with a gcc patch by
4092 Jakub Jelinek which adds calls to these functions if possible.
4093 Implemented by Jakub Jelinek and Ulrich Drepper.
4095 * Old code for several operating systems and machine architectures that
4096 have not been in working condition in a long time have been removed from
4097 the main source tree maintained by the GNU C Library's maintainers.
4098 These files are now reside in the separate `ports' source module
4099 that is usable as an add-on when building the library.
4103 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
4104 interrogate the dynamic linker, compatible with the Solaris interface.
4106 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
4107 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
4109 * getifaddrs now uses the netlink interface on Linux to get its information.
4110 Implemented by Thorsten Kukuk.
4112 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
4113 Implemented by Ulrich Drepper.
4115 * support for non-executable stacks on x86 has been added. Changes mostly
4118 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
4121 * getaddrinfo now performs destination address selection according to
4126 * Thread-safe interfaces for many functions that access locale data
4127 were added in version 2.3, but these features were omitted from NEWS.
4128 Many functions have variants with an `_l' suffix that take a `locale_t'
4129 object as a parameter rather than consulting the current locale.
4130 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
4131 create and maintain `locale_t' objects. Additionally, the new function
4132 `uselocale' sets "the current locale" (as used by functions not so
4133 parameterized) set for an individual thread. These features were added
4134 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
4136 * The functions getresuid, getresgid, setresuid, and setresgid, which
4137 have long been available on Linux, are now declared in <unistd.h>
4138 and are now also available on the Hurd.
4140 * ELF thread-local storage support (TLS) now works on x86-64.
4142 * The new dynamic string token $LIB is expanded in shared library names.
4143 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
4145 * Aldy Hernandez contributed complete software floating point support for
4146 PowerPC machines with no FPU.
4148 * fexecve is implemented on Linux.
4150 * The `btowc' function should work at least twice as fast due to
4151 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
4153 * With approriate thread add-ons cancelable functions are now implemented
4154 in libc.so as well. No need to call the function in libpthread. This
4155 change allowed to finally disable the incorrect and expensive handling
4156 of weak definition in ld.so.
4158 * Yet more PLT entries in libc.so have been removed. We finally arrived
4159 at the bare minimum. Startup times improved appropriately.
4161 * Support for the new Linux/x86 system call interface was added. The
4162 AT_SYSINFO auxiliary vector entry is recognized and handled.
4166 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
4169 * iconv (the program and the interface) now accepts empty names (excluding
4170 options like //TRANSLIT) to mean "use charset of current locale".
4172 * localedef can now transliterate characters in strings which are not in
4173 the provided charmap. The information from the input locale is used.
4175 * Prelinking support was added for ELF targets. This requires additional
4176 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
4178 * Read-only stdio streams now use mmap to speed up operation by eliminating
4179 copying and buffer underflows. To use add 'm' to the mode string of
4180 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
4182 * The malloc functions were completely rewritten by Wolfram Gloger based
4183 on Doug Lea's malloc-2.7.0.c.
4185 * Isamu Hasegawa contributed a completely new and POSIX-conformant
4186 implementation of regex.
4188 * Bruno Haible upgraded the iconv and locale implementation to support
4191 * Contents of the LC_* and LANG environment variables in the CEN style are
4192 not recognized anymore. It never was used. Change by Ulrich Drepper.
4194 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
4195 thread-local storage (TLS) ABI on some platforms.
4196 Changes by Ulrich Drepper. SH support by Kaz Kojima.
4198 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
4199 EUC-JISX0213, and TSCII.
4201 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
4202 BSD-compatible interface for getting all network interface addresses.
4203 Implementation for IPv4 by Roland McGrath.
4205 * Loading of locale data is faster due to the introduction of a locale
4206 archive. Implemented by Roland McGrath and Ulrich Drepper.
4208 * Startup times are significantly reduced by not using exported functions
4209 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
4212 * Steven Munroe contributed a port to PowerPC64/Linux.
4216 * The Hurd now uses the GNU libio implementation of stdio rather than the
4217 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
4219 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4220 with the same behavior as the Linux system call of the same name.
4224 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4225 128-bit long double format.
4227 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4228 IBM1161, and IBM1162 charsets.
4230 * Andreas Jaeger contributed a port to x86-64/Linux.
4232 * Peter Bruin contributed a port to PowerPC/Hurd.
4234 * libc functions using I/O streams now can handle wide-oriented streams
4237 * optimizations in the dynamic linker. Binaries created by recent binutils
4238 versions start up quicker due to reduced time spent on relocations.
4240 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4244 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4245 asinh, atanh, j0 for the 128-bit long double format.
4247 * Bruno Haible updated all the code handling Unicode in some form to
4248 support Unicode 3.1.
4250 * Speed of regex for single-byte locales is back to previous levels.
4251 Patch by Isamu Hasegawa.
4253 * Alpha, SPARC, and IA-64 now also using floating stacks.
4255 * Startup time of internationalized applications greatly improved through
4256 iconv cache. Use iconvconfig to generate the cache file.
4257 Contributed by Ulrich Drepper.
4259 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4260 severe performance problems. Patch by David Mosberger.
4262 * The Hurd port got a lot more functionality like AIO, various stdio
4263 extensions, etc. Mainly done by Roland McGrath.
4265 * mtrace can now lookup symbols in shared libraries.
4269 * Intel's IA-64 math library is largely integrated. It provides fast and
4270 accurate implementatations for most basic and standard math functions
4271 in float, double, and long double format.
4273 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4274 and asin for the 96-bit long double format and asin, log, tan for the
4275 128-bit long double format.
4277 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4278 The basic double functions exist today. Contributed by Abraham Ziv
4279 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4280 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4282 * An asynchronous name lookup library was added. The interface is designed
4283 after POSIX AIO. The proposal was circulated beforehand to get comments.
4284 No negative ones came in. Implemented by Ulrich Drepper.
4286 * Port to S390/64bit contributed by Martin Schwidefsky
4287 <schwidefsky@de.ibm.com>.
4289 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4290 of functions for Linux/IA-64.
4292 * The RPC code is now thread safe. Threads can now use the same service
4293 of different services at the same time. Patch by Eric Norum
4294 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4296 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4297 family of functions for Linux/S390.
4299 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4300 of functions for Linux/x86.
4302 * Port to Linux/CRIS contributed by Axis Communications.
4306 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4307 we can now check for namespace violations and missing declarations. The
4308 result is that almost all headers are now Unix-compliant (as defined in
4309 the upcoming XPG6). The negative side is that some programs might need
4310 corrections, too, if they depend on the incorrect form of the headers in
4311 previous versions which defined too many symbols and included too many
4314 * regex now handles multibyte character sets correctly.
4315 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4317 * iconv (the program) does now conform to the upcoming XPG6 and handles
4318 charmaps. Instead of the charset names the path of charmaps can be
4319 provided and the conversion happens based on this data.
4320 Contributed by Ulrich Drepper.
4322 * The locale program now provides detailed information about the installed
4327 only lists the names of the supported locales
4331 provides details such as country, language, and codeset name.
4332 Contributed by Ulrich Drepper.
4336 * The gencat program now parses the input file according to the charset
4337 selected by the LC_CTYPE category. This is important for stateful
4338 character sets. To make generating catalogs easier there is a way
4339 to overwrite the charset selected by the locale: before the first
4340 message or $ quote line the catalog can contain a line like
4342 $ codeset=ISO-8859-2
4344 to select the charset (ISO-8859-2 in this case).
4346 Implemented by Shinya Hanataka and Ulrich Drepper.
4348 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4349 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4350 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4352 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4353 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4355 * The LANGUAGE environment variable is now ignored unless the locale is
4356 changed from the default "C" locale.
4358 * The usual bug fixes.
4362 * Greg McGary added runtime support for bounds checking using gcc's
4363 new -fbounded-pointers option. ix86 target is complete. PowerPC
4366 * Thorsten Kukuk added secure mode support to nscd.
4368 * The Berkeley db libraries have been removed.
4370 Related, the nss_db module is now in a separate package since it
4371 obviously requires a database library being available.
4373 * Wide character I/O streams implemented by Ulrich Drepper.
4375 * Functions from the extended socket API added by Ulrich Drepper.
4377 * Functions feenableexcept and fedisableexcept to control the
4378 behaviour of individual exceptions have been added by Andreas Jaeger.
4380 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4382 * The resolver code has been updated from bind 8.2.3-T5B which supports
4383 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4386 This change could in some situations effect backward compatibility. Since
4387 now `_res' is a thread-local instead of a global variable, modifying it
4388 in one thread does not have any effect in other threads.
4390 The resolver library was also extended to allow IPv6 as the transport
4391 protocol for the requests. This work was done by Stig Venaas.
4393 * Compatibility code for K&R C compilers has been removed from the
4394 header files. A ISO C compiler is needed to use the library
4395 (conforming to either C89 or C99 standard).
4397 * Complete rewrite of the localedef program to support multibyte character
4398 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4399 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4400 Implemented by Ulrich Drepper.
4402 Bruno Haible significantly improved the generation and use of the data
4403 structures for the wide character tables.
4405 * Plural handling in gettext implemented by Ulrich Drepper.
4407 * The utmp daemon has been removed.
4409 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4411 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4414 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4416 * POSIX barriers implemented by Kaz Kylheku.
4418 * POSIX spawn function family implemented by Ulrich Drepper.
4420 * POSIX spinlocks are now available.
4422 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4424 * the configure option --enable-kernel=X.Y.Z allows to strip out
4425 compatibility for kernel versions before X.Y.Z. This is currently only
4426 implemented for Linux.
4428 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4429 field sin6_scope_id. This means that all programs using IPv6 should be
4430 recompiled. Don't expect binary compatibility with previous glibc
4433 * various conversion modules for IBM character sets contributed by
4436 * IA-64 port by Jes Sorensen and HJ Lu.
4450 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4452 * New xdr functions are added; some rpc functions are now 64bit clean.
4454 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4456 * Fixed known incompatibilities with glibc 2.0.
4458 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4460 * Optimized a number of functions (especially the ELF dynamic loader).
4462 * Update timezone data files.
4464 * lots of charmaps corrections
4466 * some new locale definitions and charmaps
4471 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4472 conform to POSIX member type requirements. Symbol versions have been
4473 adjusted as needed within the library, and for direct use by applications,
4474 but there is potential for problems if third-party libraries use
4475 struct timeval as part of their interface. This does not present
4476 a problem for X and other "essential" system libraries.
4478 * An additional locale model to support C++ Standard Library locale
4479 model and probably more was implemented by Ulrich Drepper.
4481 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4484 * Miles Bader provided the `argp' function family to support hierachical
4485 command line argument parsing, layered on top of getopt.
4487 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4489 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4492 * scanf recognizes the %a and %A format for scanning floating point numbers.
4494 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4495 information and interfaces for the available integer types.
4497 * about 130 new math functions were added to implement the ISO C9x math
4500 * the new header <complex.h> contains definitions of the complex math
4501 functions from ISO C 9X.
4503 * the new header <tgmath.h> defines generic macros to use complex or
4504 real valued functions.
4506 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4508 * Andreas Jaeger provided a test suite for the math library.
4510 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4512 * Ulrich Drepper added character set conversion functions (iconv).
4514 * Optimized string functions have been added.
4516 * The localedata addon is now part of glibc.
4518 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4520 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4521 daemon for NSS (nscd).
4523 Missing a better place here are some numbers on improvements. Under
4524 Linux 2.1.125 un-tar-ing the kernel sources takes
4528 using local files 12.19s 6.88s 22.91s
4530 using NIS 13.92s 8.91s 26.34s
4532 using NIS & nscd 10.37s 7.34s 25.30s
4534 using NIS+ 27.57s 30.37s 640.46s
4536 using NIS+ & nscd 10.25s 7.83s 26.51s
4538 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4540 Keep in mind that non-namelookup related operations dominate above times.
4541 It was just a common complain that using NIS+ unpacking the kernel is
4544 [1] The old nscd implementation is not available anymore since it was
4545 distributed with glibc up to version 2.0.98 and thus is now replaced.
4547 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4549 * Mark Kettenis provided a Hesiod NSS module.
4551 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4552 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4554 * rcmd can now handle netgroups (Dick Streefland).
4556 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4559 * Support for the IPv6 protocol has been added to the socket API, as per the
4560 latest draft standards.
4562 * Support for Linux 2.2 has been added.
4564 * Interface changes relative to the latest 2.0.x release:
4565 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4566 addseverity NEW: Unix98
4567 alphasort64 NEW: LFS
4568 argp_err_exit_status NEW: argp, GNU ext
4569 argp_error NEW: argp, GNU ext
4570 argp_failure NEW: argp, GNU ext
4571 argp_help NEW: argp, GNU ext
4572 argp_parse NEW: argp, GNU ext
4573 argp_program_bug_address NEW: argp, GNU ext
4574 argp_program_version NEW: argp, GNU ext
4575 argp_program_version_hook NEW: argp, GNU ext
4576 argp_state_help NEW: argp, GNU ext
4577 argp_usage NEW: argp, GNU ext
4578 authdes_create NEW: Secure RPC
4579 authdes_getucred NEW: Secure RPC
4580 authdes_pk_create NEW: Secure RPC
4581 backtrace NEW: GNU ext.
4582 backtrace_symbols NEW: GNU ext.
4583 backtrace_symbols_fd NEW: GNU ext.
4585 cacosf NEW: ISO C 9x
4586 cacosh NEW: ISO C 9x
4587 cacoshf NEW: ISO C 9x
4588 cacoshl NEW: ISO C 9x
4589 cacosl NEW: ISO C 9x
4596 casinf NEW: ISO C 9x
4597 casinh NEW: ISO C 9x
4598 casinhf NEW: ISO C 9x
4599 casinhl NEW: ISO C 9x
4600 casinl NEW: ISO C 9x
4602 catanf NEW: ISO C 9x
4603 catanh NEW: ISO C 9x
4604 catanhf NEW: ISO C 9x
4605 catanhl NEW: ISO C 9x
4606 catanl NEW: ISO C 9x
4607 cbc_crypt NEW: Secure RPC
4611 ccoshf NEW: ISO C 9x
4612 ccoshl NEW: ISO C 9x
4618 cimagf NEW: ISO C 9x
4619 cimagl NEW: ISO C 9x
4620 clearerr_locked REMOVED
4621 clntunix_create NEW: sunrpc ext
4623 clog10 NEW: ISO C 9x
4624 clog10f NEW: ISO C 9x
4625 clog10l NEW: ISO C 9x
4635 cprojf NEW: ISO C 9x
4636 cprojl NEW: ISO C 9x
4638 crealf NEW: ISO C 9x
4639 creall NEW: ISO C 9x
4644 csinhf NEW: ISO C 9x
4645 csinhl NEW: ISO C 9x
4648 csqrtf NEW: ISO C 9x
4649 csqrtl NEW: ISO C 9x
4653 ctanhf NEW: ISO C 9x
4654 ctanhl NEW: ISO C 9x
4656 des_setparity NEW: Secure RPC
4657 ecb_crypt NEW: Secure RPC
4658 endutxent NEW: Unix98
4660 exp10f NEW: ISO C 9x
4661 exp10l NEW: ISO C 9x
4665 fattach NEW: STREAMS
4666 fdetach NEW: STREAMS
4670 feclearexcept NEW: ISO C 9x
4671 fegetenv NEW: ISO C 9x
4672 fegetexceptflag NEW: ISO C 9x
4673 fegetround NEW: ISO C 9x
4674 feholdexcept NEW: ISO C 9x
4676 feraiseexcept NEW: ISO C 9x
4677 ferror_locked REMOVED
4678 fesetenv NEW: ISO C 9x
4679 fesetexceptflag NEW: ISO C 9x
4680 fesetround NEW: ISO C 9x
4681 fetestexcept NEW: ISO C 9x
4682 feupdateenv NEW: ISO C 9x
4683 fflush_locked REMOVED
4687 fgets_unlocked NEW: GNU ext.
4688 fileno_locked REMOVED
4700 fputc_locked REMOVED
4701 fputs_unlocked NEW: GNU ext.
4702 fread_unlocked NEW: GNU ext.
4707 fstatvfs NEW: Unix98
4711 ftruncate64 NEW: LFS
4713 fwrite_unlocked NEW: GNU ext.
4714 gai_strerror NEW: IPv6
4718 getchar_locked REMOVED
4720 getdate_err NEW: Unix98
4721 getdate_r NEW: GNU ext.
4723 getnameinfo NEW: IPv6
4724 getnetname NEW: Secure RPC
4725 getpmsg NEW: STREAMS
4726 getpt NEW: Unix98 PTY
4727 getrlimit64 NEW: LFS
4728 getutxent NEW: Unix98
4729 getutxid NEW: Unix98
4730 getutxline NEW: Unix98
4731 glob64 NEW: GNU ext.
4732 globfree64 NEW: GNU ext.
4733 gnu_get_libc_release NEW: GNU ext.
4734 gnu_get_libc_version NEW: GNU ext.
4735 grantpt NEW: Unix98 PTY
4736 host2netname NEW: Secure RPC
4738 iconv_close NEW: iconv
4739 iconv_open NEW: iconv
4740 if_freenameindex NEW: IPv6
4741 if_indextoname NEW: IPv6
4742 if_nameindex NEW: IPv6
4743 if_nametoindex NEW: IPv6
4744 in6addr_any NEW: IPv6
4745 in6addr_loopback NEW: IPv6
4746 inet6_isipv4mapped NEW: IPv6
4747 isastream NEW: STREAMS
4748 iswblank NEW: GNU ext.
4749 key_decryptsession NEW: Secure RPC
4750 key_decryptsession_pk NEW: Secure RPC
4751 key_encryptsession NEW: Secure RPC
4752 key_encryptsession_pk NEW: Secure RPC
4753 key_gendes NEW: Secure RPC
4754 key_get_conv NEW: Secure RPC
4755 key_secretkey_is_set NEW: Secure RPC
4756 key_setnet NEW: Secure RPC
4757 key_setsecret NEW: Secure RPC
4758 llrint NEW: ISO C 9x
4759 llrintf NEW: ISO C 9x
4760 llrintl NEW: ISO C 9x
4761 llround NEW: ISO C 9x
4762 llroundf NEW: ISO C 9x
4763 llroundl NEW: ISO C 9x
4768 lrintf NEW: ISO C 9x
4769 lrintl NEW: ISO C 9x
4770 lround NEW: ISO C 9x
4771 lroundf NEW: ISO C 9x
4772 lroundl NEW: ISO C 9x
4774 makecontext NEW: Unix98
4775 mempcpy NEW: GNU ext.
4778 modify_ldt NEW: kernel
4782 nearbyint NEW: ISO C 9x
4783 nearbyintf NEW: ISO C 9x
4784 nearbyintl NEW: ISO C 9x
4785 netname2host NEW: Secure RPC
4786 netname2user NEW: Secure RPC
4787 nexttoward NEW: ISO C 9x
4788 nexttowardf NEW: ISO C 9x
4789 nexttowardl NEW: ISO C 9x
4793 passwd2des NEW: Secure RPC
4795 pow10f NEW: GNU ext.
4796 pow10l NEW: GNU ext.
4799 printf_size NEW: GNU ext.
4800 printf_size_info NEW: GNU ext.
4801 profil_counter REMOVED
4802 pthread_mutexattr_getkind_np REPLACED
4803 pthread_mutexattr_setkind_np REPLACED
4804 ptsname NEW: Unix98 PTY
4805 ptsname_r NEW: Unix98 PTY
4807 putchar_locked REMOVED
4808 putgrent NEW: GNU ext.
4810 putpmsg NEW: STREAMS
4811 pututxline NEW: Unix98
4815 readdir64_r NEW: LFS
4816 remquo NEW: ISO C 9x
4817 remquof NEW: ISO C 9x
4818 remquol NEW: ISO C 9x
4820 roundf NEW: ISO C 9x
4821 roundl NEW: ISO C 9x
4823 scalbln NEW: ISO C 9x
4824 scalblnf NEW: ISO C 9x
4825 scalblnl NEW: ISO C 9x
4827 sendfile NEW: kernel
4828 setcontext NEW: Unix98
4829 setrlimit64 NEW: LFS
4830 setutxent NEW: Unix98
4832 sigignore NEW: Unix98
4833 sigqueue NEW: POSIX.1b
4834 sigrelse NEW: Unix98
4835 sigset NEW: POSIX.1b
4836 sigtimedwait NEW: POSIX.1b
4837 sigwaitinfo NEW: POSIX.1b
4838 sincos NEW: GNU ext.
4839 sincosf NEW: GNU ext.
4840 sincosl NEW: GNU ext.
4844 strcasestr NEW: GNU ext.
4845 strtoimax NEW: ISO C 9x
4846 strtoumax NEW: ISO C 9x
4847 strverscmp NEW: GNU ext.
4848 svcauthdes_stats NEW: Secure RPC
4849 svcunix_create NEW: sunrpc etx
4850 svcunixfd_create NEW: sunrpc ext
4851 swapcontext NEW: Unix98
4852 tcgetsid NEW: Unix98 PTY
4853 tdestroy NEW: GNU ext.
4854 tgamma NEW: ISO C 9x
4855 tgammaf NEW: ISO C 9x
4856 tgammal NEW: ISO C 9x
4860 truncf NEW: ISO C 9x
4861 truncl NEW: ISO C 9x
4863 unlockpt NEW: Unix98 PTY
4864 updwtmpx NEW: Unix98
4865 user2netname NEW: Secure RPC
4866 utmpxname NEW: Unix98
4867 versionsort NEW: GNU ext.
4868 versionsort64 NEW: GNU ext.
4870 wcscasecmp NEW: GNU ext.
4871 wcsncasecmp NEW: GNU ext.
4872 wcsnlen NEW: GNU ext.
4873 wcstoimax NEW: ISO C 9x
4874 wcstoll NEW: ISO C 9x
4875 wcstoull NEW: ISO C 9x
4876 wcstoumax NEW: ISO C 9x
4878 wordexp NEW: POSIX.2
4879 wordfree NEW: POSIX.2
4880 write_profiling REMOVED
4881 xdecrypt NEW: Secure RPC
4882 xdr_authdes_cred NEW: Secure RPC
4883 xdr_authdes_verf NEW: Secure RPC
4884 xdr_cryptkeyarg NEW: Secure RPC
4885 xdr_cryptkeyarg2 NEW: Secure RPC
4886 xdr_cryptkeyres NEW: Secure RPC
4887 xdr_getcredres NEW: Secure RPC
4888 xdr_key_netstarg NEW: Secure RPC
4889 xdr_key_netstres NEW: Secure RPC
4890 xdr_keybuf NEW: Secure RPC
4891 xdr_keystatus NEW: Secure RPC
4892 xdr_netnamestr NEW: Secure RPC
4893 xdr_sizeof NEW: Secure RPC
4894 xdr_unixcred NEW: sunrpc ext
4895 xencrypt NEW: Secure RPC
4896 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4907 * inet_ntoa is thread-safe
4909 * updwtmp is moved from libutil to libc
4911 * rewrite of cbrt function
4913 * update of timezone data
4927 * add atoll function
4929 * fix complex problems in Berkeley DB code
4931 * fix math functions
4935 * fixed lots of header problems (especially Linux/GNU specific)
4937 * dynamic loader preserves all registers
4939 * Roland McGrath provided support for handling of auxiliary objects in
4940 the ELF dynamic loader.
4942 * support for parallel builds is improved
4946 * GNU extensions are no longer declared by default. To enable them you
4947 must define the macro `_GNU_SOURCE' in your program or compile with
4950 * The library has changed from using GNU ld symbol aliases to using weak
4951 symbols where available. The ELF object file format supports weak
4952 symbols; GNU ld also supports weak symbols in the a.out format. (There
4953 is also now support for other GNU ld extensions in ELF. Use the
4954 `--with-elf' option to configure to indicate you have ELF, and
4955 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4956 of many files which contained only symbol aliases, reducing the size of
4957 the source and the compiled library; many other files were renamed to
4958 less cryptic names previously occupied by the symbol alias files.
4959 There is a new header file <elf.h> for programs which operate on
4960 files in the ELF format.
4962 * Converted to Autoconf version 2, so `configure' has more options.
4963 Run `configure --help' to see the details.
4965 * The library can now be configured to build profiling, highly-optimized
4966 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4967 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4968 `configure' enable building these extra libraries. The shared library is
4969 built by default when using both ELF and GNU ld. When shared libraries
4970 are enabled, the new library `-ldl' is available for arbitrary run-time
4971 loading of shared objects; its interface is defined in <dlfcn.h>. The
4972 new header file <link.h> gives access to the internals of the run-time
4973 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4974 application of same name on other systems and it provides information
4975 about dynamically linked binaries.
4977 * The C library now provides the run-time support code for profiling
4978 executables compiled with `-pg'. Programs can control the profiling code
4979 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4980 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4981 the support for this file format was contributed by David Mosberger-Tang.
4983 * The math code has been replaced with a math library based on fdlibm from
4984 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4985 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4986 functions. The math functions now reside in a separate library, so
4987 programs using them will need to use `-lm' their linking commands.
4989 * John C. Bowman contributed optimized ix87 assembler inline functions.
4991 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4992 mechanism similar to that found in Solaris 2. This is now used for the
4993 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4994 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4995 file controls what services are used for each individual database. This
4996 works by loading shared libraries with names specified in `nsswitch.conf',
4997 so service modules can be changed or added at any time without even
4998 relinking any program. Currently there are the file, db, and NIS based
4999 NSS services available.
5001 * The new functions `strtoq' and `strtouq' parse integer values from
5002 strings, like `strtol' and `strtoul', but they return `long long int' and
5003 `unsigned long long int' values, respectively (64-bit quantities).
5005 * The new functions `strtof' and `strtold' parse floating-point values from
5006 strings, like `strtod', but they return `float' and `long double' values,
5007 respectively (on some machines `double' and `long double' are the same).
5009 * Ulrich Drepper has contributed new implementations of the floating-point
5010 printing and reading code used in the `printf' family of functions and
5011 `strtod', `strtof', and `strtold'. These new functions are perfectly
5012 accurate, and much faster than the old ones.
5014 * The implementation of the POSIX locale model was completely rewritten by
5015 Ulrich Drepper. This includes the new programs `localedef' and `locale'
5016 to compile the POSIX locale definition.
5018 * The former dummy implementations of the strcoll and strxfrm function are
5019 now replaced by fully functional code contributed by Ulrich Drepper. The
5020 collation information comes from the POSIX locale definitions.
5022 * The new header <langinfo.h> defines an interface for accessing
5023 various locale-dependent data (using the locale chosen with `setlocale').
5025 * Ulrich Drepper has contributed a new suite of functions for operation on
5026 wide-character and multibyte-character strings, in <wchar.h>;
5027 and classification and case conversion of wide characters, in <wctype.h>.
5028 These new functions are conforming to the ISO C, Amendement 1 specification.
5030 * There is now a second implementation of the standard I/O library available.
5031 It comes from GNU libg++ as was written by Per Bothner, heavily modified
5032 by Hongjiu Lu and made thread safe by Ulrich Drepper.
5034 * You can now use positional parameter specifications in format strings
5035 for the `printf' and `scanf' families of functions. For example,
5036 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
5037 ``Number 6, Mr Jones''. This is mainly useful when providing different
5038 format strings for different languages, whose grammars may dictate
5039 different orderings of the values being printed. To support this
5040 feature, the interface for `register_printf_handler' has changed; see
5041 the header file <printf.h> for details.
5043 * The `printf' and `scanf' families of functions now understand a new
5044 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
5045 to group numbers as indicated by the locale; for `scanf' and friends, this
5046 says to accept as valid only a number with all the proper grouping
5047 separators in the right places. In the default "C" locale, numbers are
5048 not grouped; but locales for specific countries will define the usual
5049 conventions (i.e. separate thousands with `,' in the US locale).
5051 * The pgrp functions have been regularized, slightly incompatibly but much
5052 less confusingly. The core functions are now `getpgid' and `setpgid',
5053 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
5054 argument) and BSD `setpgrp' (identical to `setpgid') functions are
5055 provided for compatibility. There is no longer an incompatible `getpgrp'
5056 with an argument declared under _BSD_SOURCE; no BSD code uses it.
5058 * The new header file <fts.h> and suite of functions simplify programs that
5059 operate on directory trees. This code comes from 4.4 BSD.
5061 * The resolver code has been updated from the BIND 4.9.5-P1 release.
5062 Parts of the code were heavily modified by Ulrich Drepper to fit in the
5063 NSS scheme used in glibc.
5065 * There is a new malloc debugging hook `__memalign_hook'.
5067 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
5068 `unsigned int' in <sys/types.h>. These are for compatibility only and
5069 their use is discouraged.
5071 * The `-lmcheck' library to enable standard malloc debugging hooks is now
5072 done differently, so that it works even without GNU ld.
5074 * New function `euidaccess' checks allowed access to a file like `access',
5075 but using the effective IDs instead of the real IDs.
5077 * The time zone data files have been updated for the latest and greatest
5078 local time conventions of the countries of the world.
5080 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
5083 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
5084 for formatting floating-point numbers. They are provided only for
5085 compatibility; new programs should use `sprintf' instead. There are
5086 also equivalent function for the `long double' floating-point type and
5087 all functions also exist in a reentrant form.
5089 * The new auxiliary library `-lutil' from 4.4 BSD contains various
5090 functions for maintaining the login-record files (primarily of use to
5091 system programs such as `login'), and convenient functions for
5092 allocating and initializing a pseudo-terminal (pty) device.
5094 * Ulrich Drepper has contributed new support for System V style
5095 shared memory and IPC on systems that support it.
5097 * Ulrich Drepper has contributed several miscellaneous new functions found
5098 in System V: The `hsearch' family of functions provide an effective
5099 implementation of hash tables; `a64l' and `l64a' provide a very simple
5100 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
5103 * Ulrich Drepper has contributed new reentrant counterparts for the
5104 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
5106 * Ulrich Drepper has contributed new, highly-optimized versions of several
5107 string functions for the i486/Pentium family of processors.
5109 * Ulrich Drepper has updated the Linux-specific code, based largely
5110 on work done in Hongjiu Lu's version of GNU libc for Linux.
5111 The GNU library now supports Linux versions 2.0.10 and later,
5112 using the ELF object file format (i[3456]86-*-linux).
5114 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
5116 * David Mosberger-Tang and Richard Henderson have ported the C library
5117 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
5118 dynamic linking support for ELF/Alpha.
5120 * Richard Henderson contributed several Alpha optimized assembler function
5121 for arithmetic and string handling.
5123 * Ulrich Drepper has contributed a new set of message catalog functions to
5124 support multiple languages using the <libintl.h> interface, for use with
5125 his new package GNU gettext. Translation volunteers have contributed
5126 catalogs of the library's messages in Spanish, German, and Korean.
5128 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
5129 program and the `catgets' function for reading the catalog files it
5130 creates. (The <libintl.h> interface is preferred; we include the
5131 <nl_types.h> interface using `catgets' only for source compatibility with
5132 programs already written to use it.)
5134 * New header file <values.h> gives SVID-compatible names for <limits.h>
5137 * Various new macros, declarations, and small header files for compatibility
5140 * New function `group_member' is a convenient way to check if a process has
5141 a given effective group ID.
5143 * When using GCC 2.7 and later, the socket functions are now declared in a
5144 special way so that passing an argument of type `struct sockaddr_in *',
5145 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
5146 `struct sockaddr *' type, does not generate a type-clash warning.
5148 * New function `error' declared in header file <error.h> is a convenient
5149 function for printing error messages and optionally exiting; this is the
5150 canonical function used in GNU programs. The new functions `err', `warn',
5151 and friends in header file <err.h> are the canonical 4.4 BSD interface for
5152 doing the same thing.
5154 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
5155 POSIX.2 `glob' function to do ~ and {...} expansion.
5157 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
5158 `clearenv' which is used in POSIX.9 is also available.
5160 * New function `getsid' returns session ID number on systems that support it.
5162 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
5163 files <db.h> and <mpool.h> provide a rich set of functions for several
5164 types of simple databases stored in memory and in files, and <ndbm.h> is
5165 an old `ndbm'-compatible interface using the `db' functions. Link with
5166 `-ldb' to get these functions.
5168 * New macro `strdupa' copies a string like `strdup', but uses local stack
5169 space from `alloca' instead of dynamic heap space from `malloc'.
5171 * New function `strnlen' is like `strlen' but searches only a given maximum
5172 number of characters for the null terminator. `stpncpy', `strndup' and
5173 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
5176 * New function `statfs' in header <sys/statfs.h>.
5178 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
5179 provide convenient functions for operating on blocks of null-terminated
5182 * A new suite of functions in <utmp.h> handle all the details of reading
5183 and writing the utmp file.
5185 * An implementation of the NIS/YP(tm) based NSS service was contributed by
5188 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
5189 completely POSIX compliant and also implemented the extended functionality
5190 to handle alternate digit representation and alternate era date formats.
5192 * Ulrich Drepper provided an implementation of the `strptime' function
5193 defined in XPG4.2 which transforms a string into a `struct tm' value.
5195 * Paul Eggert provided the tzselect shell script as part of the timezone
5196 code. The shell script makes it easy to select the correct timezone
5199 * The implementation of the malloc family of functions is completely replaced
5200 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
5201 The implementation uses the mmap function (if available) and it is
5202 optimized for the use in multi threaded programs.
5204 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
5205 functions. This new functionality is usable by specifying a special
5206 salt string and it is compatible with implementation on *BSD systems.
5208 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
5209 `getsubopt' to handle second level command line options, `bsd_signal'
5210 to access BSD style `signal' functionality, the obsolete `regexp' style
5213 * the `lchown' function is available on system which support this
5216 * The implementation of the shadow password handling function was contributed
5219 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5221 * POSIX.1g support was added. The <sys/select.h> header is available,
5222 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5223 implementation of `getaddrinfo'.
5227 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5229 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5230 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5232 * The new function `daemon' from 4.4 BSD is useful for server programs that
5233 want to put themselves in the background.
5235 * Joel Sherrill has contributed support for several standalone boards that
5236 run without an operating system.
5238 * `printf', `scanf' and friends now accept a `q' type modifier for long
5239 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5241 * All of the code taken from BSD (notably most of the math and networking
5242 routines) has been updated from the BSD 4.4-Lite release.
5244 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5246 * The new functions `getdomainname' and `setdomainname' fetch or change the
5247 YP/NIS domain name. These are system calls which exist on systems which
5250 * The time zone data files have been updated for the latest international
5253 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5254 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5258 * The C library now includes support for Sun RPC, from Sun's free
5259 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5260 are included. (There is still no support for YP.)
5262 * Tom Quinn has contributed a port of the C library to SGI machines running
5263 Irix 4 (mips-sgi-irix4).
5265 * The new `lockf' function is a simplified interface to the locking
5266 facilities of `fcntl', included for compatibility.
5268 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5270 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5272 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5275 * The error message printed by `assert' for a failed assertion now includes
5276 the name of the program (if using GNU ld) and the name of the calling
5277 function (with versions of GCC that support this).
5279 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5281 * The library now includes the <sys/mman.h> header file and memory
5282 management functions `mmap', `munmap', `mprotect', `msync', and
5283 `madvise', on systems that support those facilities.
5285 * The interface for `mcheck' has changed slightly: the function called to
5286 abort the program when an allocation inconsistency is detected now takes
5287 an argument that indicates the type of failure. The new function
5288 `mprobe' lets you request a consistency check for a particular block at
5289 any time (checks are normally done only when you call `free' or `realloc'
5292 * It is now possible to easily cross-compile the C library, building on one
5293 system a library to run on another machine and/or operating system. All
5294 you need to do is set the variable `HOST_CC' in `configparms' to the
5295 native compiler for programs to run on the machine you are building on (a
5296 few generator programs are used on Unix systems); set `CC' to the
5299 * The new function `fexecve' (only implemented on the GNU system) executes
5300 a program file given a file descriptor already open on the file.
5304 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5305 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5307 * You can set the variable `libprefix' in `configparms' to specify a prefix
5308 to be prepended to installed library files; this makes it easy to install
5309 the GNU C library to be linked as `-lgnuc' or whatever.
5311 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5312 copies a limited number of characters from a string, and returns the
5313 address of the last character written.
5315 * You no longer need to check for whether the installed `stddef.h' is
5316 compatible with the GNU C library. configure now checks for you.
5318 * You can now define a per-stream `fileno' function to convert the
5319 stream's cookie into an integral file descriptor.
5321 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5322 allocates zero bytes of storage, and returns a unique pointer which
5323 you can pass to `realloc' or `free'. The behavior is undefined if
5324 you dereference this pointer.
5326 * The C library now runs on Sony NEWS m68k machines running either
5327 NewsOS 3 or NewsOS 4.
5329 * The new `syscall' function is a system-dependent primitive function
5330 for invoking system calls. It has the canonical behavior on Unix
5331 systems, including unreliable return values for some calls (such as
5332 `pipe', `fork' and `getppid').
5334 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5335 to `EAGAIN', which is the preferred name. On systems whose kernels
5336 use two distinct codes, the C library now translates EWOULDBLOCK to
5337 EAGAIN in every system call function.
5341 * The GNU C Library Reference Manual is now distributed with the library.
5342 `make dvi' will produce a DVI file of the printed manual.
5343 `make info' will produce Info files that you can read on line using C-h i
5344 in Emacs or the `info' program.
5345 Please send comments on the manual to bug-glibc-manual@gnu.org.
5347 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5349 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5351 * Jason Merrill has contributed a port to the Sequent Symmetry running
5352 Dynix version 3 (i386-sequent-dynix).
5354 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5355 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5357 * New function `memory_warnings' lets you arrange to get warnings when
5358 malloc is running out of memory to allocate, like Emacs gives you.
5360 * The C library now contains the relocating allocator used in Emacs 19 for
5361 its editing buffers. This allocator (ralloc) minimizes allocation
5362 overhead and fragmentation by moving allocated regions around whenever it
5363 needs to. You always refer to a ralloc'd region with a "handle" (a
5364 pointer to a pointer--an object of type `void **').
5366 * There is a new `printf' format: `%m' gives you the string corresponding
5367 to the error code in `errno'.
5369 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5370 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5371 pass, the `a' modifier says to fill in a `char **' you pass with a
5374 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5375 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5376 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5378 * `mkstemp' is a traditional Unix function to atomically create and open a
5379 uniquely-named temporary file.
5383 * The standard location for the file that says what the local timezone is
5384 has changed again. It is now `/usr/local/etc/localtime' (or more
5385 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5387 * The distribution no longer contains any files with names longer than 14
5390 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5391 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5393 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5397 * The configuration process has changed quite a bit. The `configure'
5398 script is now used just like the configuration scripts for other GNU
5399 packages. The `sysdeps' directory hierarchy is much rearranged.
5400 The file `INSTALL' explains the new scheme in detail.
5402 * The header files no longer need to be processed into ANSI C and
5403 traditional C versions. There is just one set of files to install, and
5404 it will work with ANSI or old C compilers (including `gcc -traditional').
5406 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5407 MIPS DECStation running Ultrix 4.
5409 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5410 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5411 made itself into a shared library.
5413 * Yet further improved support for the i386, running 4.3 BSD-like systems
5414 (such as Mach 3 with the Unix single-server), or System V.
5416 * New function `strncasecmp' to do case-insensitive string comparison
5417 with limited length.
5419 * New function `strsep' is a reentrant alternative to `strtok'.
5421 * New functions `scandir' and `alphasort' for searching directories.
5423 * New function `setenv' is a better interface to `putenv'.
5425 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5426 function for traversing a directory tree.
5428 * The GNU obstack package is now also part of the C library.
5429 The new function `open_obstack_stream' creates a stdio stream that
5430 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5431 formatted output directly to an obstack.
5433 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5434 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5436 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5438 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5439 things to your strings.
5441 * There are some new test programs: test-fseek, testmb, and testrand.
5443 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5444 These ports are not finished, but are a good starting place for really
5445 supporting those systems.
5447 * `/etc/localtime' is now the standard location for the file that says what
5448 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5449 This follows the general principle that `/etc' is the place for all local
5450 configuration files.
5452 * The C library header files now use `extern "C"' when used by the C++
5453 compiler, so the C library should now work with C++ code.
5455 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5456 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5459 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5460 function which uses the `merge sort' algorithm, and is said to be
5461 significantly faster than the old GNU `qsort' function. Merge sort is
5462 now the standard `qsort' function. The new algorithm can require a lot
5463 of temporary storage; so, the old sorting function is called when the
5464 required storage is not available.
5466 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5467 provides the Unix `crypt' function, plus some other entry points.
5469 * The code and header files taken from 4.4 BSD have been updated with the
5470 latest files released from Berkeley.
5472 ----------------------------------------------------------------------
5473 Copyright information:
5475 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5477 Permission is granted to anyone to make or distribute verbatim copies
5478 of this document as received, in any medium, provided that the
5479 copyright notice and this permission notice are preserved,
5480 thus giving the recipient permission to redistribute in turn.
5482 Permission is granted to distribute modified versions
5483 of this document, or of portions of it,
5484 under the above conditions, provided also that they
5485 carry prominent notices stating who last changed them.
5488 version-control: never