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 12.1.0 Support: Character encoding, character type info, and
13 transliteration tables are all updated to Unicode 12.1.0, 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 * The twalk_r function has been added. It is similar to the existing
20 twalk function, but it passes an additional caller-supplied argument
21 to the callback function.
23 * On Linux, the gettid and tgkill functions have been added.
25 * Minguo (Republic of China) calendar support has been added as an
26 alternative calendar for the following locales: zh_TW, cmn_TW, hak_TW,
29 * The entry for the new Japanese era has been added for ja_JP locale.
31 * Memory allocation functions malloc, calloc, realloc, reallocarray, valloc,
32 pvalloc, memalign, and posix_memalign fail now with total object size
33 larger than PTRDIFF_MAX. This is to avoid potential undefined behavior with
34 pointer subtraction within the allocated object, where results might
35 overflow the ptrdiff_t type.
37 Deprecated and removed features, and other changes affecting compatibility:
39 * The functions clock_gettime, clock_getres, clock_settime,
40 clock_getcpuclockid, clock_nanosleep were removed from the librt library
41 for new applications (on architectures which had them). Instead, the
42 definitions in libc will be used automatically, which have been available
45 * The obsolete and never-implemented XSI STREAMS header files <stropts.h>
46 and <sys/stropts.h> have been removed.
48 * Support for the "inet6" option in /etc/resolv.conf and the RES_USE_INET6
49 resolver flag (deprecated in glibc 2.25) have been removed.
51 * The obsolete RES_INSECURE1 and RES_INSECURE2 option flags for the DNS stub
52 resolver have been removed from <resolv.h>.
54 * With --enable-bind-now, installed programs are now linked with the
57 * Support for the PowerPC SPE ISA extension (powerpc-*-*gnuspe*
58 configurations) has been removed, following the deprecation of this
59 subarchitecture in version 8 of GCC, and its removal in version 9.
61 Changes to build and runtime requirements:
63 * GCC 6.2 or later is required to build the GNU C Library.
65 Older GCC versions and non-GNU compilers are still supported when
66 compiling programs that use the GNU C Library.
68 Security related changes:
70 CVE-2019-7309: x86-64 memcmp used signed Jcc instructions to check
71 size. For x86-64, memcmp on an object size larger than SSIZE_MAX
72 has undefined behavior. On x32, the size_t argument may be passed
73 in the lower 32 bits of the 64-bit RDX register with non-zero upper
74 32 bits. When it happened with the sign bit of RDX register set,
75 memcmp gave the wrong result since it treated the size argument as
76 zero. Reported by H.J. Lu.
78 CVE-2019-9169: Attempted case-insensitive regular-expression match
79 via proceed_next_node in posix/regexec.c leads to heap-based buffer
80 over-read. Reported by Hongxu Chen.
82 The following bugs are resolved with this release:
84 [The release manager will add the list generated by
85 scripts/list-fixed-bugs.py just before the release.]
92 * The getcpu wrapper function has been added, which returns the currently
93 used CPU and NUMA node. This function is Linux-specific.
95 * A new convenience target has been added for distribution maintainers
96 to build and install all locales as directories with files. The new
97 target is run by issuing the following command in your build tree:
98 'make localedata/install-locale-files', with an optional DESTDIR
99 to set the install root if you wish to install into a non-default
102 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
104 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
105 for _GNU_SOURCE, to match BSD environments.
107 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
108 indicates that it will abort the transaction prior to entering the kernel
109 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
110 suspended, and this caused some undefined side-effects issues by aborting
111 transactions manually. Glibc avoided it by abort transactions manually on
112 each syscall, but it lead to performance issues on newer kernels where the
113 HTM state is saved and restore lazily (the state being saved even when the
114 process actually does not use HTM).
116 * The functions posix_spawn_file_actions_addchdir_np and
117 posix_spawn_file_actions_addfchdir_np have been added, enabling
118 posix_spawn and posix_spawnp to run the new process in a different
119 directory. These functions are GNU extensions. The function
120 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
123 * The popen and system do not run atfork handlers anymore (BZ#17490).
124 Although it is a possible POSIX violation, the POSIX rationale in
125 pthread_atfork documentation regarding atfork handlers is to handle
126 inconsistent mutex state after a fork call in a multi-threaded process.
127 In both popen and system there is no direct access to user-defined mutexes.
129 * Support for the C-SKY ABIV2 running on Linux has been added. This port
130 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
132 - C-SKY ABIV2 soft-float little-endian
133 - C-SKY ABIV2 hard-float little-endian
135 * strftime's default formatting of a locale's alternative year (%Ey)
136 has been changed to zero-pad the year to a minimum of two digits,
137 like "%y". This improves the display of Japanese era years during
138 the first nine years of a new era, and is expected to be harmless
139 for all other locales (only Japanese locales regularly have
140 alternative year numbers less than 10). Zero-padding can be
141 overridden with the '_' or '-' flags (which are GNU extensions).
143 * As a GNU extension, the '_' and '-' flags can now be applied to
144 "%EY" to control how the year number is formatted; they have the
145 same effect that they would on "%Ey".
147 Deprecated and removed features, and other changes affecting compatibility:
149 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
150 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
152 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
153 in <sys/procfs.h>, has been corrected to match the type actually used by
154 the Linux kernel. This affects the size and layout of that structure on
155 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
157 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
158 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
159 defined in <sys/procfs.h>, has been corrected to match the type actually
160 used by the Linux kernel. This affects the size and layout of those
163 * An archaic GNU extension to scanf, under which '%as', '%aS', and '%a[...]'
164 meant to scan a string and allocate space for it with malloc, is now
165 restricted to programs compiled in C89 or C++98 mode with _GNU_SOURCE
166 defined. This extension conflicts with C99's use of '%a' to scan a
167 hexadecimal floating-point number, which is now available to programs
168 compiled as C99 or C++11 or higher, regardless of _GNU_SOURCE.
170 POSIX.1-2008 includes the feature of allocating a buffer for string input
171 with malloc, using the modifier letter 'm' instead. Programs using
172 '%as', '%aS', or '%a[...]' with the old GNU meaning should change to
173 '%ms', '%mS', or '%m[...]' respectively. Programs that wish to use the
174 C99 '%a' no longer need to avoid _GNU_SOURCE.
176 GCC's -Wformat warnings can detect most uses of this extension, as long
177 as all functions that call vscanf, vfscanf, or vsscanf are annotated with
178 __attribute__ ((format (scanf, ...))).
180 Changes to build and runtime requirements:
182 * Python 3.4 or later is required to build the GNU C Library.
184 * On most architectures, GCC 5 or later is required to build the GNU C
185 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
187 Older GCC versions and non-GNU compilers are still supported when
188 compiling programs that use the GNU C Library.
190 Security related changes:
192 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
193 denial of service due to resource exhaustion when processing getaddrinfo
194 calls with crafted host names. Reported by Guido Vranken.
196 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
197 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
198 happened, accessing the 32-bit size_t value as the full 64-bit register
199 in the assembly string/memory functions would cause a buffer overflow.
202 CVE-2016-10739: The getaddrinfo function could successfully parse IPv4
203 addresses with arbitrary trailing characters, potentially leading to data
204 or command injection issues in applications.
206 The following bugs are resolved with this release:
208 [10425] localedata: it_IT/it_CH: LC_TIME format is wrong
209 [10496] localedata: 12h time representation in multiple locales faulty
210 [10797] localedata: it_IT locale numeric does not have a separator for
212 [11319] libc: dprintf doesn't handle errors properly
213 [16346] time: mktime: potentially unsafe use of localtime_offset
214 [17248] build: glibc should not sort CFLAGS (support gcc plugins and
216 [17405] libc: Implement posix_spawn_file_actions_addchdir_np,
217 posix_spawn_file_actions_addfchdir_np
218 [17426] localedata: Indian locales: set the correct date format
219 [17490] stdio: popen should not invoke atfork handlers
220 [17783] libc: TIOCSER_TEMT conditions inconsistent
221 [18040] regex: use-after-free in regexec/get_subexp
222 [18093] libc: Corrupted aux-cache causes ldconfig to segfault
223 [20018] network: getaddrinfo should reject IP addresses with trailing
224 characters (CVE-2016-10739)
225 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
226 [20271] libc: Missing "\n" in __libc_fatal calls
227 [20480] dynamic-link: Patch: ifunc not executable, crashes sudo qemu
228 [20544] libc: RFE: atexit, __cxa_atexit, on_exit should assert function
229 pointer argument is non-NULL
230 [21037] stdio: open_memstream and freopen
231 [21286] libc: bits/siginfo.h is missing enum definition for TRAP_HWBKPT
232 [21716] time: Crash in glibc's mktime in low-memory situations
233 [22834] stdio: Subprocess forked by popen may crash in Linux when
234 multithreads call popen
235 [22927] network: crash in vn_gai_enqueue_request if requests_tail was NULL
236 and pthread_create fails.
237 [23032] hurd: sysdeps/htl/pt-barrier-init.c:39: bad call to memcmp ?
238 [23125] libc: riscv64: endless loop when throwing an exception from a
240 [23275] nptl: Race in pthread_mutex_lock while promoting to
241 PTHREAD_MUTEX_ELISION_NP.
242 [23400] libc: stdlib/test-bz22786.c creates temporary files in glibc
244 [23479] math: [mips] bits/fenv.h should not define some macros for soft-
246 [23490] libc: sysdeps/unix/sysv/linux/x86/tst-cet-property-2.c:49: off by
248 [23497] libc: readdir64@GLIBC_2.1 cannot parse the kernel directory stream
249 [23509] dynamic-link: CET enabled glibc is incompatible with the older
251 [23520] nscd: nscd: Use-after-free in addgetnetgrentX and its callers
252 [23521] nss: get_next_alias nss_files file stream leak
253 [23538] nptl: Hang in pthread_cond_broadcast
254 [23562] libc: Wrong type for si_band in Linux-specific siginfo_t
255 [23578] regex: Invalid memory access if regex pattern contains NUL byte
256 [23579] libc: Errors misreported in preadv2
257 [23597] build: support/test-container.c doesn't work with different
259 [23603] time: mktime signed integer overflow on large timestamps
260 [23606] libc: Missing ENDBR32 in sysdeps/i386/start.S
261 [23614] libc: powerpc: missing CFI register information in __mpn_*
263 [23637] string: Generic strstr/strcasestr fails with huge needles
264 [23640] libc: no way to easily clear FD_CLOEXEC in
265 posix_spawn_file_actions_adddup2()
266 [23649] libc: [microblaze/mips/nios2/riscv] sys/procfs.h pr_uid, pr_gid
268 [23656] libc: [mips n32] sys/procfs.h pr_sigpend, pr_sighold, pr_flag have
270 [23679] libc: gethostid: Missing NULL check for gethostbyname_r result
271 [23689] libc: Bug in documentation for rusage.ru_ixrss in
272 bits/types/struct_rusage.h
273 [23690] dynamic-link: Segfault in _dl_profile_fixup with a high number of
275 [23707] dynamic-link: Missing unwind info in sysdeps/powerpc/powerpc32/dl-
277 [23709] string: glibc 2.25 lacks sse2 optimized strstr()
278 [23716] dynamic-link: _dl_runtime_resolve_shstk isn't selected properly
279 [23717] libc: glibc: stdlib/tst-setcontext9 test suite failure on
281 [23724] localedata: Albanian date formats are incorrect
282 [23735] math: libnldbl_nonshared.a references internal libm symbols
283 [23740] localedata: kl_GL: Month names and date formats need update
284 [23744] regex: regex refactorings to remove BE, avoid duplication
285 [23745] time: mktime fix for Gnulib + coreutils
286 [23758] time: Improve the width of alternate representation for year in
288 [23783] libc: [mips] Missing CMSPAR bits/termios.h
289 [23789] time: mktime does not set errno on failure
290 [23791] localedata: Wrong monetary format for ca_ES locale
291 [23793] locale: c32rtomb and mbrtoc32 should not alias wcrtomb and mbrtowc
292 [23794] locale: c16rtomb does not handle surrogate pairs
293 [23821] libc: si_band in siginfo_t has wrong type long int on sparc64
294 [23822] math: ia64 static libm.a is missing exp2f, log2f and powf symbols
295 [23836] time: time/tst-mktime2 test failure on Arm (32-bit)
296 [23848] libc: [sparc] Some socket syscalls wrongly assumed to be present
297 [23861] nptl: rdlock stalls indefinitely on an unlocked pthread rwlock
298 [23862] libc: [sh] missing kernel-features.h undefines
299 [23864] libc: [riscv] missing kernel-features.h undefines
300 [23867] libc: [arm/microblaze] __ASSUME_MLOCK2 incorrect
301 [23907] malloc: Incorrect double-free malloc tcache check disregards
303 [23913] libc: off-by-one in function maybe_script_execute in
304 sysdeps/posix/spawni.c
305 [23915] libc: [arm] __ASSUME_COPY_FILE_RANGE incorrect
306 [23923] locale: Add --no-hard-links option to localedef
307 [23927] network: Linux if_nametoindex() does not close descriptor
309 [23961] math: powf can overflow to inf without setting errno in non-
310 nearest rounding mode
311 [23967] libc: [2.28 Regression]: New sigaction implementation breaks m68k
312 [23972] libc: __old_getdents64 uses wrong d_off value on overflow
313 [23993] libc: glibc 2.29 doesn't build with gcc 4.9
314 [23995] localedata: Remove execution flags from localedata/locales/bi_VU
315 [24011] localedata: Fixed small type in comment for locale bs_BA
316 [24018] libc: gettext() may return NULL
317 [24022] build: riscv build failure with Linux kernel 4.20-rc7
318 [24023] build: [2.29 Regression] FAIL: elf/check-localplt
319 [24024] string: strerror() might set errno to ENOMEM due to -fno-math-
321 [24027] malloc: glibc: realloc() ncopies 32-bit integer overflow
322 [24034] libc: tst-cancel21-static fails with SIGBUS on pre-ARMv7 when
324 [24046] localedata: en_US locale doesn't define date_fmt
325 [24063] manual: @var{errno} should be @code{errno}
326 [24066] soft-fp: Inconsistent _FP_W_TYPE_SIZE check
327 [24088] libc: VSCR field is not being correctly read in ucontext_t on
329 [24096] time: Specifying '_' or '-' flag for "%EY" does not produce the
331 [24097] string: Can't use 64-bit register for size_t in assembly codes for
333 [24110] hurd: SS_DISABLE never set in stack_t value returned by
335 [24112] network: Do not send DNS queries for non-host names (where all
336 answers will be rejected)
337 [24130] libc: alpha __remqu corrupts $f3 register
344 * The localization data for ISO 14651 is updated to match the 2016
345 Edition 4 release of the standard, this matches data provided by
346 Unicode 9.0.0. This update introduces significant improvements to the
347 collation of Unicode characters. This release deviates slightly from
348 the standard in that the collation element ordering for lowercase and
349 uppercase LATIN script characters is adjusted to ensure that regular
350 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
351 is not matched by [a-z]. With the update many locales have been
352 updated to take advantage of the new collation information. The new
353 collation information has increased the size of the compiled locale
354 archive or binary locales.
356 * The GNU C Library can now be compiled with support for Intel CET, AKA
357 Intel Control-flow Enforcement Technology. When the library is built
358 with --enable-cet, the resulting glibc is protected with indirect
359 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
360 compatible with all existing executables and shared libraries. This
361 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
362 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
363 capable of multi-byte NOPs, like x86-64 processors as well as Intel
364 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
365 x86_64 and x32 on non-CET processors. --enable-cet has been tested
366 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
369 * The GNU C Library now has correct support for ABSOLUTE symbols
370 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
371 relocated incorrectly or in some cases discarded. The GNU linker can
372 make use of the newer semantics, but it must communicate it to the
373 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
374 field) to indicate such support is required.
376 * Unicode 11.0.0 Support: Character encoding, character type info, and
377 transliteration tables are all updated to Unicode 11.0.0, using
378 generator scripts contributed by Mike FABIAN (Red Hat).
380 * <math.h> functions that round their results to a narrower type are added
381 from TS 18661-1:2014 and TS 18661-3:2015:
383 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
386 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
389 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
392 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
395 * Two grammatical forms of month names are now supported for the following
396 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
397 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
398 support two grammatical forms in abbreviated month names: Catalan, Greek,
401 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
402 include the support for two grammatical forms of month names.
404 * Building and running on GNU/Hurd systems now works without out-of-tree
407 * The renameat2 function has been added, a variant of the renameat function
408 which has a flags argument. If the flags are zero, the renameat2 function
409 acts like renameat. If the flag is not zero and there is no kernel
410 support for renameat2, the function will fail with an errno value of
411 EINVAL. This is different from the existing gnulib function renameatu,
412 which performs a plain rename operation in case of a RENAME_NOREPLACE
413 flags and a non-existing destination (and therefore has a race condition
414 that can clobber the destination inadvertently).
416 * The statx function has been added, a variant of the fstatat64
417 function with an additional flags argument. If there is no direct
418 kernel support for statx, glibc provides basic stat support based on
419 the fstatat64 function.
421 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
422 library if installed. libidn2 version 2.0.5 or later is recommended. If
423 libidn2 is not available, internationalized domain names are not encoded
424 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
425 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
426 with an encoding error.) Flags which used to change the IDN encoding and
427 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
428 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
429 deprecated. They no longer have any effect.
431 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
432 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
433 range of ELF gABI expressions including such constructs as
434 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
435 have been further restricted, and where in the past a dynamic string
436 token sequence may have been interpreted as a literal string it will
437 now cause a load failure. These load failures were always considered
438 unspecified behaviour from the perspective of the dynamic loader, and
439 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
440 results in a load failure now.
442 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
443 implementation includes all the standard functions provided by
446 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
447 thrd_detach, thrd_exit, and thrd_join for thread management.
449 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
450 mtx_destroy for mutual exclusion.
452 - call_once for function call synchronization.
454 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
455 cnd_wait for conditional variables.
457 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
459 Application developers must link against libpthread to use ISO C threads.
461 Deprecated and removed features, and other changes affecting compatibility:
463 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
464 installed. Software that was using either header should be updated to
465 use standard <stdio.h> interfaces instead.
467 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
468 This was never required by the C standard, and the macros just expanded
469 to call alternative names for the same functions. If you hoped getc and
470 putc would provide performance improvements over fgetc and fputc, instead
471 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
472 necessary, flockfile and funlockfile.
474 * All stdio functions now treat end-of-file as a sticky condition. If you
475 read from a file until EOF, and then the file is enlarged by another
476 process, you must call clearerr or another function with the same effect
477 (e.g. fseek, rewind) before you can read the additional data. This
478 corrects a longstanding C99 conformance bug. It is most likely to affect
479 programs that use stdio to read interactive input from a terminal.
482 * The macros 'major', 'minor', and 'makedev' are now only available from
483 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
484 headers that happen to include <sys/types.h>. These macros are rarely
485 used, not part of POSIX nor XSI, and their names frequently collide with
486 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
489 <sys/sysmacros.h> is a GNU extension. Portable programs that require
490 these macros should first include <sys/types.h>, and then include
491 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
493 * The tilegx*-*-linux-gnu configurations are no longer supported.
495 * The obsolete function ustat is no longer available to newly linked
496 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
497 function has been deprecated in favor of fstatfs and statfs.
499 * The obsolete function nfsservctl is no longer available to newly linked
500 binaries. This function was specific to systems using the Linux kernel
501 and could not usefully be used with the GNU C Library on systems with
502 version 3.1 or later of the Linux kernel.
504 * The obsolete function name llseek is no longer available to newly linked
505 binaries. This function was specific to systems using the Linux kernel
506 and was not declared in a header. Programs should use the lseek64 name
507 for this function instead.
509 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
510 getaddrinfo and getnameinfo functions have been deprecated. The behavior
511 previously selected by them is now always enabled.
513 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
514 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
515 restriction (rejecting '_' in host names, among other things) has been
516 removed, for increased compatibility with non-IDN name resolution.
518 * The fcntl function now have a Long File Support variant named fcntl64. It
519 is added to fix some Linux Open File Description (OFD) locks usage on non
520 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
521 fcntl and LFS support is handled transparently. Also for Linux, the OFD
522 locks act as a cancellation entrypoint.
524 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
525 ecb_crypt, and des_setparity are no longer available to newly linked
526 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
527 longer installed. These functions encrypted and decrypted data with the
528 DES block cipher, which is no longer considered secure. Software that
529 still uses these functions should switch to a modern cryptography library,
532 * Reflecting the removal of the encrypt and setkey functions above, the
533 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
534 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
537 * The obsolete function fcrypt is no longer available to newly linked
538 binaries. It was just another name for the standard function crypt,
539 and it has not appeared in any header file in many years.
541 * We have tentative plans to hand off maintenance of the passphrase-hashing
542 library, libcrypt, to a separate development project that will, we hope,
543 keep up better with new passphrase-hashing algorithms. We will continue
544 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
545 'crypt_r' should not need to change at all; however, distributions will
546 need to install <crypt.h> and libcrypt from a separate project.
548 In this release, if the configure option --disable-crypt is used, glibc
549 will not install <crypt.h> or libcrypt, making room for the separate
550 project's versions of these files. The plan is to make this the default
551 behavior in a future release.
553 Changes to build and runtime requirements:
555 GNU make 4.0 or later is now required to build glibc.
557 Security related changes:
559 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
560 been fixed by removing the glibc-internal IDNA implementation and using
561 the system-provided libidn2 library instead. Originally reported by Hanno
562 Böck and Christian Weisgerber.
564 CVE-2017-18269: An SSE2-based memmove implementation for the i386
565 architecture could corrupt memory. Reported by Max Horn.
567 CVE-2018-11236: Very long pathname arguments to realpath function could
568 result in an integer overflow and buffer overflow. Reported by Alexey
571 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
572 architecture could write beyond the target buffer, resulting in a buffer
573 overflow. Reported by Andreas Schwab.
575 The following bugs are resolved with this release:
577 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
578 [6889] manual: 'PWD' mentioned but not specified
579 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
580 when __WORDSIZE != 64
581 [13762] regex: re_search etc. should return -2 on memory exhaustion
582 [13888] build: /tmp usage during testing
583 [13932] math: dbl-64 pow unexpectedly slow for some inputs
584 [14092] nptl: Support C11 threads
585 [14095] localedata: Review / update collation data from Unicode / ISO
587 [14508] libc: -Wformat warnings
588 [14553] libc: Namespace pollution loff_t in sys/types.h
589 [14890] libc: Make NT_PRFPREG canonical.
590 [15105] libc: Extra PLT references with -Os
591 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
593 [16335] manual: Feature test macro documentation incomplete and out of
595 [16552] libc: Unify umount implementations in terms of umount2
596 [17082] libc: htons et al.: statement-expressions prevent use on global
597 scope with -O1 and higher
598 [17343] libc: Signed integer overflow in /stdlib/random_r.c
599 [17438] localedata: pt_BR: wrong d_fmt delimiter
600 [17662] libc: please implement binding for the new renameat2 syscall
601 [17721] libc: __restrict defined as /* Ignore */ even in c11
602 [17979] libc: inconsistency between uchar.h and stdint.h
603 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
604 [18023] libc: extend_alloca is broken (questionable pointer comparison,
605 horrible machine code)
606 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
608 [18471] libc: llseek should be a compat symbol
609 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
611 [18991] nss: nss_files skips large entry in database
612 [19239] libc: Including stdlib.h ends up with macros major and minor being
614 [19463] libc: linknamespace failures when compiled with -Os
615 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
616 [19527] locale: Normalized charset name not recognized by setlocale
617 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
618 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
620 [19728] network: out of bounds stack read in libidn function
621 idna_to_ascii_4i (CVE-2016-6261)
622 [19729] network: out of bounds heap read on invalid utf-8 inputs in
623 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
624 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
626 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
627 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
628 [20419] dynamic-link: files with large allocated notes crash in
630 [20530] libc: bswap_16 should use __builtin_bswap16() when available
631 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
632 [20980] manual: CFLAGS environment variable replaces vital options
633 [21163] regex: Assertion failure in pop_fail_stack when executing a
634 malformed regexp (CVE-2015-8985)
635 [21234] manual: use of CFLAGS makes glibc detect no optimization
636 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
637 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
638 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
639 [21508] locale: intl/tst-gettext failure with latest msgfmt
640 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
641 [21812] network: getifaddrs() returns entries with ifa_name == NULL
642 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
644 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
646 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
647 [22247] network: Integer overflow in the decode_digit function in
648 puny_decode.c in libidn (CVE-2017-14062)
649 [22342] nscd: NSCD not properly caching netgroup
650 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
651 [22550] localedata: es_ES locale (and other es_* locales): collation
652 should treat ñ as a primary different character, sync the collation
653 for Spanish with CLDR
654 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
655 by gcc configured with --enable-default-pie
656 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
657 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
658 crossing 2GB threshold (CVE-2017-18269)
659 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
661 [22735] time: Misleading typo in time.h source comment regarding
663 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
664 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
665 `__assert_perror_fail ()' to `__assert_fail_base ()'
666 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
668 [22786] libc: Stack buffer overflow in realpath() if input size is close
669 to SSIZE_MAX (CVE-2018-11236)
670 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
671 through an absolute DT_NEEDED path
672 [22792] build: tcb-offsets.h dependency dropped
673 [22797] libc: pkey_get() uses non-reserved name of argument
674 [22807] libc: PTRACE_* constants missing for powerpc
675 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
676 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
677 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
678 [22848] localedata: ca_ES: update date definitions from CLDR
679 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
680 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
681 [22896] localedata: Update locale data for an_ES
682 [22902] math: float128 test failures with GCC 8
683 [22918] libc: multiple common of `__nss_shadow_database'
684 [22919] libc: sparc32: backtrace yields infinite backtrace with
686 [22926] libc: FTBFS on powerpcspe
687 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
689 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
690 [22947] libc: FAIL: misc/tst-preadvwritev2
691 [22963] localedata: cs_CZ: Add alternative month names
692 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
693 [22996] localedata: change LC_PAPER to en_US in es_BO locale
694 [22998] dynamic-link: execstack tests are disabled when SELinux is
696 [23005] network: Crash in __res_context_send after memory allocation
698 [23007] math: strtod cannot handle -nan
699 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
700 [23036] regex: regex equivalence class regression
701 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
702 [23069] libc: sigaction broken on riscv64-linux-gnu
703 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
704 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
705 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
706 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
707 and libc build with -Os)
708 [23140] localedata: More languages need two forms of month names
709 [23145] libc: _init/_fini aren't marked as hidden
710 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
711 [23171] math: C++ iseqsig for long double converts arguments to double
712 [23178] nscd: sudo will fail when it is run in concurrent with commands
713 that changes /etc/passwd
714 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
716 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
717 [23208] localedata: New locale - Lower Sorbian (dsb)
718 [23233] regex: Memory leak in build_charclass_op function in file
720 [23236] stdio: Harden function pointers in _IO_str_fields
721 [23250] nptl: Offset of __private_ss differs from GCC
722 [23253] math: tgamma test suite failures on i686 with -march=x86-64
723 -mtune=generic -mfpmath=sse
724 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
726 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
727 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
729 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
730 [23277] math: nan function should not have const attribute
731 [23279] math: scanf and strtod wrong for some hex floating-point
732 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
734 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
735 [23303] build: undefined reference to symbol
736 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
737 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
739 [23313] stdio: libio vtables validation and standard file object
741 [23329] libc: The __libc_freeres infrastructure is not properly run across
743 [23349] libc: Various glibc headers no longer compatible with
745 [23351] malloc: Remove unused code related to heap dumps and malloc
747 [23363] stdio: stdio-common/tst-printf.c has non-free license
748 [23396] regex: Regex equivalence regression in single-byte locales
749 [23422] localedata: oc_FR: More updates of locale data
750 [23442] build: New warning with GCC 8
751 [23448] libc: Out of bounds access in IBM-1390 converter
752 [23456] libc: Wrong index_cpu_LZCNT
753 [23458] build: tst-get-cpu-features-static isn't added to tests
754 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
756 [23467] dynamic-link: x86/CET: A property note parser bug
763 * The GNU C Library can now be compiled with support for building static
764 PIE executables (See --enable-static-pie in INSTALL). These static PIE
765 executables are like static executables but can be loaded at any address
766 and provide additional security hardening benefits at the cost of some
767 memory and performance. When the library is built with --enable-static-pie
768 the resulting libc.a is usable with GCC 8 and above to create static PIE
769 executables using the GCC option '-static-pie'. This feature is currently
770 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
771 aarch64 with binutils 2.30 or later.
773 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
774 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
777 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
779 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
781 * In order to support faster and safer process termination the malloc API
782 family of functions will no longer print a failure address and stack
783 backtrace after detecting heap corruption. The goal is to minimize the
784 amount of work done after corruption is detected and to avoid potential
785 security issues in continued process execution. Reducing shutdown time
786 leads to lower overall process restart latency, so there is benefit both
787 from a security and performance perspective.
789 * The abort function terminates the process immediately, without flushing
790 stdio streams. Previous glibc versions used to flush streams, resulting
791 in deadlocks and further data corruption. This change also affects
792 process aborts as the result of assertion failures.
794 * On platforms where long double has the IEEE binary128 format (aarch64,
795 alpha, mips64, riscv, s390 and sparc), the math library now implements
796 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
797 These are the same interfaces added in version 2.26 for some platforms where
798 this format is supported but is not the format of long double.
800 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
801 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
802 implements interfaces for that type, as defined by ISO/IEC TS
803 18661-3:2015. These are corresponding interfaces to those supported for
806 * The math library now implements interfaces for the _Float32, _Float64 and
807 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
808 corresponding interfaces to those supported for _Float128.
810 * glibc now implements the memfd_create and mlock2 functions on Linux.
812 * Support for memory protection keys was added. The <sys/mman.h> header now
813 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
816 * The copy_file_range function was added.
818 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
820 * The ldconfig utility now processes `include' directives using the C/POSIX
821 collation ordering. Previous glibc versions used locale-specific
822 ordering, the change might break systems that relied on that.
824 * Support for two grammatical forms of month names has been added.
825 In a call to strftime, the "%B" and "%b" format specifiers will now
826 produce the grammatical form required when the month is used as part
827 of a complete date. New "%OB" and "%Ob" specifiers produce the form
828 required when the month is named by itself. For instance, in Greek
829 and in many Slavic and Baltic languages, "%B" will produce the month
830 in genitive case, and "%OB" will produce the month in nominative case.
832 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
833 are all valid and will all accept any known form of month
834 name---standalone or complete, abbreviated or full. In a call to
835 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
836 the strings used by "%B" and "%b", respectively. New query
837 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
838 used by "%OB" and "%Ob", respectively.
840 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
841 define the strings for %OB and %Ob, respectively; these have the
842 same syntax as "mon" and "abmon". These arrays are optional; if they
843 are not provided then they have the same content as "mon" and "abmon",
846 These features are provided for locales which define "alt_mon" and/or
847 "ab_alt_mon" in their locale source data. This release includes such
848 alternative month name data for the following languages: Belarusian,
849 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
851 This feature is currently a GNU extension, but it is expected to
852 be added to the next revision of POSIX, and it is also already
853 available on some BSD-derived operating systems.
855 This feature will cause existing statically compiled applications
856 to fail to load locales and fall back to the builtin C/POSIX locales.
857 See notes below for other changes affecting compatibility.
859 * Support for the RISC-V ISA running on Linux has been added. This port
860 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
861 for the following ISA and ABI pairs:
867 Deprecated and removed features, and other changes affecting compatibility:
869 * Statically compiled applications attempting to load locales compiled for the
870 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
871 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
872 support for two grammatical forms of the month names, also extends the locale
873 data binary format. Static applications needing locale support must be
874 recompiled to match the runtime and data they are deployed with. In some
875 distributions there is an upgrade window where dynamically linked applications
876 may use a new library but the old locale data and also fall back to the
877 builtin C/POSIX locales; restarting the application process is sufficient to
880 * Support for statically linked applications which call dlopen is deprecated
881 and will be removed in a future version of glibc. Applications which call
882 dlopen need to be linked dynamically instead.
884 * Support for old programs which use internal stdio data structures and
885 functions is deprecated. This includes programs which use the C++ streams
886 provided by libstdc++ in GCC 2.95. Programs which use the internal
887 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
888 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
889 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
890 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
891 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
892 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
893 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
894 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
895 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
896 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
897 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
898 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
899 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
900 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
901 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
902 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
903 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
904 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
905 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
906 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
907 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
908 future version of glibc. Unlike other symbol removals, these old
909 applications will not be supported using compatibility symbols.
911 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
912 defined by <sys/ptrace.h>.
914 * libm no longer supports SVID error handling (calling a user-provided
915 matherr function on error) or the _LIB_VERSION variable to control error
916 handling. (SVID error handling and the _LIB_VERSION variable still work
917 for binaries linked against older versions of the GNU C Library.) The
918 libieee.a library is no longer provided. math.h no longer defines struct
919 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
920 TLOSS, PLOSS and HUGE.
922 * The libm functions pow10, pow10f and pow10l are no longer supported for
923 new programs. Programs should use the standard names exp10, exp10f and
924 exp10l for these functions instead.
926 * The mcontext_t type is no longer the same as struct sigcontext. On
927 platforms where it was previously the same, this changes the C++ name
928 mangling for interfaces involving this type.
930 * The add-ons mechanism for building additional packages at the same time as
931 glibc has been removed. The --enable-add-ons configure option is now
934 * The --without-fp configure option is now ignored. Whether hardware
935 floating-point instructions are used is now configured based on whether
936 the compiler used at configure time (without any options implied by a
937 --with-cpu= configure option) uses such instructions.
939 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
940 the specified string can be parsed as a domain name.
942 * In the malloc_info output, the <heap> element may contain another <aspace>
943 element, "subheaps", which contains the number of sub-heaps.
945 * The libresolv function p_secstodate is no longer supported for new
948 * The tilepro-*-linux-gnu configuration is no longer supported.
950 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
951 and will be removed in a future release. Software that is still using
952 either header should be updated to use standard <stdio.h> interfaces
955 libio.h was originally the header for a set of supported GNU extensions,
956 but they have not been maintained as such in many years, they are now
957 standing in the way of improvements to stdio, and we don't think there are
958 any remaining external users. _G_config.h was never intended for public
959 use, but predates the bits convention.
961 Changes to build and runtime requirements:
963 * bison version 2.7 or later is required to generate code in the 'intl'
966 Security related changes:
968 CVE-2009-5064: The ldd script would sometimes run the program under
969 examination directly, without preventing code execution through the
970 dynamic linker. (The glibc project disputes that this is a security
971 vulnerability; only trusted binaries must be examined using the ldd
974 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
975 suffered from a one-byte overflow during ~ operator processing (either
976 on the stack or the heap, depending on the length of the user name).
977 Reported by Tim Rühsen.
979 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
980 would sometimes fail to free memory allocated during ~ operator
981 processing, leading to a memory leak and, potentially, to a denial
984 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
985 without GLOB_NOESCAPE, could write past the end of a buffer while
986 unescaping user names. Reported by Tim Rühsen.
988 CVE-2017-17426: The malloc function, when called with an object size near
989 the value SIZE_MAX, would return a pointer to a buffer which is too small,
990 instead of NULL. This was a regression introduced with the new malloc
991 thread cache in glibc 2.26. Reported by Iain Buclaw.
993 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
994 to the allocation of too much memory. (This is not a security bug per se,
995 it is mentioned here only because of the CVE assignment.) Reported by
998 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
999 of the number of search path components. (This is not a security
1000 vulnerability per se because no trust boundary is crossed if the fix for
1001 CVE-2017-1000366 has been applied, but it is mentioned here only because
1002 of the CVE assignment.) Reported by Qualys.
1004 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
1005 for AT_SECURE or SUID binaries could be used to load libraries from the
1008 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
1009 succeeds without returning an absolute path due to unexpected behaviour
1010 of the Linux kernel getcwd syscall. Reported by halfdog.
1012 CVE-2018-6485: The posix_memalign and memalign functions, when called with
1013 an object size near the value of SIZE_MAX, would return a pointer to a
1014 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
1016 CVE-2018-6551: The malloc function, when called with an object size near
1017 the value of SIZE_MAX, would return a pointer to a buffer which is too
1018 small, instead of NULL.
1020 The following bugs are resolved with this release:
1022 [866] glob: glob should match dangling symlinks
1023 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
1024 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
1026 [5997] math: Very slow execution of sinf function
1027 [10580] localedata: hr_HR: updated locale
1028 [10871] locale: 'mon' array should contain both nominative and genitive
1030 [12349] localedata: eu_ES: incorrect thousands separator
1031 [13605] localedata: shn_MM: new Shan locale
1032 [13805] localedata: ru_RU: currency should use ',' as radix point
1033 [13953] localedata: km_KH: locale update
1034 [13994] localedata: mjw_IN: new locale
1035 [14121] build: make writes .mo files in po directory
1036 [14333] libc: Fix the race between atexit() and exit()
1037 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
1038 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
1040 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
1041 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
1042 width Latin characters
1043 [15332] localedata: es_CU: locale update
1044 [15436] stdio: Don't close or flush stdio streams on abort
1045 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
1047 [16148] localedata: ca_ES: incorrect thousands separator
1048 [16750] dynamic-link: ldd should not try to execute the binaries
1050 [16777] localedata: pl_PL: incorrect thousands separator in locale
1051 [16905] localedata: hanzi: new collation
1052 [17563] localedata: cmn_TW: add hanzi collation
1053 [17750] localedata: wrong collation order of diacritics in most locales
1054 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
1056 [17956] build: Build fails on missing definitions from header file
1057 nss/nss.h when Mozilla NSS is used for cryptography
1058 [18203] libc: realpath() does not handle unreachable paths correctly
1059 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
1060 [18812] localedata: kab_DZ: new Kabyle Algeria locale
1061 [18822] libc: Internal functions are called via PLT
1062 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
1063 [19170] libc: __gmon_start__ defined in hppa in crtn.S
1064 [19574] libc: glibc should support building static PIE binaries
1065 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
1067 [19971] glob: glob: Do not skip entries with zero d_ino values
1068 [19982] localedata: fr.po: spelling mistake for error code EXDEV
1069 [20008] localedata: km_KH: convert to translit_neutral
1070 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
1071 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
1072 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
1073 [20482] localedata: de_CH: abbreviated weekdays should be two letters
1074 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
1076 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
1077 they're set, wrong errors returned by gaih_inet when lookup functions
1079 [20756] localedata: [PATCH] Use Unicode wise thousands separator
1080 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
1082 [20952] localedata: yuw_PG: new locale
1083 [21084] localedata: charmaps/IBM858: new codepage
1084 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
1085 [21242] libc: assert gives pedantic warning in old gcc versions
1086 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
1087 __regcall calling convention
1088 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
1089 [21326] libc: C99 functions are not declared for C++11 and later if
1090 _GNU_SOURCE is not predefined by g++
1091 [21457] libc: sys/ucontext.h namespace
1092 [21530] libc: tmpfile() should be implemented using O_TMPFILE
1093 [21660] math: GCC fails to compile a formula with tgmath.h
1094 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
1095 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
1096 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
1097 [21684] math: tgmath.h handling of complex integers
1098 [21685] math: tgmath.h handling of bit-fields
1099 [21686] math: tgmath.h handling of __int128
1100 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1102 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
1104 [21750] localedata: column width of characters incompatible with classical
1106 [21754] malloc: malloc: Perform as little work as possible after heap
1107 consistency check failures
1108 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
1109 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
1110 [21791] string: Unused XXX_chk_XXX functions in libc.a
1111 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
1113 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
1115 [21845] localedata: Added new Locale bho_NP
1116 [21853] localedata: Fix abday Which looks same as day in zh_SG
1117 [21854] localedata: Added New Locale en_SC
1118 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
1119 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
1120 _dl_runtime_resolve_avx_slow
1121 [21885] network: getaddrinfo: gethosts does not release resolver context
1122 on memory allocation failure
1123 [21899] libc: XPG4.2 sigaction namespace
1124 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
1126 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
1128 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
1129 errno value, causing getaddrinfo to fail
1130 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
1131 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
1132 instead of EAI_NODATA
1133 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
1134 constant PTRACE_SEIZE_DEVEL
1135 [21930] math: C-only gcc builtins used in <math.h> isinf
1136 [21932] network: Unpaired __resolv_context_get in generic get*_r
1138 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
1139 [21944] libc: sigval namespace
1140 [21951] localedata: Update hanzi collation by stroke
1141 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
1142 sysdeps/x86_64/fpu/e_expf.S
1143 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
1144 allocation in syscall loops)
1145 [21959] localedata: Fix Country name for xh_ZA
1146 [21960] localedata: Fix abmon for bem_ZM
1147 [21966] math: AVX2 mathvec functions use FMA without checking
1148 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
1149 [21971] localedata: Added New Locale for mfe_MU
1150 [21972] libc: assert macro requires operator== (int) for its argument type
1151 [21973] math: [sparc] libm missing sqrtl compat symbol
1152 [21974] libc: Remove __bb_init_func and __bb_exit_func
1153 [21982] string: stratcliff.c: error: assuming signed overflow does not
1155 [21986] stdio: __guess_grouping is called incorrectly
1156 [21987] math: [sparc32] wrong bits/long-double.h installed
1157 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
1159 [22022] localedata: Missing country_name for mni_IN
1160 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
1162 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
1164 [22026] locale: iconv_open: heap overflow on gconv_init failure
1165 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
1166 [22035] math: [m68k] bits/math-inline.h macro namespace
1167 [22038] localedata: Fix abbreviated weeks and months for Somali
1168 [22044] localedata: Remove redundant data for Limburgish Language
1169 [22050] malloc: Linking with -lmcheck does not hook
1170 __malloc_initialize_hook correctly
1171 [22051] libc: zero terminator in the middle of glibc's .eh_frame
1172 [22052] malloc: malloc failed to compile with GCC 7 and -O3
1173 [22070] localedata: charmaps/UTF-8: wcwidth for
1174 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
1175 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
1176 Jungseong and Jongseong) should be 0
1177 [22078] nss: nss_files performance issue in multi mode
1178 [22082] math: bits/math-finite.h exp10 condition
1179 [22086] libc: pcprofiledump incorrect cross-endian condition
1180 [22093] dynamic-link: ld.so no longer searches in .../x86_64
1181 [22095] network: Name server address allocation memory leak in resolv.conf
1183 [22096] network: __resolv_conf_attach can incorrectly free passed conf
1185 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
1186 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
1188 [22111] malloc: malloc: per thread cache is not returned when thread exits
1189 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
1190 [22134] libc: [linux] implement fexecve with execveat
1191 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
1193 [22145] libc: ttyname() gives up too early in the face of namespaces
1194 [22146] math: C++ build issue with float128 on x86_64
1195 [22153] nptl: nptl: save error code before process termination
1196 [22156] libc: [hppa,ia64,microblaze] Executable stack default
1197 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
1198 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
1199 [22165] libc: [hppa] Text relocations in libc.so
1200 [22180] libc: destructor registered via __cxa_atexit is called twice
1201 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
1202 symlinks") cause "make" segfaults
1203 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
1205 [22207] libc: FAIL: stdlib/test-atexit-race
1206 [22225] math: nearbyint arithmetic moved before feholdexcept
1207 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
1208 [22235] math: iscanonical in C++ and float128
1209 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
1210 the svid compat wrapper
1211 [22244] math: ynf and yn are wrong without the svid compat wrapper
1212 [22273] libc: Improper assert in Linux posix_spawn implementation
1213 [22284] libc: -pg -pie doesn't work
1214 [22292] locale: localedef exits with error 4 when it should be error 1
1215 [22294] locale: Allow "" for int_currency_symbol definition in locales.
1216 [22295] locale: Don't warn on non-symbolic characters in locale sources in
1218 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
1219 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
1221 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
1222 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
1223 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
1224 [22322] libc: [mips64] wrong bits/long-double.h installed
1225 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
1226 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
1228 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
1229 [22343] malloc: Integer overflow in posix_memalign
1230 [22347] libc: getrandom() returns the number of bytes that were copied to
1231 the buffer even though the comments say "Return 0 on success and -1 on
1233 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1234 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1236 [22370] dynamic-link: Incorrect note padding check
1237 [22375] libc: malloc returns pointer from tcache_get when should return
1238 NULL (CVE-2017-17426)
1239 [22377] math: iseqsig, float128 and C++
1240 [22382] localedata: Error in tpi_PG locale
1241 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1242 inside the ASCII printable range
1243 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1245 [22403] localedata: Slash needs escaping in some locales
1246 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1247 [22409] network: res_hnok does not accept some host names used on the
1249 [22412] network: res_dnok, res_hnok should perform syntax checks
1250 [22413] network: ns_name_pton ignores syntactically invalid trailing
1252 [22415] stdio: setvbuf can lead to invalid free/segfault
1253 [22432] build: Non-deterministic build
1254 [22439] malloc: malloc_info should compute summary statistics for all sub-
1256 [22442] network: if_nametoindex could report index for the wrong
1257 networking interface
1258 [22446] build: aliasing violation calling readlink in handle_request
1259 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1260 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1261 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1262 __nscd_hash/__nss_hash
1263 [22463] network: p_secstodate overflow handling
1264 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1265 [22478] libc: sigwait can fail with EINTR
1266 [22505] libc: ldconfig processes include directive in locale-specific
1268 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1269 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1270 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1271 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1272 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1273 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1274 same as for Croatian
1275 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1276 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1277 [22577] libc: missing newline after "cannot allocate TLS data structures
1279 [22588] manual: manual/conf.texi: missing underscore in front of
1281 [22593] math: nextafter and nexttoward are declared with const attribute
1282 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1283 [22603] string: ia64 memchr overflows internal pointer check
1284 [22605] libc: SH clone does not set the exit code correctly
1285 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1287 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1288 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1289 modified in case of success
1290 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1291 [22615] manual: manual: ambiguous wording about errno value in case of
1293 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1295 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1296 binaries or if /proc is not mounted (CVE-2017-16997)
1297 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1298 [22630] build: $(no-pie-ldflag) is no longer effective
1299 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1300 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1301 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1302 [22637] nptl: guard size is subtracted from thread stack size instead of
1304 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1305 [22657] localedata: hu_HU: Avoid double space in date
1306 [22660] math: fmax, fmin sNaN handling on alpha
1307 [22664] libc: New warning of GCC8
1308 [22665] math: alpha: ceil and floor raise inexact exceptions
1309 [22666] math: alpha: trunc raise inexact exceptions
1310 [22667] libc: makecontext lacks stack alignment on i386
1311 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1312 [22679] libc: getcwd(3) can succeed without returning an absolute path
1314 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1315 elision and tunables
1316 [22687] math: [powerpc-nofpu] complex long double functions spurious
1318 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1319 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1320 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1322 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1323 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1325 [22701] nis: Incomplete removal of libnsl
1326 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1327 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1328 [22715] dynamic-link: FAIL: elf/tst-audit10
1329 [22719] libc: Backtrace tests fail on hppa
1330 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1331 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1332 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1333 before the first call to crypt_r () in crypt/badsalttest.c
1340 * A per-thread cache has been added to malloc. Access to the cache requires
1341 no locks and therefore significantly accelerates the fast path to allocate
1342 and free small amounts of memory. Refilling an empty cache requires locking
1343 the underlying arena. Performance measurements show significant gains in a
1344 wide variety of user workloads. Workloads were captured using a special
1345 instrumented malloc and analyzed with a malloc simulator. Contributed by
1346 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1348 * Unicode 10.0.0 Support: Character encoding, character type info, and
1349 transliteration tables are all updated to Unicode 10.0.0, using
1350 generator scripts contributed by Mike FABIAN (Red Hat).
1351 These updates cause user visible changes, especially the changes in
1352 wcwidth for many emoji characters cause problems when emoji sequences
1353 are rendered with pango, see for example:
1354 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1356 * Collation of Hungarian has been overhauled and is now consistent with "The
1357 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1360 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1362 - The GNU C Library will now detect when /etc/resolv.conf has been
1363 modified and reload the changed configuration. The new resolver option
1364 “no-reload” (RES_NORELOAD) disables this behavior.
1366 - The GNU C Library now supports an arbitrary number of search domains
1367 (configured using the “search” directive in /etc/resolv.conf);
1368 previously, there was a hard limit of six domains. For backward
1369 compatibility, applications that directly modify the ‘_res’ global
1370 object are still limited to six search domains.
1372 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1373 Library will now randomly pick a name server from the configuration as a
1374 starting point. (Previously, the second name server was always used.)
1376 * The tunables feature is now enabled by default. This allows users to tweak
1377 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1379 * New function reallocarray, which resizes an allocated block (like realloc)
1380 to the product of two sizes, with a guaranteed clean failure upon integer
1381 overflow in the multiplication. Originally from OpenBSD, contributed by
1382 Dennis Wölfing and Rüdiger Sonderfeld.
1384 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1385 These are extended versions of preadv and pwritev, respectively, taking an
1386 additional flags argument. The set of supported flags depends on the
1387 running kernel; full support currently requires kernel 4.7 or later.
1389 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1390 create a new session ID for the spawned process. This feature is
1391 scheduled to be added to the next major revision of POSIX; for the time
1392 being, it is available under _GNU_SOURCE.
1394 * errno.h is now safe to use from C-preprocessed assembly language on all
1395 supported operating systems. In this context, it will only define the
1396 Exxxx constants, as preprocessor macros expanding to integer literals.
1398 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1399 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1400 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1401 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1403 To compile programs that use this feature, the compiler must support
1404 128-bit floating point with the type name _Float128 (as defined by TS
1405 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1406 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1407 must be defined to make the new interfaces visible.
1409 The new functions and macros correspond to those present for other
1410 floating-point types (except for a few obsolescent interfaces not
1411 supported for the new type), with F128 or f128 suffixes; for example,
1412 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1413 printf or scanf formats for the new type; the strfromf128 and strtof128
1414 interfaces should be used instead.
1416 Deprecated and removed features, and other changes affecting compatibility:
1418 * The synchronization that pthread_spin_unlock performs has been changed to
1419 now be equivalent to a C11 atomic store with release memory order to the
1420 spin lock's memory location. Previously, several (but not all)
1421 architectures used stronger synchronization (e.g., containing what is
1422 often called a full barrier). This change can improve performance, but
1423 may affect odd fringe uses of spin locks that depend on the previous
1424 behavior (e.g., using spin locks as atomic variables to try to implement
1425 Dekker's mutual exclusion algorithm).
1427 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1430 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1431 will only be built and installed when the GNU C Library is configured with
1432 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1433 as TIRPC or rpcsvc-proto, to be used.
1435 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1436 libnss_compat, are deprecated, and will not be built or installed by
1439 The NIS(+) support library, libnsl, is also deprecated. By default, a
1440 compatibility shared library will be built and installed, but not headers
1441 or development libraries. Only a few NIS-related programs require this
1442 library. (In particular, the GNU C Library has never required programs
1443 that use 'gethostbyname' to be linked with libnsl.)
1445 Replacement implementations based on TIRPC, which additionally support
1446 IPv6, are available from <https://github.com/thkukuk/>. The configure
1447 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1448 name service modules, to be built and installed.
1450 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1451 support is enabled, the configured recursive resolver must support EDNS.
1452 (Responding to EDNS-enabled queries with responses which are not
1453 EDNS-enabled is fine, but FORMERR responses are not.)
1455 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1456 servers have not supported this opcode for a long time.
1458 * The _res_opcodes variable has been removed from libresolv. It had been
1459 exported by accident.
1461 * <string.h> no longer includes inline versions of any string functions,
1462 as this kind of optimization is better done by the compiler. The macros
1463 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1465 * The nonstandard header <xlocale.h> has been removed. Most programs should
1466 use <locale.h> instead. If you have a specific need for the definition of
1467 locale_t with no other declarations, please contact
1468 libc-alpha@sourceware.org and explain.
1470 * The obsolete header <sys/ultrasound.h> has been removed.
1472 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1474 * The obsolete function cfree has been removed. Applications should use
1477 * The stack_t type no longer has the name struct sigaltstack. This changes
1478 the C++ name mangling for interfaces involving this type.
1480 * The ucontext_t type no longer has the name struct ucontext. This changes
1481 the C++ name mangling for interfaces involving this type.
1483 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1484 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1485 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1486 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1487 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1488 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1489 fpregset_t type no longer has the name struct fpu. This changes the C++
1490 name mangling for interfaces involving those types.
1492 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1493 synced with the kernel:
1495 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1496 are not supported on this architecture and have been removed.
1498 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1499 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1500 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1502 Programs that assume the GET/SETREGS ptrace requests are universally
1503 available will now fail to build, instead of malfunctioning at runtime.
1505 Changes to build and runtime requirements:
1507 * Linux kernel 3.2 or later is required at runtime, on all architectures
1508 supported by that kernel. (This is a change from version 2.25 only for
1511 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1513 * On most architectures, GCC 4.9 or later is required to build the GNU C
1514 Library. On powerpc64le, GCC 6.2 or later is required.
1516 Older GCC versions and non-GNU compilers are still supported when
1517 compiling programs that use the GNU C Library. (We do not know exactly
1518 how old, and some GNU extensions to C may be _de facto_ required. If you
1519 are interested in helping us make this statement less vague, please
1520 contact libc-alpha@sourceware.org.)
1522 Security related changes:
1524 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1525 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1527 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1528 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1530 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1531 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1534 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1535 fixed (CVE-2017-12133).
1537 The following bugs are resolved with this release:
1539 [984] network: Respond to changed resolv.conf in gethostbyname
1540 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1541 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1543 [12189] libc: __stack_chk_fail should not attempt a backtrace
1545 [14096] time: Race condition on timezone/tst-timezone.out
1546 [14172] localedata: az_IR: new locale
1547 [14995] build: glibc fails to build if gold is the default linker, even if
1549 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1550 [16637] network: inet_pton function is accepting IPv6 with bad format
1551 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1552 x86 and other generic code
1553 [16875] localedata: ko_KR: fix lang_name
1554 [17225] localedata: ar_SY: localized month names for May and June are
1556 [17297] localedata: da_DK: wrong date_fmt string
1557 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1558 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1559 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1560 [19066] localedata: ar_SA abbreviated day and month names are in English
1561 [19569] network: resolv: Support an arbitrary number of search domains
1562 [19570] network: Implement random DNS server selection in the stub
1564 [19838] locale: localedef fails on PA-RISC
1565 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1566 order of 0D36 and 0D37
1567 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1569 [20098] libc: FAIL: debug/backtrace-tst on hppa
1570 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1572 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1574 [20313] localedata: Update locale data to Unicode 9.0
1575 [20424] manual: Document how to provide a malloc replacement
1576 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1578 [20686] locale: Add el_GR@euro to SUPPORTED.
1579 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1580 failures consistently
1581 [21015] dynamic-link: Document and fix --enable-bind-now
1582 [21016] nptl: pthread_cond support is broken on hppa
1583 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1584 frame-pointer on i386
1585 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1587 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1588 [21088] libc: Build fails with --enable-static-nss
1589 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1590 generic c code is used
1591 [21109] libc: Tunables broken on big-endian
1592 [21112] math: powf has large ulp errors with base close to 1 and exponent
1594 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1596 [21120] malloc: glibc malloc is incompatible with GCC 7
1597 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1599 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1600 (-0.0) when linking with -lieee
1601 [21171] math: log10, log2 and lgamma return incorrect results
1602 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1603 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1604 [21207] localedata: ce_RU: update weekdays from CLDR
1605 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1606 [21217] localedata: Update months from CLDR-31
1607 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1608 [21243] libc: support_delete_temp_file should issue warning for failed
1610 [21244] libc: support resolv_test_start() socket fd close should be
1612 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1613 new posix_spawn implementation
1614 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1615 leads to lower CPU frequency
1616 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1617 [21261] libc: [sparc64] bits/setjmp.h namespace
1618 [21267] network: [mips] bits/socket.h IOC* namespace
1619 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1620 [21270] libc: mmap64 silently truncates large offset values
1621 [21275] libc: posix_spawn always crashes on ia64 now
1622 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1623 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1624 [21289] libc: Incorrect declaration for 32-bit platforms with
1625 _FILE_OFFSET_BITS=64 causes build error
1626 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1627 not support gethostbyname4_r
1628 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1630 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1632 [21340] libc: Support POSIX_SPAWN_SETSID
1633 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1634 [21359] network: ns_name_pack needs additional byte in destination buffer
1635 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1636 against fragmentation attacks (CVE-2017-12132)
1637 [21369] network: resolv: Remove EDNS fallback
1638 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1640 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1641 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1642 [21393] stdio: Missing dup3 error check in freopen, freopen64
1643 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1644 [21399] localedata: Bad description for U00EC in
1645 localedata/charmaps/CP1254
1646 [21411] malloc: realloc documentation error
1647 [21426] network: sys/socket.h uio.h namespace
1648 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1649 [21445] libc: signal.h bsd_signal namespace
1650 [21455] network: Network headers stdint.h namespace
1651 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1653 [21475] network: resolv: Overlong search path is truncated mid-label
1654 [21511] libc: sigstack namespace
1655 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1656 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1657 bits/syscall.h: No such file or directory
1658 [21517] libc: struct sigaltstack namespace
1659 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1660 [21533] localedata: Update locale data to Unicode 10.0
1662 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1663 junk at end of line, first unrecognized character is `@'
1664 [21538] libc: SIG_HOLD missing for XPG4
1665 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1666 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1667 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1668 [21543] libc: sigevent namespace
1669 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1670 size for general purpose registers in mcontext_t structure
1671 [21550] libc: sigwait namespace
1672 [21552] libc: XPG4 bsd_signal namespace
1673 [21554] libc: sigpause namespace
1674 [21560] libc: sys/wait.h signal.h namespace
1675 [21561] libc: waitid namespace
1676 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1677 '/usr/include/stdlib.h': Permission denied
1678 [21575] libc: sys/wait.h missing struct rusage definition
1679 [21584] libc: sigaltstack etc namespace
1680 [21597] libc: siginterrupt namespace
1681 [21607] math: hppa: FAIL: math/test-tgmath
1682 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1684 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1685 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1686 stack and heap (CVE-2017-1000366)
1687 [21625] libc: wait3 namespace
1688 [21654] nss: Incorrect pointer alignment in NSS group merge result
1690 [21657] network: Parse interface zone id for node-local multicast
1691 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1693 [21666] libc: .symver is used on common symbol
1694 [21668] network: resolv: res_init cross-thread broadcast introduces race
1696 [21687] math: tgmath.h totalorder, totalordermag return type
1697 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1699 [21696] libc: Incorrect assumption of of __cpu_mask in
1700 posix/sched_cpucount.c
1701 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1702 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1704 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1705 leading to relocation crash
1706 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1707 `resolv_conf_matches (resp, conf)' failed.
1708 [21710] localedata: Added Samoan language locale for Samoa
1709 [21711] localedata: Pashto yesstr/nostr locale are missing
1710 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1711 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1712 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1713 [21723] localedata: yesstr/nostr missing for Chinese language locale
1714 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1716 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1718 [21728] localedata: New Locale for Tongan language
1719 [21729] localedata: incorrect LC_NAME fields for hi_IN
1720 [21733] localedata: yesstr and nostr are missing for zh_HK
1721 [21734] localedata: Missing yesstr and nostr are for kw_GB
1722 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1723 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1724 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1725 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1727 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1729 [21749] localedata: Wrong abbreviated day name (“abday”) for
1731 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1732 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1733 [21759] localedata: missing yesstr and nostr for Tigrinya
1734 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1735 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1736 [21767] localedata: Missing Bislama locales
1737 [21768] localedata: Missing yesstr and nostr for aa_ET
1738 [21770] localedata: Missing Field in li_NL
1739 [21778] nptl: Robust mutex may deadlock
1740 [21779] libc: MicroBlaze segfaults when loading libpthread
1741 [21783] localedata: Fix int_select international_call_prefixes
1742 [21784] localedata: Inconsistency in country_isbn
1743 [21788] localedata: Missing Country Postal Abbreviations
1744 [21794] localedata: Added-country_isbn-for-Italy
1745 [21795] localedata: Add/Fix country_isbn for France
1746 [21796] localedata: Added country_isbn for Republic of Korea
1747 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1748 [21799] localedata: Added int_select international_call_prefixes
1749 [21801] localedata: Added int_select international_call_prefixes
1750 [21804] nptl: Double semicolon in thread-shared-types.h
1751 [21807] localedata: LC_ADDRESS fix for pap_CW
1752 [21808] localedata: Fix LC_ADDRESS for pap_AW
1753 [21821] localedata: Added country_name in mai_IN
1754 [21822] localedata: Fix LC_TIME for mai_IN
1755 [21823] localedata: missing yesstr, nostr for sa_IN
1756 [21825] localedata: Fix name_mrs for mag_IN
1757 [21828] localedata: 2.26 changelog should mention user visible changes
1759 [21835] localedata: Added Maithili language locale for Nepal
1760 [21838] localedata: Removed redundant data for the_NP
1761 [21839] localedata: Fix LC_MONETARY for ta_LK
1762 [21844] localedata: Fix Latin characters and Months Sequence.
1763 [21848] localedata: Fix mai_NP Title Name
1768 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1769 24731-2:2010, is supported to enable declarations of functions from that
1770 TR. Note that not all functions from that TR are supported by the GNU C
1773 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1774 18661-1:2014, is supported to enable declarations of functions and macros
1775 from that TS. Note that not all features from that TS are supported by
1778 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1779 18661-4:2015, is supported to enable declarations of functions and macros
1780 from that TS. Note that most features from that TS are not supported by
1783 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1784 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1785 Since the GNU C Library defaults to a much newer revision of POSIX, this
1786 will only affect programs that specifically request an old conformance
1787 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1788 see a change in the visible declarations, but a program compiled with
1789 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1792 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1793 defined by all multithreaded code, but glibc has not required this for
1796 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1797 means that in a future release, the macros “major”, “minor”, and “makedev”
1798 will only be available from <sys/sysmacros.h>.
1800 These macros are not part of POSIX nor XSI, and their names frequently
1801 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1802 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1803 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1806 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1807 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1808 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1810 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1811 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1812 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1813 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1814 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1815 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1816 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1817 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1818 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1819 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1820 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1823 * New <math.h> features are added from TS 18661-1:2014:
1825 - Signaling NaN macros: SNANF, SNAN, SNANL.
1827 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1828 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1829 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1831 - llogb functions: the llogb, llogbf and llogbl functions, and the
1832 FP_LLOGB0 and FP_LLOGBNAN macros.
1834 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1837 - Comparison macros: iseqsig.
1839 - Classification macros: iscanonical, issubnormal, iszero.
1841 - Total order functions: totalorder, totalorderf, totalorderl,
1842 totalordermag, totalordermagf, totalordermagl.
1844 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1846 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1847 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1849 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1850 are added to libc. They convert a floating-point number into string.
1852 * Most of glibc can now be built with the stack smashing protector enabled.
1853 It is recommended to build glibc with --enable-stack-protector=strong.
1854 Implemented by Nick Alcock (Oracle).
1856 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1857 intended to be used instead of memset() to erase sensitive data after use;
1858 the compiler will not optimize out calls to explicit_bzero even if they
1859 are "unnecessary" (in the sense that no _correct_ program can observe the
1860 effects of the memory clear).
1862 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1863 to float instead of double. This does not affect the ABI of any libraries
1864 that are part of the GNU C Library, but may affect the ABI of other
1865 libraries that use this type in their interfaces.
1867 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1868 float_t and double_t types are now defined to long double instead of float
1869 and double. These options are not the default, and this does not affect
1870 the ABI of any libraries that are part of the GNU C Library, but it may
1871 affect the ABI of other libraries that use this type in their interfaces,
1872 if they are compiled or used with those options.
1874 * The getentropy and getrandom functions, and the <sys/random.h> header file
1877 * The buffer size for byte-oriented stdio streams is now limited to 8192
1878 bytes by default. Previously, on Linux, the default buffer size on most
1879 file systems was 4096 bytes (and thus remains unchanged), except on
1880 network file systems, where the buffer size was unpredictable and could be
1881 as large as several megabytes.
1883 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1884 for the Linux quota interface which predates kernel version 2.4.22 has
1887 * The malloc_get_state and malloc_set_state functions have been removed.
1888 Already-existing binaries that dynamically link to these functions will
1889 get a hidden implementation in which malloc_get_state is a stub. As far
1890 as we know, these functions are used only by GNU Emacs and this change
1891 will not adversely affect already-built Emacs executables. Any undumped
1892 Emacs executables, which normally exist only during an Emacs build, should
1893 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1895 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1896 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1897 “no-ip6-dotint” had already been the default, and support for the
1898 “ip6-dotint” option was removed from the Internet in 2006.
1900 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1901 flag from <resolv.h> have been removed. The option relied on a
1902 backwards-incompatible DNS extension which was never deployed on the
1905 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1906 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1907 They were already unimplemented.
1909 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1910 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1911 removed again from the IETF name lookup interface specification in RFC
1912 2553. Applications should use getaddrinfo instead.
1914 * DNSSEC-related declarations and definitions have been removed from the
1915 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1916 decode the data part of DNSSEC record types. Previous versions of glibc
1917 only implemented minimal support for the previous version of DNSSEC, which
1918 is incompatible with the currently deployed version.
1920 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1921 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1922 <arpa/nameser.h> header file because the distinction between RR types and
1923 meta-RR types is not officially standardized, subject to revision, and
1924 thus not suitable for encoding in a macro.
1926 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1927 and rhook members of the res_state type in <resolv.h> have been removed.
1928 The glibc stub resolver did not support these hooks, but the header file
1929 did not reflect that.
1931 * For multi-arch support it is recommended to use a GCC which has
1932 been built with support for GNU indirect functions. This ensures
1933 that correct debugging information is generated for functions
1934 selected by IFUNC resolvers. This support can either be enabled by
1935 configuring GCC with '--enable-gnu-indirect-function', or by
1936 enabling it by default by setting 'default_gnu_indirect_function'
1937 variable for a particular architecture in the GCC source file
1940 * GDB pretty printers have been added for mutex and condition variable
1941 structures in POSIX Threads. When installed and loaded in gdb these pretty
1942 printers show various pthread variables in human-readable form when read
1943 using the 'print' or 'display' commands in gdb.
1945 * Tunables feature added to allow tweaking of the runtime for an application
1946 program. This feature can be enabled with the '--enable-tunables' configure
1947 flag. The GNU C Library manual has details on usage and README.tunables has
1948 instructions on adding new tunables to the library.
1950 * A new version of condition variables functions have been implemented in
1951 the NPTL implementation of POSIX Threads to provide stronger ordering
1954 * A new version of pthread_rwlock functions have been implemented to use a more
1955 scalable algorithm primarily through not using a critical section anymore to
1958 Security related changes:
1960 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1961 have been created with makecontext could fail to terminate due to a
1962 missing .cantunwind annotation. This has been observed to lead to a hang
1963 (denial of service) in some Go applications compiled with gccgo. Reported
1964 by Andreas Schwab. (CVE-2016-6323)
1966 * The DNS stub resolver functions would crash due to a NULL pointer
1967 dereference when processing a query with a valid DNS question type which
1968 was used internally in the implementation. The stub resolver now uses a
1969 question type which is outside the range of valid question type values.
1972 The following bugs are resolved with this release:
1974 [4099] stdio: Overly agressive caching by stream i/o functions.
1975 [7065] build: Support building glibc with -fstack-protector or -fstack-
1977 [9842] localedata: en_CA: incorrect date format
1978 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1979 before it started waiting
1980 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1981 [15765] nptl: sem_open is wrongly a cancellation point
1982 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1983 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1985 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1986 library linked with pthread
1987 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1988 "FIXME: Ingo" issue)
1989 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1990 [17252] libc: getrandom and getentropy syscall
1991 [17730] malloc: thread-local storage is sometimes improperly free()'d
1992 after being __libc_memalign()'d
1993 [18241] stdio: failed fseek on memstream does not set errno and can fail
1995 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1997 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1998 PTHREAD_PRIO_INHERIT mutex on ARM
1999 [18784] network: res_query and related function crash for special record
2000 type queries (CVE-2015-5180)
2001 [19380] math: strtod does not raise "inexact"
2002 [19387] string: Integer overflow in memchr
2003 [19390] string: Integer overflow in strncat
2004 [19398] build: linknamespace tests fail with massively parallel build
2005 [19402] nptl: Deadlock with robust shared mutex and asynchronous
2007 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
2008 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
2009 [19514] libc: [PATCH] Fix spelling errors in spelling
2010 "implement"/"implementation" in several places
2011 [19582] network: Deprecate RES_USE_INET6
2012 [19673] manual: clog10 docs appear to be erroneous
2013 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
2014 causes a segmentation fault
2015 [19826] libc: invalid pointer returned from __tls_get_addr with static
2017 [20016] network: resolv: Remove hooks support from the API
2018 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
2020 [20033] math: [x86_64] vectorized math function don't call the __finite
2022 [20116] nptl: use after free in pthread_create
2023 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
2024 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
2026 [20311] nptl: please install proc_service.h
2027 [20366] libc: Compilation errors in installed headers in strict-compliance
2029 [20370] malloc: malloc: Arena free list management is still racy
2030 (incorrect fix in bug 19243)
2031 [20386] libc: assert (X = 0) does not result in GCC warning
2032 [20432] malloc: malloc: Minimize interface required for interposition
2033 [20435] libc: Missing unwind info in __startcontext causes infinite loop
2034 in _Unwind_Backtrace (CVE-2016-6323)
2035 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
2036 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
2038 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
2039 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
2040 [20477] network: resolv: incorrect double-checked locking related to
2042 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
2044 [20495] math: x86_64 performance degradation due to AVX/SSE transition
2046 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
2047 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
2048 cause transition penalty
2049 [20517] math: sparcv9 missing fdiml compat symbol
2050 [20524] manual: strverscmp is inconsistent
2051 [20525] libc: <sys/quota.h> should be based on kernel headers
2052 [20539] math: GCC 7 -static -lm fails to link at -O3
2053 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
2055 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
2057 [20591] network: Remove obsolete DNSSEC support
2058 [20592] network: DNS resource record type classification macros in
2059 <arpa/nameser.h> are incorrect
2060 [20593] network: Update DNS RR type definitions
2061 [20611] network: getaddrinfo accepts invalid numeric scope IDs
2062 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
2064 [20629] network: libresolv: Remove support for bitstring labels
2066 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
2067 assertion: "maxidx >= 2"
2068 [20660] math: [arm] Use VSQRT
2069 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
2070 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
2071 [20689] libc: Test for FMA should also check for AVX.
2072 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
2073 [20715] math: iszero macro breaks existing code
2074 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
2075 [20728] libc: powerpc: Missing TOC stub in clone
2076 [20729] build: glibc-2.24 fails to build for i486 with -Os
2077 [20750] build: Build fails with default PIE enabled gcc-6.x
2078 [20768] math: [alpha] sqrt fegetenv namespace
2079 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
2081 [20787] math: float_t is defined as float incorrectly on x86_64 even if
2082 __FLT_EVAL_METHOD__ is 2
2083 [20790] network: rpcgen buffer overrun in get_prog_declaration
2084 [20822] nptl: race condition in __lll_unlock_elision on powerpc
2085 [20829] libc: crypt snprintf namespace
2086 [20847] libc: tst-vfork3 failure
2087 [20855] math: Default bits/mathdef.h has inappropriate float_t
2088 [20859] math: [sh4] FP_ILOGB0 invalid
2089 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
2090 U+20AC), not same as GBK
2091 [20915] dynamic-link: global-dynamic TLS broken on aarch64
2092 [20916] math: pow handling of sNaN arguments
2093 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
2094 [20919] math: Bad pow (qNaN, 0) result with -lieee
2095 [20940] math: hypot sNaN handling
2096 [20947] math: fmax, fmin sNaN handling
2097 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
2098 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
2100 [20971] string: powerpc64/power7 memchr overflows internal pointer check
2101 [20973] nptl: robust mutexes: Lost wake-ups
2102 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
2103 [20978] nis: strlen on null pointer in nss_nisplus
2104 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
2106 [21014] string: i686 memchr overflows internal pointer check
2107 [21019] libc: [mips] n32 lseek incorrect on overflow
2108 [21022] libc: [microblaze] __backtrace get_frame_size namespace
2109 [21026] libc: [MIPS] readahead syscall is broken on n64
2110 [21028] math: Fallback fesetexceptflag should always succeed
2111 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
2112 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
2114 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
2115 [21061] librt: [microblaze] librt lost clock_* exports
2116 [21073] libc: tunables: insecure environment variables passed to
2117 subprocesses with AT_SECURE
2118 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
2122 * The minimum Linux kernel version that this version of the GNU C Library
2123 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
2124 kernel version 2.6.32 or later suffices (on architectures that already
2125 required kernel versions more recent than 3.2, those requirements remain
2126 unchanged). Linux 3.2 or later kernel headers are required on all
2129 * The pap_AN locale has been deleted. This has been deprecated for a long
2130 time. It has been replaced by pap_AW & pap_CW, both of which have long
2131 been included in previous releases.
2133 * The readdir_r and readdir64_r functions have been deprecated. It is
2134 recommended to use readdir and readdir64 instead.
2136 * The type “union wait” has been removed. It was deprecated in the early
2137 1990s and never part of POSIX. Application code should use the int type
2138 instead of “union wait”.
2140 * A new NSS action is added to facilitate large distributed system
2141 administration. The action, MERGE, allows remote user stores like LDAP
2142 to be merged into local user stores like /etc/groups in order to provide
2143 easy to use, updated, and managed sets of merged credentials. The new
2144 action can be used by configuring it in /etc/nsswitch.conf:
2145 group: files [SUCCESS=merge] nis
2146 Implemented by Stephen Gallagher (Red Hat).
2148 * The deprecated __malloc_initialize_hook variable has been removed from the
2151 * The long unused localedef --old-style option has been removed. It hasn't
2152 done anything in over 16 years. Scripts using this option can safely
2155 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
2156 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
2157 return the next representable value in the direction of positive infinity
2158 and the nextdown functions return the next representable value in the
2159 direction of negative infinity. These are currently enabled as GNU
2162 Security related changes:
2164 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
2165 could result in a stack overflow when getnetbyname was called with an
2166 overly long name. (CVE-2016-3075)
2168 * Previously, getaddrinfo copied large amounts of address data to the stack,
2169 even after the fix for CVE-2013-4458 has been applied, potentially
2170 resulting in a stack overflow. getaddrinfo now uses a heap allocation
2171 instead. Reported by Michael Petlan. (CVE-2016-3706)
2173 * The glob function suffered from a stack-based buffer overflow when it was
2174 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
2175 Reported by Alexander Cherepanov. (CVE-2016-1234)
2177 * The Sun RPC UDP client could exhaust all available stack space when
2178 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
2179 alloca plugin for GCC. (CVE-2016-4429)
2181 * The IPv6 name server management code in libresolv could result in a memory
2182 leak for each thread which is created, performs a failing naming lookup,
2183 and exits. Over time, this could result in a denial of service due to
2184 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
2186 The following bugs are resolved with this release:
2188 [1170] localedata: ne_NP: update Nepali locale definition file
2189 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
2191 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
2192 [6796] math: fdim() does not set errno on overflow
2193 [10354] libc: posix_spawn should use vfork() in more cases than presently
2194 [11213] localedata: localedata: add copyright disclaimer to locale files
2195 [12143] localedata: chr_US: new Cherokee locale
2196 [12450] localedata: sgs_LT: new locale
2197 [12676] localedata: ln_CD: new locale
2198 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
2200 [13304] math: fma, fmaf, fmal produce wrong results
2201 [14259] build: --localedir arg to configure is ignored
2202 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
2203 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
2204 [14934] localedata: es_CL: wrong first weekday chilean locale
2205 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
2207 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
2209 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
2210 [15368] nptl: raise() is not async-signal-safe
2211 [15479] math: ceil, floor, round and trunc raise inexact exception
2212 [15578] localedata: kk_KZ: various updates
2213 [16003] localedata: pap_AN: punt old locale
2214 [16137] localedata: iw_IL: punt old locale
2215 [16190] localedata: eo: new esperanto locale
2216 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
2217 [16742] malloc: race condition: pthread_atfork() called before first
2218 malloc() results in unexpected locking behaviour/deadlocks
2219 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
2221 [16983] localedata: postal_fmt does not allow %l and %n modifiers
2222 [17565] localedata: pt_PT: wrong (work-)week start
2223 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
2224 [17950] build: Build fails with -msse
2225 [18205] localedata: be_BY*: wrong first_weekday and first_workday
2226 [18433] libc: posix_spawn does not return correctly upon failure to
2228 [18453] localedata: charmaps/IBM875: incorrect codes
2229 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2231 [18896] localedata: he_IL: improvements for currency
2232 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2233 Romanian locale data
2234 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2236 [19056] libc: Deprecate readdir_r
2237 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2239 [19198] localedata: nl_NL: small improvements for Dutch locales
2240 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2241 nameservers (CVE-2016-5417)
2242 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2244 [19400] locale: Language missing in "iso-639.def", trivial fix in
2246 [19431] malloc: Deadlock between fflush, getdelim, and fork
2247 [19505] libc: Incorrect file descriptor validity checks in
2248 posix_spawn_file_actions_add{open,close,dup2}
2249 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2250 when using RTLD_NEXT
2251 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2252 `intl/{gettextP,loadinfo}.h'
2253 [19534] libc: execle, execlp may use malloc
2254 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2255 [19573] network: res_nclose and __res_maybe_init disagree about name
2256 server initialization, breaking Hesiod
2257 [19575] localedata: Status of GB18030 tables
2258 [19581] localedata: sr_* date_fmt string contains additional newline
2259 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2261 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2262 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2263 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2264 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2266 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2268 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2270 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2271 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2272 [19633] locale: strfmon_l applies global locale to number formatting
2273 [19642] network: Memory leak in getnameinfo
2274 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2275 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2277 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2278 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2280 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2281 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2283 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2285 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2287 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2288 pointers and lengths in error-case.
2289 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2290 report errors on UTF-16 surrogates values.
2291 [19755] nscd: nscd assertion failure in gc
2292 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2293 [19759] libc: mempcpy shouldn't be inlined
2294 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2295 [19765] libc: s390 needs an optimized mempcpy
2296 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2297 NAME_MAX limit assumption (CVE-2016-1234)
2298 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2299 [19787] network: Missing and incorrect truncation checks in getnameinfo
2300 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2302 [19791] network: Assertion failure in res_query.c with un-connectable name
2304 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2305 [19822] math: libm.so install clobbers old version
2306 [19825] network: resolv: send_vc can return uninitialized data in second
2307 response to getaddrinfo
2308 [19830] network: nss_dns: should check RDATA length against buffer length
2309 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2310 confronted with A/AAAA records of invalid size
2311 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2312 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2313 [19853] stdio: Printing IBM long double in decimal with high precision is
2315 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2316 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2317 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2319 [19865] network: Assertion failure or memory leak in
2320 _nss_dns_getcanonname_r
2321 [19868] network: nss_dns: netent code does not skip over non-PTR records
2322 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2324 [19881] string: Improve x86-64 memset
2325 [19907] string: Incorrect memcpy tests
2326 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2328 [19925] libc: termios.h XCASE namespace
2329 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2330 [19929] libc: limits.h NL_NMAX namespace
2331 [19931] stdio: Memory leak in vfprintf
2332 [19957] libc: clone(CLONE_VM) access invalid parent memory
2333 [19963] localedata: en_IL: New locale
2334 [19989] stdio: stdio.h cuserid namespace
2335 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2337 [19996] locale: langinfo.h nl_langinfo_l namespace
2338 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2339 [20010] network: getaddrinfo: Stack overflow in hostent translation
2341 [20012] stdio: libio: fmemopen append mode failure
2342 [20014] stdio: stdio.h namespace for pre-threads POSIX
2343 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2344 [20023] libc: fcntl.h timespec namespace
2345 [20024] math: [x86_64] vectorized sincos trashes the stack
2346 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2347 [20041] time: sys/time.h timespec namespace
2348 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2349 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2350 [20051] libc: ttyslot in wrong header under wrong conditions
2351 [20054] libc: gethostname not declared for XPG4
2352 [20055] libc: termios.h missing tcgetsid for XPG4
2353 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2355 [20073] libc: sys/stat.h fchmod namespace
2356 [20074] libc: stdlib.h rand_r namespace
2357 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2358 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2360 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2361 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2363 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2364 [20119] libc: Wrong mask for processors level type from CPUID
2365 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2366 AS not supporting AVX512
2367 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2369 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2370 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2372 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2373 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2374 [20178] libc: posix_spawn{p} should not call exit
2375 [20191] stdio: libio: vtables hardening
2376 [20195] string: FMA4 detection requires CPUID execution with register
2378 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2379 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2381 [20212] math: acos (sNaN) returns sNaN
2382 [20213] math: asin (sNaN) returns sNaN
2383 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2384 [20218] math: [i386] asinhl (sNaN) returns sNaN
2385 [20219] math: [i386] atanhl (sNaN) returns sNaN
2386 [20222] stdio: fopencookie: Mangle function pointers
2387 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2388 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2389 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2391 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2392 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2393 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2394 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2395 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2396 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2397 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2398 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2399 [20240] math: modf (sNaN) returns sNaN
2400 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2401 [20250] math: frexp (sNaN) returns sNaN
2402 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2403 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2405 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2406 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2407 subscript is below array bounds [-Werror=array-bounds]
2408 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2409 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2411 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2412 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2414 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2415 "invalid" exceptions
2416 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2417 [20316] localedata: id_ID: Februari instead of Pebruari
2418 [20327] string: POWER8 strcasecmp returns incorrect result
2419 [20347] math: Failure: Test: j0_downward (0xap+0)
2420 [20348] libc: FAIL: misc/tst-preadvwritev64
2421 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2422 [20350] libc: There is no test for p{read,write}64
2423 [20357] math: Incorrect cos result for 1.5174239687223976
2424 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2428 * Unicode 8.0.0 Support: Character encoding, character type info, and
2429 transliteration tables are all updated to Unicode 8.0.0, using new
2430 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2431 These updates cause user visible changes, such as the fixes for bugs
2432 89, 16061, and 18568.
2434 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2435 kernel-internal CPU set size. This means that requests that change the
2436 CPU affinity which failed before (for example, an all-ones CPU mask) will
2437 now succeed. Applications that need to determine the effective CPU
2438 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2439 setting it because the kernel can adjust it (and the previous size check
2440 would not detect this in the majority of cases).
2442 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2443 the following new symbols are used: fts64_children, fts64_close,
2444 fts64_open, fts64_read and fts64_set.
2446 * getaddrinfo now detects certain invalid responses on an internal netlink
2447 socket. If such responses are received, an affected process will
2448 terminate with an error message of "Unexpected error <number> on netlink
2449 descriptor <number>" or "Unexpected netlink response of size <number> on
2450 descriptor <number>". The most likely cause for these errors is a
2451 multi-threaded application which erroneously closes and reuses the netlink
2452 file descriptor while it is used by getaddrinfo.
2454 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2455 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2456 unnecessary serialization of memory allocation requests across threads.
2457 The defect is now corrected. Users should see a substantial increase in
2458 the concurent throughput of allocation requests for applications which
2459 trigger this bug. Affected applications typically create and
2460 destroy threads frequently. (Bug 19048 was reported and analyzed by
2463 * There is now a --disable-timezone-tools configure option for disabling the
2464 building and installing of the timezone related utilities (zic, zdump, and
2465 tzselect). This is useful for people who build the timezone data and code
2466 independent of the GNU C Library.
2468 * The obsolete header <regexp.h> has been removed. Programs that require
2469 this header must be updated to use <regex.h> instead.
2471 * The obsolete functions bdflush, create_module, get_kernel_syms,
2472 query_module and uselib are no longer available to newly linked binaries;
2473 the header <sys/kdaemon.h> has been removed. These functions and header
2474 were specific to systems using the Linux kernel and could not usefully be
2475 used with the GNU C Library on systems with version 2.6 or later of the
2478 * Optimized string, wcsmbs and memory functions for IBM z13.
2479 Implemented by Stefan Liebler.
2481 * Newly linked programs that define a variable called signgam will no longer
2482 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2483 require signgam to be set by those functions must ensure that they use the
2484 variable provided by the GNU C Library and declared in <math.h>, without
2485 defining their own copy.
2487 * The minimum GCC version that can be used to build this version of the GNU
2488 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2489 still be used to compile programs using the GNU C Library.
2491 Security related changes:
2493 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2494 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2496 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2497 the pointer guard feature. It is always enabled. Previously,
2498 LD_POINTER_GUARD could be used to disable security hardening in binaries
2499 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2502 * An integer overflow in hcreate and hcreate_r could lead to an
2503 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2505 * The catopen function no longer has unbounded stack usage. Reported by
2506 Max. (CVE-2015-8779)
2508 * The nan, nanf and nanl functions no longer have unbounded stack usage
2509 depending on the length of the string passed as an argument to the
2510 functions. Reported by Joseph Myers. (CVE-2014-9761)
2512 * A stack-based buffer overflow was found in libresolv when invoked from
2513 libnss_dns, allowing specially crafted DNS responses to seize control
2514 of execution flow in the DNS client. The buffer overflow occurs in
2515 the functions send_dg (send datagram) and send_vc (send TCP) for the
2516 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2517 family. The use of AF_UNSPEC triggers the low-level resolver code to
2518 send out two parallel queries for A and AAAA. A mismanagement of the
2519 buffers used for those queries could result in the response of a query
2520 writing beyond the alloca allocated buffer created by
2521 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2522 the overflow. Thanks to the Google Security Team and Red Hat for
2523 reporting the security impact of this issue, and Robert Holiday of
2524 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2526 The following bugs are resolved with this release:
2528 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2529 [887] math: Math library function "logb" and "nextafter" inconsistent
2530 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2531 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2532 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2533 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2535 [4404] localedata: German translation of "Alarm clock" is misleading
2536 [6799] math: nextafter() and nexttoward() doen't set errno on
2537 overflow/underflow errors
2538 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2540 [10432] nis: _nss_nis_setnetgrent assertion failure
2541 [11460] libc: fts has no LFS support
2542 [12926] network: getaddrinfo()/make_request() may spin forever
2543 [13065] nptl: Race condition in pthread barriers
2544 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2545 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2547 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2548 [14912] libc: Rename non-installed bits/*.h headers
2549 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2550 [15367] math: Let gcc use __builtin_isinf
2551 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2552 [15421] math: lgamma wrongly sets signgam for ISO C
2553 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2554 FE_INVALID with argument out of range
2555 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2557 [15786] dynamic-link: ifunc resolver functions can smash function
2559 [15918] math: Unnecessary check for equality in hypotf()
2560 [16061] localedata: Review / update transliteration data
2561 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2563 [16141] time: strptime %z offset restriction
2564 [16171] math: drem should be alias of remainder
2565 [16296] math: fegetround is pure?
2566 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2567 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2569 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2570 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2572 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2573 "Exception "Inexact" set" on ppc32
2574 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2575 [16517] math: Missing underflow exception from tanf/tan/tanl
2576 [16519] math: Missing underflow exception from sinhf
2577 [16520] math: Missing underflow exception from tanhf
2578 [16521] math: Missing underflow exception from exp2
2579 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2581 [16734] stdio: fopen calls mmap to allocate its buffer
2582 [16961] math: nan function incorrect handling of bad sequences
2583 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2584 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2585 [16985] locale: localedef: confusing error message when opening output
2587 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2588 [17197] locale: Redundant shift character in iconv conversion output at
2590 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2591 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2592 [17250] dynamic-link: static linking breaks nss loading
2593 (getaddrinfo/getpwnam/etc...)
2594 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2596 [17441] math: isnan() should use __builtin_isnan() in GCC
2597 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2598 (related to lock elision)
2599 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2600 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2601 [17887] time: strptime should be able to parse "+01:00" style timezones
2602 [17905] libc: catopen() Multiple unbounded stack allocations
2604 [18084] libc: backtrace (..., 0) dumps core on x86
2605 [18086] libc: nice() sets errno to 0 on success
2606 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2607 is too large (CVE-2015-8778)
2608 [18251] dynamic-link: SONAME missing when audit modules provides path
2609 [18265] libc: add attributes for wchar string and memory functions
2610 [18370] math: csqrt missing underflows
2611 [18421] libc: [hppa] read-only segment has dynamic relocations
2612 [18472] libc: Obsolete syscall wrappers should be compat symbols
2613 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2614 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2615 [18525] localedata: Remove locale timezone information
2616 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2617 [18568] localedata: Update locale data to Unicode 8.0
2618 [18589] locale: sort-test.sh fails at random
2619 [18595] math: ctan, ctanh missing underflows
2620 [18604] libc: assert macro-expands its argument
2621 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2622 contains a vector instruction exception.
2623 [18611] math: j1, jn missing errno setting on underflow
2624 [18618] localedata: sync Chechen locale definitions with other *_RU
2626 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2627 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2628 [18665] network: In send_dg, the recvfrom function is NOT always using the
2629 buffer size of a newly created buffer (CVE-2015-7547)
2630 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2632 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2634 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2635 [18699] math: tilegx cproj() for various complex infinities does not yield
2637 [18724] libc: Harden put*ent functions against data injection
2638 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2639 [18755] build: build errors with -DNDEBUG
2640 [18757] stdio: fmemopen fails to set errno on failure
2641 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2643 [18781] libc: openat64 lacks O_LARGEFILE
2644 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2645 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2646 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2647 [18790] math: [ldbl-128ibm] tanhl inaccurate
2648 [18795] libc: stpncpy fortification misses buffer lengths that are
2649 statically too large
2650 [18796] build: build fails for --disable-mathvec
2651 [18803] math: hypot missing underflows
2652 [18820] stdio: fmemopen may leak memory on failure
2653 [18823] math: csqrt spurious underflows
2654 [18824] math: fma spurious underflows
2655 [18825] math: pow missing underflows
2656 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2657 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2658 [18870] build: sem_open.c fails to compile with missing symbol
2660 [18872] stdio: Fix memory leak in printf_positional
2661 [18873] libc: posix_fallocate overflow check ineffective
2662 [18875] math: Excess precision leads incorrect libm
2663 [18877] libc: arm: mmap offset regression
2664 [18887] libc: memory corruption when using getmntent on blank lines
2665 [18918] localedata: hu_HU: change time to HH:MM:SS format
2666 [18921] libc: Regression: extraneous stat() and fstat() performed by
2668 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2669 binaries (CVE-2015-8777)
2670 [18951] math: tgamma missing underflows
2671 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2673 [18953] localedata: lt_LT: change currency symbol to the euro
2674 [18956] math: powf inaccuracy
2675 [18961] math: [i386] exp missing underflows
2676 [18966] math: [i386] exp10 missing underflows
2677 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2678 [18969] build: multiple string test failures due to missing locale
2680 [18970] libc: Reference of pthread_setcancelstate in libc.a
2681 [18977] math: float / long double Bessel functions not in XSI POSIX
2682 [18980] math: i386 libm functions return with excess range and precision
2683 [18981] math: i386 scalb*, ldexp return with excess range and precision
2684 [18982] stdio: va_list and vprintf
2685 [18985] time: Passing out of range data to strftime() causes a segfault
2687 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2688 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2689 [19012] locale: iconv_open leaks memory on error path
2690 [19016] math: clog, clog10 inaccuracy
2691 [19018] nptl: Mangle function pointers in tls_dtor_list
2692 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2693 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2694 [19048] malloc: malloc: arena free list can become cyclic, increasing
2696 [19049] math: [powerpc] erfc incorrect zero sign
2697 [19050] math: [powerpc] log* incorrect zero sign
2698 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2699 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2700 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2701 [19074] network: Data race in _res_hconf_reorder_addrs
2702 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2703 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2704 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2706 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2708 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2709 [19086] manual: posix_fallocate64 documented argument order is wrong.
2710 [19088] math: lround, llround missing exceptions close to overflow
2712 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2713 [19095] math: dbl-64 lrint incorrect for 64-bit long
2714 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2715 [19124] dynamic-link: ld.so failed to build with older assmebler
2716 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2717 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2718 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2719 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2720 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2721 pthread_setaffinity_np
2722 [19156] math: [ldbl-128] j0l spurious underflows
2723 [19164] nptl: tst-getcpu fails with many possible CPUs
2724 [19168] math: math/test-ildoubl and math/test-ldouble failure
2725 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2726 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2728 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2729 (FE_NOMASK_ENV) do not clear SSE exceptions
2730 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2731 _int_new_arena/reused_arena
2732 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2733 [19201] math: dbl-64 remainder incorrect sign of zero result
2734 [19205] math: bits/math-finite.h conditions do not match math.h and
2736 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2737 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2738 for C99-based standards
2739 [19212] libc: features.h not -Wundef clean
2740 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2742 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2743 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2744 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2746 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2747 "inexact" exceptions
2748 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2750 [19242] libc: strtol incorrect in Turkish locales
2751 [19243] malloc: reused_arena can pick an arena on the free list, leading
2752 to an assertion failure and reference count corruption
2753 [19253] time: tzset() ineffective when temporary TZ did not include DST
2755 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2756 [19270] math: [hppa] Shared libm missing __isnanl
2757 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2758 MAP_HUGETLB and MAP_STACK defines
2759 [19313] nptl: Wrong __cpu_mask for x32
2760 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2761 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2762 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2763 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2764 [19363] time: x32: times() return value wrongly truncates/sign extends
2766 [19367] dynamic-link: Improve branch prediction on Silvermont
2767 [19369] network: Default domain name not reset by res_ninit when "search"
2768 / "domain" entry is removed from resolv.conf
2769 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2771 [19385] localedata: bg_BG: time separator should be colon, not comma
2772 [19408] libc: linux personality syscall wrapper may erroneously return an
2773 error on 32-bit architectures
2774 [19415] libc: dladdr returns wrong names on hppa
2775 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2776 IBM905, IBM907, and IBM909
2777 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2778 [19443] build: build failures with -DDEBUG
2779 [19451] build: Make check fails on test-double-vlen2
2780 [19462] libc: Glibc failed to build with -Os
2781 [19465] math: Wrong code with -Os
2782 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2784 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2786 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2787 [19486] math: S390: Math tests fail with "Exception Inexact set".
2788 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2789 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2791 [19590] math: Fail to build shared objects that use libmvec.so functions.
2795 * The following bugs are resolved with this release:
2797 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2798 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2799 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2800 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2801 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2802 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2803 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2804 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2805 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2806 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2807 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2808 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2809 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2810 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2811 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2812 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2813 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2814 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2815 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2816 18657, 18676, 18694, 18696, 18887.
2818 * Cache information can be queried via sysconf() function on s390 e.g. with
2819 _SC_LEVEL1_ICACHE_SIZE as argument.
2821 * A buffer overflow in gethostbyname_r and related functions performing DNS
2822 requests has been fixed. If the NSS functions were called with a
2823 misaligned buffer, the buffer length change due to pointer alignment was
2824 not taken into account. This could result in application crashes or,
2825 potentially arbitrary code execution, using crafted, but syntactically
2826 valid DNS responses. (CVE-2015-1781)
2828 * The time zone file parser has been made more robust against crafted time
2829 zone files, avoiding heap buffer overflows related to the processing of
2830 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2831 large time zone data files. Overly long time zone specifiers in the TZ
2832 variable no longer result in stack overflows and crashes.
2834 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2835 for LD and GD on x86 and x86-64, has been implemented. You will need
2836 binutils-2.24 or later to enable this optimization.
2838 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2839 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2840 Hat). These updates cause user visible changes, such as the fix for bug
2843 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2844 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2845 condition in some applications.
2847 * Added vector math library named libmvec with the following vectorized x86_64
2848 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2850 The library can be disabled with --disable-mathvec. Use of the functions is
2851 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2852 Shared library libmvec.so is linked in as needed when using -lm (no need to
2853 specify -lmvec explicitly for not static builds).
2854 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2856 * A new fmemopen implementation has been added with the goal of POSIX
2857 compliance. The new implementation fixes the following long-standing
2858 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2859 old implementation is still present for use be by existing binaries.
2861 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2862 releases. It has been fixed to match 2.19 and older, but binaries built
2863 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2865 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2866 Contributed by Roland McGrath (Google).
2868 * The header <regexp.h> is deprecated, and will be removed in a future
2869 release. Use of this header will trigger a deprecation warning.
2870 Application developers should update their code to use <regex.h> instead.
2872 This header was formerly part of SUS, but was deprecated in 1994 and
2873 removed from the standard in 2001. Also, the glibc implementation
2874 leaks memory. See BZ#18681 for more details.
2878 * The following bugs are resolved with this release:
2880 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2881 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2882 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2883 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2884 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2885 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2886 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2887 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2888 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2889 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2892 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2893 too little memory for the to-be-scanned arguments and overflow the
2894 allocated buffer. The implementation now correctly computes the required
2895 buffer size when using malloc, and switches to malloc from alloca as
2898 * A new semaphore algorithm has been implemented in generic C code for all
2899 machines. Previous custom assembly implementations of semaphore were
2900 difficult to reason about or ensure that they were safe. The new version
2901 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2902 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2903 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2905 * Port to Altera Nios II has been contributed by Mentor Graphics.
2907 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2908 implementations for powerpc64/powerpc64le.
2909 Implemented by Adhemerval Zanella (IBM).
2911 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2912 and powerpc64le. This may improve lock scaling of existing programs on
2913 HTM capable systems. The lock elision code is only enabled with
2914 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2915 powerpc will issue a transaction abort on every syscall to avoid side
2916 effects being visible outside transactions.
2918 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2919 AArch64. Contributed by ARM Ltd.
2921 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2923 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2924 under certain input conditions resulting in the execution of a shell for
2925 command substitution when the application did not request it. The
2926 implementation now checks WRDE_NOCMD immediately before executing the
2927 shell and returns the error WRDE_CMDSUB as expected.
2929 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2930 processing format strings with a large number of format specifiers.
2932 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2933 infinite loop if the DNS response contained a PTR record of an unexpected
2936 * The minimum GCC version that can be used to build this version of the GNU
2937 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2938 still be used to compile programs using the GNU C Library.
2940 * The GNU C Library is now built with -Werror by default. This can be
2941 disabled by configuring with --disable-werror.
2943 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2945 * The obsolete sigvec function has been removed. This was the original
2946 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2947 programs have been using instead for about 25 years. Of course, ABI
2948 compatibility for old binaries using sigvec remains intact.
2950 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2951 with newer versions of bison.
2953 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2954 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2955 registers overlay two consecutive single-precision registers. MIPS32R2
2956 introduced a new FPU mode (FR=1) where double-precision registers extend the
2957 corresponding single-precision registers which is incompatible with the
2958 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2959 require the use of FR=1 making a transition necessary. New o32 ABI
2960 extensions enable users to migrate over time from the original o32 ABI
2961 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2962 tracks the ABI of any loaded object and verifies that new objects are
2963 compatible. Mode transitions will also be requested as required and
2964 unsupportable objects will be rejected. The ABI checks include both soft and
2965 hard float ABIs for o32, n32 and n64.
2967 GCC 5 with GNU binutils 2.25 onwards:
2968 It is strongly recommended that all o32 system libraries are built using the
2969 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2970 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2971 compiler using --with-fp-32=xx to set this by default.
2975 * The following bugs are resolved with this release:
2977 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2978 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2979 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2980 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2981 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2982 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2983 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2984 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2985 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2986 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2987 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2988 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2989 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2990 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2991 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2993 * Reverted change of ABI data structures for s390 and s390x:
2994 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2995 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2996 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2997 are preserved pointing straight to the same implementation as the old ones.
2998 Given that, new callers will simply provide a too-big buffer to these
2999 functions. Any applications/libraries out there that embed jmp_buf or
3000 ucontext_t in an ABI-relevant data structure that have already been rebuilt
3001 against 2.19 headers will have to rebuilt again. This is necessary in any
3002 case to revert the breakage in their ABI caused by the glibc change.
3004 * Support for file description locks is added to systems running the
3005 Linux kernel. The standard file locking interfaces are extended to
3006 operate on file descriptions, not file descriptors, via the use of
3007 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
3008 are associated with an open file instead of a process.
3010 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
3012 * The minimum Linux kernel version that this version of the GNU C Library
3013 can be used with is 2.6.32.
3015 * Running the testsuite no longer terminates as soon as a test fails.
3016 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
3017 with PASS or FAIL lines for individual tests. A summary of the results is
3018 printed, including a list of failing lists, and "make check" exits with
3019 error status if there were any unexpected failures. "make check
3020 stop-on-test-failure=y" may be used to keep the old behavior.
3022 * The am33 port, which had not worked for several years, has been removed
3025 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
3026 supported; they now act the same as _DEFAULT_SOURCE (but generate a
3027 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
3028 conflicted with POSIX (support for which was removed in 2.19), the
3029 interfaces those macros enabled remain available when compiling with
3030 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
3031 test macros defined.
3033 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
3035 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
3036 This may improve lock scaling of existing programs on TX capable systems.
3037 The lock elision code is only built with --enable-lock-elision=yes and
3038 then requires a GCC version supporting the TX builtins. With lock elision
3039 default mutexes are elided via __builtin_tbegin, if the cpu supports
3040 transactions. By default lock elision is not enabled and the elision code
3043 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
3044 copy the path argument. This allowed programs to cause posix_spawn to
3045 deference a dangling pointer, or use an unexpected pathname argument if
3046 the string was modified after the posix_spawn_file_actions_addopen
3049 * All supported architectures now use the main glibc sysdeps directory
3050 instead of some being in a separate "ports" directory (which was
3051 distributed separately before glibc 2.17).
3053 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
3054 On configurations that support it (all Linux configurations), it's now
3055 used regardless of the --enable-add-ons switch to configure. It is no
3056 longer possible to build such configurations without pthreads support.
3058 * Locale names, including those obtained from environment variables (LANG
3059 and the LC_* variables), are more tightly checked for proper syntax.
3060 setlocale will now fail (with EINVAL) for locale names that are overly
3061 long, contain slashes without starting with a slash, or contain ".." path
3062 components. (CVE-2014-0475) Previously, some valid locale names were
3063 silently replaced with the "C" locale when running in AT_SECURE mode
3064 (e.g., in a SUID program). This is no longer necessary because of the
3067 * On x86-64, the dynamic linker's lazy-binding support is now compatible
3068 with application code using Intel MPX instructions. (With all previous
3069 versions, the MPX register state could be clobbered when making calls
3070 into or out of a shared library.) Note that while the new dynamic
3071 linker is compatible with all known x86 hardware whether or not it
3072 supports Intel MPX, some x86 instruction-set emulators might fail to
3073 handle the new instruction encodings. This is known to affect Valgrind
3074 versions up through 3.9 (but will be fixed in the forthcoming 3.10
3075 release), and might affect other tools that do instruction emulation.
3077 * Support for loadable gconv transliteration modules has been removed.
3078 The support for transliteration modules has been non-functional for
3079 over a decade, and the removal is prompted by security defects. The
3080 normal gconv conversion modules are still supported. Transliteration
3081 with //TRANSLIT is still possible, and the //IGNORE specifier
3082 continues to be supported. (CVE-2014-5119)
3084 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
3085 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
3086 resulting a denial-of-service security vulnerability in applications which
3087 use functions related to iconv. (CVE-2014-6040)
3091 * The following bugs are resolved with this release:
3093 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
3094 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
3095 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
3096 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
3097 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
3098 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
3099 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
3100 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
3101 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
3102 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
3103 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
3104 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
3105 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
3106 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
3107 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
3108 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
3109 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
3110 16431, 16453, 16474, 16506, 16510, 16529
3112 * Slovenian translations for glibc messages have been contributed by the
3113 Translation Project's Slovenian team of translators.
3115 * The public headers no longer use __unused nor __block. This change is to
3116 support compiling programs that are derived from BSD sources and use
3117 __unused internally, and to support compiling with Clang's -fblock
3118 extension which uses __block.
3120 * CVE-2012-4412 The strcoll implementation caches indices and rules for
3121 large collation sequences to optimize multiple passes. This cache
3122 computation may overflow for large collation sequences and may cause a
3123 stack or buffer overflow. This is now fixed to use a slower algorithm
3124 which does not use a cache if there is an integer overflow.
3126 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
3127 rules for large collation sequences to optimize multiple passes and falls
3128 back to alloca if malloc fails, resulting in a possible stack overflow.
3129 The implementation now falls back to an uncached collation sequence lookup
3132 * CVE-2013-4788 The pointer guard used for pointer mangling was not
3133 initialized for static applications resulting in the security feature
3134 being disabled. The pointer guard is now correctly initialized to a
3135 random value for static applications. Existing static applications need
3136 to be recompiled to take advantage of the fix (bug 15754).
3138 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
3139 to the d_name member of struct dirent, or omit the terminating NUL
3140 character. (Bugzilla #14699).
3142 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
3143 aligned_alloc functions could allocate too few bytes or corrupt the
3144 heap when passed very large allocation size values (Bugzilla #15855,
3147 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
3148 for AF_INET6 has been fixed (Bugzilla #16072).
3150 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
3151 pap_CW, quz_PE, the_NP.
3153 * Substantially revised locales: gd_GB, ht_HT
3155 * The LC_ADDRESS field was updated to support country_car for almost all
3158 * ISO 1427 definitions were updated.
3160 * ISO 3166 definitions were updated.
3162 * The localedef utility now supports --big-endian and --little-endian
3163 command-line options to generate locales for a different system from that
3164 for which the C library was built.
3166 * Binary locale files now only depend on the endianness of the system for
3167 which they are generated and not on other properties of that system. As a
3168 consequence, binary files generated with new localedef may be incompatible
3169 with old versions of the GNU C Library, and binary files generated with
3170 old localedef may be incompatible with this version of the GNU C Library,
3171 in the following circumstances:
3173 + Locale files may be incompatible on m68k systems.
3175 + Locale archive files (but not separate files for individual locales) may
3176 be incompatible on systems where plain "char" is signed.
3178 * The configure option --disable-versioning has been removed. Builds with
3179 --disable-versioning had not worked for several years.
3181 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
3182 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
3184 * SystemTap probes for malloc have been introduced.
3186 * SystemTap probes for slow multiple precision fallback paths of
3187 transcendental functions have been introduced.
3189 * Support for powerpc64le has been added.
3191 * The soft-float powerpc port now supports e500 processors.
3193 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
3195 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
3196 set of header declarations that are enabled by default, even when other
3197 feature test macros or compiler options such as -std=c99 would otherwise
3198 disable some of those declarations.
3200 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
3201 conflict with POSIX. The libbsd-compat library (which was a dummy library
3202 that did nothing) has also been removed.
3204 * Preliminary documentation about Multi-Thread, Async-Signal and
3205 Async-Cancel Safety has been added.
3207 * Change of ABI data structures for s390 and s390x:
3208 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
3209 allow for future hardware extensions. All library functions that accept or
3210 return these structures were versioned in order to provide backwards
3211 compatibility with existing code. However, not all cases can be handled
3212 with interface versioning. If an application embeds either structure into
3213 another structure and that structure is passed to another compilation unit
3214 which expects the newer larger-sized object then any access to the new
3215 fields will result in undefined behaviour. Similarly any access to fields
3216 that were moved by the enlarging of the structures will cause undefined
3217 behaviour. To fix the undefined behaviour all objects that are part of the
3218 final in-memory image must agree on the size of structures, and this may
3219 require recompilation.
3223 * The following bugs are resolved with this release:
3225 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
3226 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
3227 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
3228 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
3229 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3230 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3231 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3232 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3233 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3234 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3235 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3236 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3237 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3240 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3241 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3242 Distributions can re-enable building and using pt_chown via the new configure
3243 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3244 considerable security risks and should only be used if the distribution
3245 understands and accepts the risks.
3247 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3250 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3251 fixed (Bugzilla #15330).
3253 * Add support for calling C++11 thread_local object destructors on thread
3254 and program exit. This needs compiler support for offloading C++11
3255 destructor calls to glibc.
3257 * Improved worst case performance of libm functions with double inputs and
3260 * Support for priority inherited mutexes in pthread condition variables on
3261 non-x86 architectures.
3263 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3265 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3267 * Optimized string functions for ARM. Implemented by Will Newton and
3270 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3272 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3275 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3276 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3278 * Added a benchmark framework to track performance of functions in glibc.
3280 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3281 It is based on draft TS 18661 and currently enabled as a GNU extension.
3283 * On Linux, the clock function now uses the clock_gettime system call
3284 for improved precision, rather than old times system call.
3286 * Added support for version-3 tz format files. This is needed when using
3287 the tz database release 2013e or later, and affects a few unusual cases --
3288 currently only TZ='America/Godthab' for time stamps after 2037.
3290 * Added new API functions pthread_getattr_default_np and
3291 pthread_setattr_default_np to get and set the default pthread
3292 attributes of a process.
3294 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3295 This may improve lock scaling of existing programs on TSX capable systems.
3296 When the --enable-lock-elision=yes parameter is specified at configure
3297 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3300 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3301 availability). Implemented by Ryan S. Arnold.
3303 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3307 * The following bugs are resolved with this release:
3309 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3310 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3311 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3312 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3313 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3314 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3315 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3316 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3317 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3318 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3319 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3320 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3321 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3322 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3323 14879, 14889, 14893, 14898, 14914.
3325 * Optimization of memcpy for MIPS.
3327 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3328 EMFILE has been fixed (Bugzilla #14889).
3330 * The add-on ports collection is now distributed in the "ports" subdirectory
3331 of the main GNU C Library distribution, rather than separately.
3333 * Port to ARM AArch64 contributed by Linaro.
3335 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3336 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3338 Implemented by Andreas Krebbel.
3340 * The new function secure_getenv allows secure access to the environment,
3341 returning NULL if running in a SUID/SGID process. This function replaces
3342 the internal function __secure_getenv.
3344 * SystemTap static probes have been added into the dynamic linker.
3345 Implemented by Gary Benson.
3347 * Optimizations of string functions strstr, strcasestr and memmem.
3348 Implemented by Maxim Kuvyrkov.
3350 * The minimum Linux kernel version that this version of the GNU C Library
3351 can be used with is 2.6.16.
3353 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3354 powerpc POWER7. Implemented by Will Schmidt.
3356 * New configure option --disable-nscd builds the C library such that it
3357 never attempts to contact the Name Service Caching Daemon (nscd).
3358 New configure option --disable-build-nscd avoids building nscd itself;
3359 this is the default if --disable-nscd is used.
3361 * Improved support for cross-compilation, including cross-testing and
3362 bootstrap builds without a previously built glibc.
3364 * Several testsuite tests are now able to test multiple IFUNC variants of an
3365 interface, rather than just testing the one that would be chooen by
3368 * New configure options --with-bugurl and --with-pkgversion, for
3369 distributors to use to embed their bug-reporting and package version
3370 information in --help and --version output.
3372 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3373 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3374 allows creation of chroots without the procfs mounted on /proc.
3376 * The `crypt' function now fails if passed salt bytes that violate the
3377 specification for those values. On Linux, the `crypt' function will
3378 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3379 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3380 when the mode is enabled.
3382 * The `clock_*' suite of functions (declared in <time.h>) is now available
3383 directly in the main C library. Previously it was necessary to link with
3384 -lrt to use these functions. This change has the effect that a
3385 single-threaded program that uses a function such as `clock_gettime' (and
3386 is not linked with -lrt) will no longer implicitly load the pthreads
3387 library at runtime and so will not suffer the overheads associated with
3388 multi-thread support in other code such as the C++ runtime library.
3390 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3395 * The following bugs are resolved with this release:
3397 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3398 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3399 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3400 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3401 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3402 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3403 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3404 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3405 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3406 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3407 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3408 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3409 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3410 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3411 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3412 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3413 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3414 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3415 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3416 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3417 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3418 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3421 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3422 configuring glibc with:
3423 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3424 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3425 Implemented by H.J. Lu.
3429 + define static_assert
3431 + do not declare gets
3433 + declare at_quick_exit and quick_exit also for ISO C11
3435 + aligned_alloc. NB: The code is deliberately allows the size parameter
3436 to not be a multiple of the alignment. This is a moronic requirement
3437 in the standard but it is only a requirement on the caller, not the
3440 + timespec_get added
3442 + uchar.h support added
3444 + CMPLX, CMPLXF, CMPLXL added
3446 Implemented by Ulrich Drepper.
3448 * Support for the IA-64 has been moved to ports.
3450 * Remove support for anything but ELF binary format
3452 * Checking versions of poll, ppoll added.
3453 Implemented by Ulrich Drepper.
3455 * More generic and 64-bit performance optimizations to math functions.
3456 Implemented by Ulrich Drepper.
3458 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3459 headers and functions available at compile time as they were before
3460 version 2.14. This option will be removed at some time in the future
3461 after the TI-RPC library becomes fully sufficient for the needs of
3462 existing applications.
3464 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3465 Note that glibc is not expected to work with any Linux kernel version
3468 * New header <sys/auxv.h> and function getauxval allowing easy access to
3469 the AT_* key-value pairs passed from the Linux kernel. The header also
3470 defines the HWCAP_* bits associated with the AT_HWCAP key.
3472 * New locales: mag_IN
3474 * New configure option --enable-systemtap builds SystemTap static probes
3475 into libc for setjmp and longjmp and into libpthread for various operations.
3476 So far the setjmp/longjmp probes and some of the libpthread probes are
3477 provided only for i*86 and x86_64.
3478 Implemented by Roland McGrath and Rayson Ho.
3480 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3482 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3485 * More optimized functions for SPARC. Implemented by David S. Miller.
3487 * Improved support for cross-compilation, especially bootstrap builds
3488 without a previously built glibc.
3490 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3491 Chris Metcalf from Tilera.
3493 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3494 now supported for ARM processors.
3496 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3497 as the name of the dynamic linker, to distinguish it from the
3498 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3500 * Support for CRIS has been removed from ports.
3502 * A new class of installed header has been documented for low-level
3503 platform-specific functionality. PowerPC added the first instance with a
3504 function to provide time base register access. Contributed by Tulio
3505 Magno Quites Machado Filho.
3507 * ix86 configurations now install header files that are consistent with
3508 what x86-64 configurations install. These same header files can be used
3509 for -m32, -m64, or -mx32 builds.
3510 Contributed by H.J. Lu.
3512 * Math library bug fixes. A thorough audit of all open math library bugs was
3513 conducted by Joseph Myers. Significant progress was made on many math
3514 library bugs resulting in more accurate exceptions and function results.
3515 Many thanks to all those that contributed including Andreas Jaeger for his
3516 patch review and work on the x87 trigonometric instruction issues.
3518 * Timezone data is no longer installed. Timezone-related binaries and scripts
3519 will continue to be installed. Users should obtain their timezone data from
3520 their distribution provider or from the tzdata package at
3521 <ftp://munnari.oz.au/pub/>.
3525 * The following bugs are resolved with this release:
3527 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3528 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3529 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3530 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3531 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3532 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3533 13484, 13506, 13515, 13523, 13524, 13538, 13540
3535 * New program pldd to list loaded object of a process
3536 Implemented by Ulrich Drepper.
3538 * Add nss_db support back to glibc. No more dependency on Berkeley db
3539 and support for initgroups lookups.
3540 Implemented by Ulrich Drepper.
3542 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3543 Contributed by HJ Lu.
3545 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3546 Contributed by HJ Lu.
3548 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3549 on x86-32 and x86-64.
3550 Contributed by Liubov Dmitrieva.
3552 * Optimized strchr and strrchr for SSE on x86-32.
3553 Contributed by Liubov Dmitrieva.
3555 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3556 for x86-64 and x86-32.
3557 Contributed by Liubov Dmitrieva.
3559 * New interfaces: scandirat, scandirat64
3560 Implemented by Ulrich Drepper.
3562 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3563 Implemented by Ulrich Drepper.
3565 * nscd now also caches the netgroup database.
3566 Implemented by Ulrich Drepper.
3568 * Integrate libm with gcc's -ffinite-math-only option.
3569 Implemented by Ulrich Drepper.
3571 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3572 to math functions. Implemented by Ulrich Drepper.
3574 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3575 Implemented by Ulrich Drepper.
3577 * New Linux interfaces: process_vm_readv, process_vm_writev
3579 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3580 Implemented by Ulrich Drepper.
3582 * Optimized nearbyint and strcasecmp for PPC.
3583 Implemented by Adhemerval Zanella.
3585 * New locales: bho_IN, unm_US, es_CU, ta_LK
3589 * The following bugs are resolved with this release:
3591 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3592 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3593 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3594 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3595 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3596 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3597 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3598 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3599 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3600 12795, 12811, 12813, 12814, 12841
3602 * The RPC implementation in libc is obsoleted. Old programs keep working
3603 but new programs cannot be linked with the routines in libc anymore.
3604 Programs in need of RPC functionality must be linked against TI-RPC.
3605 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3607 Visible changes of this change include (obviously) the inability to link
3608 programs using RPC functions without referencing the TI-RPC library and the
3609 removal of the RPC headers from the glibc headers.
3610 Implemented by Ulrich Drepper.
3612 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3613 syncfs, setns, sendmmsg
3615 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3616 yue_HK, lij_IT, mhr_RU
3618 * New encodings: CP770, CP771, CP772, CP773, CP774
3620 * New program sotruss to trace calls through PLTs
3621 Implemented by Ulrich Drepper.
3623 * The malloc hook implementation is marked deprecated and will be removed
3624 from the default implementation in the next version. The design never
3625 worked ever since the introduction of threads. Even programs which do
3626 not create threads themselves can use multiple threads created internally.
3630 * The following bugs are resolved with this release:
3632 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3633 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3634 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3635 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3638 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3640 * POWER7 optimizations: memset, memcmp, strncmp
3642 * New optimized string functions for x86-64: strnlen (SSE2),
3643 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3644 Implemented by Ulrich Drepper.
3648 * The following bugs are resolved with this release:
3650 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3651 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3652 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3653 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3654 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3655 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3656 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3657 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3659 * New interfaces: pthread_getname_np, pthread_setname_np
3661 * New Linux interface: recvmmsg
3663 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3665 * The dynamic linker now recognizes supported ABI versions from the
3666 EI_ABIVERSION field in the ELF header.
3667 Implemented by Ulrich Drepper.
3669 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3670 will not be used to fill in password fields in the passwd.byname replies.
3671 Instead it is used to synthesize the shadow.byname table, should it be
3672 missing. This is a useful mode in some installations involving Solaris.
3673 Implemented by Ulrich Drepper.
3675 * New locales: kok_IN, sq_MK, cv_RU
3679 * The following bugs are resolved with this release:
3681 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3682 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3683 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3684 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3685 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3686 10789, 10817, 10823, 10840, 10847
3688 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3690 Implemented by Ulrich Drepper.
3692 * Checking version of longjmp added that fails if an uninitialized stack
3693 frame would be created. Implemented by Ulrich Drepper.
3695 * STT_GNU_IFUNC is now supported in static executables.
3696 Implemented by H.J. Lu.
3698 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3700 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3701 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3702 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3703 Contributed by H.J. Lu.
3705 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3706 Implemented by Ulrich Drepper.
3708 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3710 Contributed by H.J. Lu.
3712 * Support for fma instruction in AVX on x86-64.
3713 Implemented by H.J. Lu and Ulrich Drepper.
3715 * AVX support in x86-64 auditing support in ld.so.
3716 Implemented by H.J. Lu.
3718 * STB_GNU_UNIQUE support added.
3719 Implemented by Ulrich Drepper.
3721 * Implement second fallback mode for optimized DNS lookup for even more
3722 broken environments. If two requests from the same port are not
3723 handled correctly close the socket and open a new one before sending
3724 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3725 can be used to select this mode right away, instead of rediscovering the
3726 necessity is every process again.
3727 Implemented by Ulrich Drepper.
3729 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3730 Implemented by Adam Tkac.
3732 * Optimized iconv conversions for S390x.
3733 Implemented by Andreas Krebbel.
3735 * Using condvars with PI mutexes is now more efficient due to kernel
3736 support for requeueing to PI futexes. NPTL support added for x86-64.
3737 Implemented by Ulrich Drepper.
3739 * New locales: ps_AF, my_MM
3743 * The following bugs are resolved with this release:
3745 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3746 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3747 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3748 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3749 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3751 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3752 Implemented by Ulrich Drepper.
3754 * New Linux interfaces: accept4, fallocate, fallocate64.
3755 Implemented by Ulrich Drepper.
3757 * Correct declarations of string function when used in C++ code. This
3758 could lead to compile errors for invalid C++ code.
3760 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3763 * New POSIX 2008 interface: psiginfo
3764 Implemented by Ulrich Drepper.
3766 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3767 Implemented by Ulrich Drepper.
3769 * Support for selecting between multiple function definitions at runtime
3770 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3772 * The libcrypt library can now use the hash function implementations in
3773 NSS. Implemented by Ulrich Drepper.
3775 * The malloc implementation can be compiled to be less memory efficient
3776 but higher performing in multi-threaded programs.
3777 Implemented by Ulrich Drepper.
3779 * New locales: nan_TW@latin, ks_IN
3781 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3782 Implemented by Ulrich Drepper.
3784 * Extended printf hook support. It is possible to use user-defined types
3785 and extend existing format specifiers.
3786 Implemented by Ulrich Drepper.
3788 * Handling for group shadow files has been added.
3789 Implemented by Ulrich Drepper.
3791 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3792 servers (the case, e.g., for some people using the built-in DNS
3793 server in ADSL modems/routers). There is a once-per-process timeout
3794 in case of a broken server. To avoid it, users can run nscd or put
3795 'options single-request' in /etc/resolv.conf.
3796 Implemented by Ulrich Drepper.
3800 * The following bugs are resolved with this release:
3802 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3803 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3804 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3805 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3806 6968, 6974, 6980, 6995, 7008, 7009, 7029
3808 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3809 up at the same time. Implemented by Ulrich Drepper.
3811 * TLS descriptors for LD and GD on x86 and x86-64.
3812 Implemented by Alexandre Oliva.
3814 * getaddrinfo now handles DCCP and UDPlite.
3815 Implemented by Ulrich Drepper.
3817 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3818 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3819 Implemented by Ulrich Drepper.
3821 * New implementation of memmem, strstr, and strcasestr which is O(n).
3822 Implemented by Eric Blake.
3824 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3826 * Implement "e" option for popen to open file descriptor with the
3827 close-on-exec flag set. Implemented by Ulrich Drepper.
3829 * Implement "b" mode for fmemopen. In this mode writes writes don't
3830 implicitly add a NUL byte and seeks from the end of the buffer really
3831 use the buffer end, not the string length as the basis.
3832 Implemented by Ulrich Drepper.
3834 * Many functions, exported and internal, now atomically set the close-on-exec
3835 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3837 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3839 Implemented by Pravin Satpute.
3841 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3845 * The following bugs are resolved with this release:
3847 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3848 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3849 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3850 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3851 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3852 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3853 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3855 * New locales: bo_CN, bo_IN, shs_CA.
3857 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3859 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3860 Implemented by Pravin Satpute.
3862 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3864 * nscd caches now all timeouts for DNS entries
3865 Implemented by Ulrich Drepper.
3867 * nscd is more efficient and wakes up less often.
3868 Implemented by Ulrich Drepper.
3870 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3871 vdprintf, and obstack_vprintf.
3872 Implemented by Jakub Jelinek.
3874 * Faster memset for x86-64.
3875 Implemented by Harsha Jagasia and H.J. Lu.
3877 * Faster memcpy on x86.
3878 Implemented by Ulrich Drepper.
3880 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3881 Implemented by Ulrich Drepper.
3883 * Faster sqrt and sqrtf implemention for some PPC variants.
3884 Implemented by Steven Munroe.
3888 * The following bugs are resolved with this release:
3890 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3891 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3892 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3893 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3894 5071, 5103, 5104, 5112, 5113, 5184, 5186
3896 * More checking functions: fread, fread_unlocked, open*, mq_open.
3897 Implemented by Jakub Jelinek and Ulrich Drepper.
3899 * Extend fortification to C++. Implemented by Jakub Jelinek.
3901 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3902 by not recognizing 'a' as a modifier when those specs are requested.
3903 Implemented by Jakub Jelinek.
3905 * PPC optimizations to math and string functions.
3906 Implemented by Steven Munroe.
3908 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3909 options to be passed. Implemented by Ulrich Drepper.
3911 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3913 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3914 Implemented by Ulrich Drepper.
3916 * Handle private futexes in the NPTL implementation.
3917 Implemented by Jakub Jelinek and Ulrich Drepper.
3919 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3920 Implemented by Roland McGrath and Ulrich Drepper.
3922 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3924 * SHA-256 and SHA-512 based password encryption.
3925 Implemented by Ulrich Drepper.
3927 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3928 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3931 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3932 Implemented by Ulrich Drepper.
3936 * The following bugs are resolved with this release:
3938 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3939 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3940 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3941 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3942 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3943 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3944 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3947 * New Linux interfaces: epoll_pwait, sched_getcpu.
3949 * New generic interfaces: strerror_l.
3951 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3955 * The following bugs are resolved with this release:
3957 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3958 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3959 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3960 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3961 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3962 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3963 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3964 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3965 3225, 3251, 3252, 3253, 3273
3967 * For Linux, the sorting of addresses returned by getaddrinfo now also
3968 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3969 Implemented by Ulrich Drepper.
3971 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3972 Implemented by Ulrich Drepper.
3974 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3976 * New iconv module for MIK. Contributed by Alexander Shopov.
3978 * For sites with broken group and/or passwd database, the auto-propagate
3979 option of nscd can prevent creating ID lookup entries from the results
3980 of a name lookup and vice versa. This usually is no problem but some
3981 site might have problems with the default behavior.
3982 Implemented by Ulrich Drepper.
3984 * Iterating over entire database in NIS can be slow. With the
3985 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3986 to trade time for memory. The entire database will be read at once.
3987 Implemented by Ulrich Drepper.
3989 * The interfaces introduced in RFC 3542 have been implemented by
3992 * Support for the new ELF hash table format was added by Ulrich Drepper.
3994 * Support for priority inheritance mutexes added by Jakub Jelinek and
3997 * Support for priority protected mutexes added by Jakub Jelinek.
3999 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
4000 or_IN, csb_PL, fy_NL, sr_ME.
4004 * More overflow detection functions.
4006 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
4007 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
4009 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
4010 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
4011 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
4012 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
4013 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
4014 by Masahide Washizawa.
4016 * It is now possible to install audit libraries for the dynamic linker using
4017 LD_AUDIT. Implemented by Ulrich Drepper.
4019 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
4020 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
4021 For a libc and libpthread that works well on Linux 2.4 kernels, we
4022 recommend using the stable 2.3 branch.
4024 * The new function open_wmemstream acts like open_memstream,
4025 but uses a wchar_t wide character buffer.
4027 * The new function ppoll is an improved version of poll, similar to pselect.
4029 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
4030 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
4031 renameat, unlinkat, linkat, symlinkat, readlinkat.
4033 * New Linux kernel system calls: unshare,
4034 inotify_init, inotify_add_watch, inotify_rm_watch.
4036 * The euidaccess function is now also known by the alias eaccess,
4037 for compatibility with some other systems.
4039 * Timezone data updated to 2006b version.
4043 * The following bugs are resolved with this release:
4045 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
4046 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
4047 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
4048 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
4049 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
4050 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
4052 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4054 * As of this release, GCC 4 can be used to compile the C Library.
4056 * Timezone data updated to 2005m version.
4060 * The following bugs are resolved with this release:
4062 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
4063 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
4064 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
4065 777, 787, 821, 822, 823, 825
4067 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4071 * Support for RFC 3678. Real implementations exist only for Linux so far.
4072 Implemented by Ulrich Drepper.
4074 * nscd can now cache entries persistently. Expiring entries are reloaded.
4075 For speedups the cache can be shared in memory with client processes.
4076 Implemented by Ulrich Drepper.
4078 * nscd can now perform SELinux checks.
4079 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
4081 * getaddrinfo queries are now cached. Canonical name lookup is performed
4083 Implemented by Ulrich Drepper.
4085 * The nothrow function attribute is used when headers are used by gcc when
4086 compiling C code. This can avoid significant amounts of exception
4089 * The malloc functions perform more error checking and are stricter when
4090 it comes to reacting on errors. The default action is to terminate
4091 the process after showing an error message. Implemented by Ulrich Drepper.
4093 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
4094 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
4096 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
4097 namespaces using the new function dlmopen(). This feature is of course,
4098 like most other dynamic loading functionality, not available in statically
4099 linked applications. Implemented by Ulrich Drepper.
4101 * Low-overhead boundary checking variants of string and some stdio functions
4102 were added. These are to be used in conjunction with a gcc patch by
4103 Jakub Jelinek which adds calls to these functions if possible.
4104 Implemented by Jakub Jelinek and Ulrich Drepper.
4106 * Old code for several operating systems and machine architectures that
4107 have not been in working condition in a long time have been removed from
4108 the main source tree maintained by the GNU C Library's maintainers.
4109 These files are now reside in the separate `ports' source module
4110 that is usable as an add-on when building the library.
4114 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
4115 interrogate the dynamic linker, compatible with the Solaris interface.
4117 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
4118 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
4120 * getifaddrs now uses the netlink interface on Linux to get its information.
4121 Implemented by Thorsten Kukuk.
4123 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
4124 Implemented by Ulrich Drepper.
4126 * support for non-executable stacks on x86 has been added. Changes mostly
4129 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
4132 * getaddrinfo now performs destination address selection according to
4137 * Thread-safe interfaces for many functions that access locale data
4138 were added in version 2.3, but these features were omitted from NEWS.
4139 Many functions have variants with an `_l' suffix that take a `locale_t'
4140 object as a parameter rather than consulting the current locale.
4141 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
4142 create and maintain `locale_t' objects. Additionally, the new function
4143 `uselocale' sets "the current locale" (as used by functions not so
4144 parameterized) set for an individual thread. These features were added
4145 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
4147 * The functions getresuid, getresgid, setresuid, and setresgid, which
4148 have long been available on Linux, are now declared in <unistd.h>
4149 and are now also available on the Hurd.
4151 * ELF thread-local storage support (TLS) now works on x86-64.
4153 * The new dynamic string token $LIB is expanded in shared library names.
4154 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
4156 * Aldy Hernandez contributed complete software floating point support for
4157 PowerPC machines with no FPU.
4159 * fexecve is implemented on Linux.
4161 * The `btowc' function should work at least twice as fast due to
4162 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
4164 * With approriate thread add-ons cancelable functions are now implemented
4165 in libc.so as well. No need to call the function in libpthread. This
4166 change allowed to finally disable the incorrect and expensive handling
4167 of weak definition in ld.so.
4169 * Yet more PLT entries in libc.so have been removed. We finally arrived
4170 at the bare minimum. Startup times improved appropriately.
4172 * Support for the new Linux/x86 system call interface was added. The
4173 AT_SYSINFO auxiliary vector entry is recognized and handled.
4177 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
4180 * iconv (the program and the interface) now accepts empty names (excluding
4181 options like //TRANSLIT) to mean "use charset of current locale".
4183 * localedef can now transliterate characters in strings which are not in
4184 the provided charmap. The information from the input locale is used.
4186 * Prelinking support was added for ELF targets. This requires additional
4187 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
4189 * Read-only stdio streams now use mmap to speed up operation by eliminating
4190 copying and buffer underflows. To use add 'm' to the mode string of
4191 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
4193 * The malloc functions were completely rewritten by Wolfram Gloger based
4194 on Doug Lea's malloc-2.7.0.c.
4196 * Isamu Hasegawa contributed a completely new and POSIX-conformant
4197 implementation of regex.
4199 * Bruno Haible upgraded the iconv and locale implementation to support
4202 * Contents of the LC_* and LANG environment variables in the CEN style are
4203 not recognized anymore. It never was used. Change by Ulrich Drepper.
4205 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
4206 thread-local storage (TLS) ABI on some platforms.
4207 Changes by Ulrich Drepper. SH support by Kaz Kojima.
4209 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
4210 EUC-JISX0213, and TSCII.
4212 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
4213 BSD-compatible interface for getting all network interface addresses.
4214 Implementation for IPv4 by Roland McGrath.
4216 * Loading of locale data is faster due to the introduction of a locale
4217 archive. Implemented by Roland McGrath and Ulrich Drepper.
4219 * Startup times are significantly reduced by not using exported functions
4220 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
4223 * Steven Munroe contributed a port to PowerPC64/Linux.
4227 * The Hurd now uses the GNU libio implementation of stdio rather than the
4228 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
4230 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4231 with the same behavior as the Linux system call of the same name.
4235 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4236 128-bit long double format.
4238 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4239 IBM1161, and IBM1162 charsets.
4241 * Andreas Jaeger contributed a port to x86-64/Linux.
4243 * Peter Bruin contributed a port to PowerPC/Hurd.
4245 * libc functions using I/O streams now can handle wide-oriented streams
4248 * optimizations in the dynamic linker. Binaries created by recent binutils
4249 versions start up quicker due to reduced time spent on relocations.
4251 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4255 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4256 asinh, atanh, j0 for the 128-bit long double format.
4258 * Bruno Haible updated all the code handling Unicode in some form to
4259 support Unicode 3.1.
4261 * Speed of regex for single-byte locales is back to previous levels.
4262 Patch by Isamu Hasegawa.
4264 * Alpha, SPARC, and IA-64 now also using floating stacks.
4266 * Startup time of internationalized applications greatly improved through
4267 iconv cache. Use iconvconfig to generate the cache file.
4268 Contributed by Ulrich Drepper.
4270 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4271 severe performance problems. Patch by David Mosberger.
4273 * The Hurd port got a lot more functionality like AIO, various stdio
4274 extensions, etc. Mainly done by Roland McGrath.
4276 * mtrace can now lookup symbols in shared libraries.
4280 * Intel's IA-64 math library is largely integrated. It provides fast and
4281 accurate implementatations for most basic and standard math functions
4282 in float, double, and long double format.
4284 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4285 and asin for the 96-bit long double format and asin, log, tan for the
4286 128-bit long double format.
4288 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4289 The basic double functions exist today. Contributed by Abraham Ziv
4290 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4291 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4293 * An asynchronous name lookup library was added. The interface is designed
4294 after POSIX AIO. The proposal was circulated beforehand to get comments.
4295 No negative ones came in. Implemented by Ulrich Drepper.
4297 * Port to S390/64bit contributed by Martin Schwidefsky
4298 <schwidefsky@de.ibm.com>.
4300 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4301 of functions for Linux/IA-64.
4303 * The RPC code is now thread safe. Threads can now use the same service
4304 of different services at the same time. Patch by Eric Norum
4305 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4307 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4308 family of functions for Linux/S390.
4310 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4311 of functions for Linux/x86.
4313 * Port to Linux/CRIS contributed by Axis Communications.
4317 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4318 we can now check for namespace violations and missing declarations. The
4319 result is that almost all headers are now Unix-compliant (as defined in
4320 the upcoming XPG6). The negative side is that some programs might need
4321 corrections, too, if they depend on the incorrect form of the headers in
4322 previous versions which defined too many symbols and included too many
4325 * regex now handles multibyte character sets correctly.
4326 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4328 * iconv (the program) does now conform to the upcoming XPG6 and handles
4329 charmaps. Instead of the charset names the path of charmaps can be
4330 provided and the conversion happens based on this data.
4331 Contributed by Ulrich Drepper.
4333 * The locale program now provides detailed information about the installed
4338 only lists the names of the supported locales
4342 provides details such as country, language, and codeset name.
4343 Contributed by Ulrich Drepper.
4347 * The gencat program now parses the input file according to the charset
4348 selected by the LC_CTYPE category. This is important for stateful
4349 character sets. To make generating catalogs easier there is a way
4350 to overwrite the charset selected by the locale: before the first
4351 message or $ quote line the catalog can contain a line like
4353 $ codeset=ISO-8859-2
4355 to select the charset (ISO-8859-2 in this case).
4357 Implemented by Shinya Hanataka and Ulrich Drepper.
4359 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4360 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4361 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4363 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4364 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4366 * The LANGUAGE environment variable is now ignored unless the locale is
4367 changed from the default "C" locale.
4369 * The usual bug fixes.
4373 * Greg McGary added runtime support for bounds checking using gcc's
4374 new -fbounded-pointers option. ix86 target is complete. PowerPC
4377 * Thorsten Kukuk added secure mode support to nscd.
4379 * The Berkeley db libraries have been removed.
4381 Related, the nss_db module is now in a separate package since it
4382 obviously requires a database library being available.
4384 * Wide character I/O streams implemented by Ulrich Drepper.
4386 * Functions from the extended socket API added by Ulrich Drepper.
4388 * Functions feenableexcept and fedisableexcept to control the
4389 behaviour of individual exceptions have been added by Andreas Jaeger.
4391 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4393 * The resolver code has been updated from bind 8.2.3-T5B which supports
4394 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4397 This change could in some situations effect backward compatibility. Since
4398 now `_res' is a thread-local instead of a global variable, modifying it
4399 in one thread does not have any effect in other threads.
4401 The resolver library was also extended to allow IPv6 as the transport
4402 protocol for the requests. This work was done by Stig Venaas.
4404 * Compatibility code for K&R C compilers has been removed from the
4405 header files. A ISO C compiler is needed to use the library
4406 (conforming to either C89 or C99 standard).
4408 * Complete rewrite of the localedef program to support multibyte character
4409 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4410 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4411 Implemented by Ulrich Drepper.
4413 Bruno Haible significantly improved the generation and use of the data
4414 structures for the wide character tables.
4416 * Plural handling in gettext implemented by Ulrich Drepper.
4418 * The utmp daemon has been removed.
4420 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4422 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4425 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4427 * POSIX barriers implemented by Kaz Kylheku.
4429 * POSIX spawn function family implemented by Ulrich Drepper.
4431 * POSIX spinlocks are now available.
4433 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4435 * the configure option --enable-kernel=X.Y.Z allows to strip out
4436 compatibility for kernel versions before X.Y.Z. This is currently only
4437 implemented for Linux.
4439 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4440 field sin6_scope_id. This means that all programs using IPv6 should be
4441 recompiled. Don't expect binary compatibility with previous glibc
4444 * various conversion modules for IBM character sets contributed by
4447 * IA-64 port by Jes Sorensen and HJ Lu.
4461 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4463 * New xdr functions are added; some rpc functions are now 64bit clean.
4465 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4467 * Fixed known incompatibilities with glibc 2.0.
4469 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4471 * Optimized a number of functions (especially the ELF dynamic loader).
4473 * Update timezone data files.
4475 * lots of charmaps corrections
4477 * some new locale definitions and charmaps
4482 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4483 conform to POSIX member type requirements. Symbol versions have been
4484 adjusted as needed within the library, and for direct use by applications,
4485 but there is potential for problems if third-party libraries use
4486 struct timeval as part of their interface. This does not present
4487 a problem for X and other "essential" system libraries.
4489 * An additional locale model to support C++ Standard Library locale
4490 model and probably more was implemented by Ulrich Drepper.
4492 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4495 * Miles Bader provided the `argp' function family to support hierachical
4496 command line argument parsing, layered on top of getopt.
4498 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4500 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4503 * scanf recognizes the %a and %A format for scanning floating point numbers.
4505 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4506 information and interfaces for the available integer types.
4508 * about 130 new math functions were added to implement the ISO C9x math
4511 * the new header <complex.h> contains definitions of the complex math
4512 functions from ISO C 9X.
4514 * the new header <tgmath.h> defines generic macros to use complex or
4515 real valued functions.
4517 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4519 * Andreas Jaeger provided a test suite for the math library.
4521 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4523 * Ulrich Drepper added character set conversion functions (iconv).
4525 * Optimized string functions have been added.
4527 * The localedata addon is now part of glibc.
4529 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4531 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4532 daemon for NSS (nscd).
4534 Missing a better place here are some numbers on improvements. Under
4535 Linux 2.1.125 un-tar-ing the kernel sources takes
4539 using local files 12.19s 6.88s 22.91s
4541 using NIS 13.92s 8.91s 26.34s
4543 using NIS & nscd 10.37s 7.34s 25.30s
4545 using NIS+ 27.57s 30.37s 640.46s
4547 using NIS+ & nscd 10.25s 7.83s 26.51s
4549 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4551 Keep in mind that non-namelookup related operations dominate above times.
4552 It was just a common complain that using NIS+ unpacking the kernel is
4555 [1] The old nscd implementation is not available anymore since it was
4556 distributed with glibc up to version 2.0.98 and thus is now replaced.
4558 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4560 * Mark Kettenis provided a Hesiod NSS module.
4562 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4563 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4565 * rcmd can now handle netgroups (Dick Streefland).
4567 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4570 * Support for the IPv6 protocol has been added to the socket API, as per the
4571 latest draft standards.
4573 * Support for Linux 2.2 has been added.
4575 * Interface changes relative to the latest 2.0.x release:
4576 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4577 addseverity NEW: Unix98
4578 alphasort64 NEW: LFS
4579 argp_err_exit_status NEW: argp, GNU ext
4580 argp_error NEW: argp, GNU ext
4581 argp_failure NEW: argp, GNU ext
4582 argp_help NEW: argp, GNU ext
4583 argp_parse NEW: argp, GNU ext
4584 argp_program_bug_address NEW: argp, GNU ext
4585 argp_program_version NEW: argp, GNU ext
4586 argp_program_version_hook NEW: argp, GNU ext
4587 argp_state_help NEW: argp, GNU ext
4588 argp_usage NEW: argp, GNU ext
4589 authdes_create NEW: Secure RPC
4590 authdes_getucred NEW: Secure RPC
4591 authdes_pk_create NEW: Secure RPC
4592 backtrace NEW: GNU ext.
4593 backtrace_symbols NEW: GNU ext.
4594 backtrace_symbols_fd NEW: GNU ext.
4596 cacosf NEW: ISO C 9x
4597 cacosh NEW: ISO C 9x
4598 cacoshf NEW: ISO C 9x
4599 cacoshl NEW: ISO C 9x
4600 cacosl NEW: ISO C 9x
4607 casinf NEW: ISO C 9x
4608 casinh NEW: ISO C 9x
4609 casinhf NEW: ISO C 9x
4610 casinhl NEW: ISO C 9x
4611 casinl NEW: ISO C 9x
4613 catanf NEW: ISO C 9x
4614 catanh NEW: ISO C 9x
4615 catanhf NEW: ISO C 9x
4616 catanhl NEW: ISO C 9x
4617 catanl NEW: ISO C 9x
4618 cbc_crypt NEW: Secure RPC
4622 ccoshf NEW: ISO C 9x
4623 ccoshl NEW: ISO C 9x
4629 cimagf NEW: ISO C 9x
4630 cimagl NEW: ISO C 9x
4631 clearerr_locked REMOVED
4632 clntunix_create NEW: sunrpc ext
4634 clog10 NEW: ISO C 9x
4635 clog10f NEW: ISO C 9x
4636 clog10l NEW: ISO C 9x
4646 cprojf NEW: ISO C 9x
4647 cprojl NEW: ISO C 9x
4649 crealf NEW: ISO C 9x
4650 creall NEW: ISO C 9x
4655 csinhf NEW: ISO C 9x
4656 csinhl NEW: ISO C 9x
4659 csqrtf NEW: ISO C 9x
4660 csqrtl NEW: ISO C 9x
4664 ctanhf NEW: ISO C 9x
4665 ctanhl NEW: ISO C 9x
4667 des_setparity NEW: Secure RPC
4668 ecb_crypt NEW: Secure RPC
4669 endutxent NEW: Unix98
4671 exp10f NEW: ISO C 9x
4672 exp10l NEW: ISO C 9x
4676 fattach NEW: STREAMS
4677 fdetach NEW: STREAMS
4681 feclearexcept NEW: ISO C 9x
4682 fegetenv NEW: ISO C 9x
4683 fegetexceptflag NEW: ISO C 9x
4684 fegetround NEW: ISO C 9x
4685 feholdexcept NEW: ISO C 9x
4687 feraiseexcept NEW: ISO C 9x
4688 ferror_locked REMOVED
4689 fesetenv NEW: ISO C 9x
4690 fesetexceptflag NEW: ISO C 9x
4691 fesetround NEW: ISO C 9x
4692 fetestexcept NEW: ISO C 9x
4693 feupdateenv NEW: ISO C 9x
4694 fflush_locked REMOVED
4698 fgets_unlocked NEW: GNU ext.
4699 fileno_locked REMOVED
4711 fputc_locked REMOVED
4712 fputs_unlocked NEW: GNU ext.
4713 fread_unlocked NEW: GNU ext.
4718 fstatvfs NEW: Unix98
4722 ftruncate64 NEW: LFS
4724 fwrite_unlocked NEW: GNU ext.
4725 gai_strerror NEW: IPv6
4729 getchar_locked REMOVED
4731 getdate_err NEW: Unix98
4732 getdate_r NEW: GNU ext.
4734 getnameinfo NEW: IPv6
4735 getnetname NEW: Secure RPC
4736 getpmsg NEW: STREAMS
4737 getpt NEW: Unix98 PTY
4738 getrlimit64 NEW: LFS
4739 getutxent NEW: Unix98
4740 getutxid NEW: Unix98
4741 getutxline NEW: Unix98
4742 glob64 NEW: GNU ext.
4743 globfree64 NEW: GNU ext.
4744 gnu_get_libc_release NEW: GNU ext.
4745 gnu_get_libc_version NEW: GNU ext.
4746 grantpt NEW: Unix98 PTY
4747 host2netname NEW: Secure RPC
4749 iconv_close NEW: iconv
4750 iconv_open NEW: iconv
4751 if_freenameindex NEW: IPv6
4752 if_indextoname NEW: IPv6
4753 if_nameindex NEW: IPv6
4754 if_nametoindex NEW: IPv6
4755 in6addr_any NEW: IPv6
4756 in6addr_loopback NEW: IPv6
4757 inet6_isipv4mapped NEW: IPv6
4758 isastream NEW: STREAMS
4759 iswblank NEW: GNU ext.
4760 key_decryptsession NEW: Secure RPC
4761 key_decryptsession_pk NEW: Secure RPC
4762 key_encryptsession NEW: Secure RPC
4763 key_encryptsession_pk NEW: Secure RPC
4764 key_gendes NEW: Secure RPC
4765 key_get_conv NEW: Secure RPC
4766 key_secretkey_is_set NEW: Secure RPC
4767 key_setnet NEW: Secure RPC
4768 key_setsecret NEW: Secure RPC
4769 llrint NEW: ISO C 9x
4770 llrintf NEW: ISO C 9x
4771 llrintl NEW: ISO C 9x
4772 llround NEW: ISO C 9x
4773 llroundf NEW: ISO C 9x
4774 llroundl NEW: ISO C 9x
4779 lrintf NEW: ISO C 9x
4780 lrintl NEW: ISO C 9x
4781 lround NEW: ISO C 9x
4782 lroundf NEW: ISO C 9x
4783 lroundl NEW: ISO C 9x
4785 makecontext NEW: Unix98
4786 mempcpy NEW: GNU ext.
4789 modify_ldt NEW: kernel
4793 nearbyint NEW: ISO C 9x
4794 nearbyintf NEW: ISO C 9x
4795 nearbyintl NEW: ISO C 9x
4796 netname2host NEW: Secure RPC
4797 netname2user NEW: Secure RPC
4798 nexttoward NEW: ISO C 9x
4799 nexttowardf NEW: ISO C 9x
4800 nexttowardl NEW: ISO C 9x
4804 passwd2des NEW: Secure RPC
4806 pow10f NEW: GNU ext.
4807 pow10l NEW: GNU ext.
4810 printf_size NEW: GNU ext.
4811 printf_size_info NEW: GNU ext.
4812 profil_counter REMOVED
4813 pthread_mutexattr_getkind_np REPLACED
4814 pthread_mutexattr_setkind_np REPLACED
4815 ptsname NEW: Unix98 PTY
4816 ptsname_r NEW: Unix98 PTY
4818 putchar_locked REMOVED
4819 putgrent NEW: GNU ext.
4821 putpmsg NEW: STREAMS
4822 pututxline NEW: Unix98
4826 readdir64_r NEW: LFS
4827 remquo NEW: ISO C 9x
4828 remquof NEW: ISO C 9x
4829 remquol NEW: ISO C 9x
4831 roundf NEW: ISO C 9x
4832 roundl NEW: ISO C 9x
4834 scalbln NEW: ISO C 9x
4835 scalblnf NEW: ISO C 9x
4836 scalblnl NEW: ISO C 9x
4838 sendfile NEW: kernel
4839 setcontext NEW: Unix98
4840 setrlimit64 NEW: LFS
4841 setutxent NEW: Unix98
4843 sigignore NEW: Unix98
4844 sigqueue NEW: POSIX.1b
4845 sigrelse NEW: Unix98
4846 sigset NEW: POSIX.1b
4847 sigtimedwait NEW: POSIX.1b
4848 sigwaitinfo NEW: POSIX.1b
4849 sincos NEW: GNU ext.
4850 sincosf NEW: GNU ext.
4851 sincosl NEW: GNU ext.
4855 strcasestr NEW: GNU ext.
4856 strtoimax NEW: ISO C 9x
4857 strtoumax NEW: ISO C 9x
4858 strverscmp NEW: GNU ext.
4859 svcauthdes_stats NEW: Secure RPC
4860 svcunix_create NEW: sunrpc etx
4861 svcunixfd_create NEW: sunrpc ext
4862 swapcontext NEW: Unix98
4863 tcgetsid NEW: Unix98 PTY
4864 tdestroy NEW: GNU ext.
4865 tgamma NEW: ISO C 9x
4866 tgammaf NEW: ISO C 9x
4867 tgammal NEW: ISO C 9x
4871 truncf NEW: ISO C 9x
4872 truncl NEW: ISO C 9x
4874 unlockpt NEW: Unix98 PTY
4875 updwtmpx NEW: Unix98
4876 user2netname NEW: Secure RPC
4877 utmpxname NEW: Unix98
4878 versionsort NEW: GNU ext.
4879 versionsort64 NEW: GNU ext.
4881 wcscasecmp NEW: GNU ext.
4882 wcsncasecmp NEW: GNU ext.
4883 wcsnlen NEW: GNU ext.
4884 wcstoimax NEW: ISO C 9x
4885 wcstoll NEW: ISO C 9x
4886 wcstoull NEW: ISO C 9x
4887 wcstoumax NEW: ISO C 9x
4889 wordexp NEW: POSIX.2
4890 wordfree NEW: POSIX.2
4891 write_profiling REMOVED
4892 xdecrypt NEW: Secure RPC
4893 xdr_authdes_cred NEW: Secure RPC
4894 xdr_authdes_verf NEW: Secure RPC
4895 xdr_cryptkeyarg NEW: Secure RPC
4896 xdr_cryptkeyarg2 NEW: Secure RPC
4897 xdr_cryptkeyres NEW: Secure RPC
4898 xdr_getcredres NEW: Secure RPC
4899 xdr_key_netstarg NEW: Secure RPC
4900 xdr_key_netstres NEW: Secure RPC
4901 xdr_keybuf NEW: Secure RPC
4902 xdr_keystatus NEW: Secure RPC
4903 xdr_netnamestr NEW: Secure RPC
4904 xdr_sizeof NEW: Secure RPC
4905 xdr_unixcred NEW: sunrpc ext
4906 xencrypt NEW: Secure RPC
4907 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4918 * inet_ntoa is thread-safe
4920 * updwtmp is moved from libutil to libc
4922 * rewrite of cbrt function
4924 * update of timezone data
4938 * add atoll function
4940 * fix complex problems in Berkeley DB code
4942 * fix math functions
4946 * fixed lots of header problems (especially Linux/GNU specific)
4948 * dynamic loader preserves all registers
4950 * Roland McGrath provided support for handling of auxiliary objects in
4951 the ELF dynamic loader.
4953 * support for parallel builds is improved
4957 * GNU extensions are no longer declared by default. To enable them you
4958 must define the macro `_GNU_SOURCE' in your program or compile with
4961 * The library has changed from using GNU ld symbol aliases to using weak
4962 symbols where available. The ELF object file format supports weak
4963 symbols; GNU ld also supports weak symbols in the a.out format. (There
4964 is also now support for other GNU ld extensions in ELF. Use the
4965 `--with-elf' option to configure to indicate you have ELF, and
4966 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4967 of many files which contained only symbol aliases, reducing the size of
4968 the source and the compiled library; many other files were renamed to
4969 less cryptic names previously occupied by the symbol alias files.
4970 There is a new header file <elf.h> for programs which operate on
4971 files in the ELF format.
4973 * Converted to Autoconf version 2, so `configure' has more options.
4974 Run `configure --help' to see the details.
4976 * The library can now be configured to build profiling, highly-optimized
4977 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4978 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4979 `configure' enable building these extra libraries. The shared library is
4980 built by default when using both ELF and GNU ld. When shared libraries
4981 are enabled, the new library `-ldl' is available for arbitrary run-time
4982 loading of shared objects; its interface is defined in <dlfcn.h>. The
4983 new header file <link.h> gives access to the internals of the run-time
4984 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4985 application of same name on other systems and it provides information
4986 about dynamically linked binaries.
4988 * The C library now provides the run-time support code for profiling
4989 executables compiled with `-pg'. Programs can control the profiling code
4990 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4991 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4992 the support for this file format was contributed by David Mosberger-Tang.
4994 * The math code has been replaced with a math library based on fdlibm from
4995 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4996 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4997 functions. The math functions now reside in a separate library, so
4998 programs using them will need to use `-lm' their linking commands.
5000 * John C. Bowman contributed optimized ix87 assembler inline functions.
5002 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
5003 mechanism similar to that found in Solaris 2. This is now used for the
5004 group, passwd, hosts, networks, services, protocols, rpc, ethers,
5005 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
5006 file controls what services are used for each individual database. This
5007 works by loading shared libraries with names specified in `nsswitch.conf',
5008 so service modules can be changed or added at any time without even
5009 relinking any program. Currently there are the file, db, and NIS based
5010 NSS services available.
5012 * The new functions `strtoq' and `strtouq' parse integer values from
5013 strings, like `strtol' and `strtoul', but they return `long long int' and
5014 `unsigned long long int' values, respectively (64-bit quantities).
5016 * The new functions `strtof' and `strtold' parse floating-point values from
5017 strings, like `strtod', but they return `float' and `long double' values,
5018 respectively (on some machines `double' and `long double' are the same).
5020 * Ulrich Drepper has contributed new implementations of the floating-point
5021 printing and reading code used in the `printf' family of functions and
5022 `strtod', `strtof', and `strtold'. These new functions are perfectly
5023 accurate, and much faster than the old ones.
5025 * The implementation of the POSIX locale model was completely rewritten by
5026 Ulrich Drepper. This includes the new programs `localedef' and `locale'
5027 to compile the POSIX locale definition.
5029 * The former dummy implementations of the strcoll and strxfrm function are
5030 now replaced by fully functional code contributed by Ulrich Drepper. The
5031 collation information comes from the POSIX locale definitions.
5033 * The new header <langinfo.h> defines an interface for accessing
5034 various locale-dependent data (using the locale chosen with `setlocale').
5036 * Ulrich Drepper has contributed a new suite of functions for operation on
5037 wide-character and multibyte-character strings, in <wchar.h>;
5038 and classification and case conversion of wide characters, in <wctype.h>.
5039 These new functions are conforming to the ISO C, Amendement 1 specification.
5041 * There is now a second implementation of the standard I/O library available.
5042 It comes from GNU libg++ as was written by Per Bothner, heavily modified
5043 by Hongjiu Lu and made thread safe by Ulrich Drepper.
5045 * You can now use positional parameter specifications in format strings
5046 for the `printf' and `scanf' families of functions. For example,
5047 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
5048 ``Number 6, Mr Jones''. This is mainly useful when providing different
5049 format strings for different languages, whose grammars may dictate
5050 different orderings of the values being printed. To support this
5051 feature, the interface for `register_printf_handler' has changed; see
5052 the header file <printf.h> for details.
5054 * The `printf' and `scanf' families of functions now understand a new
5055 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
5056 to group numbers as indicated by the locale; for `scanf' and friends, this
5057 says to accept as valid only a number with all the proper grouping
5058 separators in the right places. In the default "C" locale, numbers are
5059 not grouped; but locales for specific countries will define the usual
5060 conventions (i.e. separate thousands with `,' in the US locale).
5062 * The pgrp functions have been regularized, slightly incompatibly but much
5063 less confusingly. The core functions are now `getpgid' and `setpgid',
5064 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
5065 argument) and BSD `setpgrp' (identical to `setpgid') functions are
5066 provided for compatibility. There is no longer an incompatible `getpgrp'
5067 with an argument declared under _BSD_SOURCE; no BSD code uses it.
5069 * The new header file <fts.h> and suite of functions simplify programs that
5070 operate on directory trees. This code comes from 4.4 BSD.
5072 * The resolver code has been updated from the BIND 4.9.5-P1 release.
5073 Parts of the code were heavily modified by Ulrich Drepper to fit in the
5074 NSS scheme used in glibc.
5076 * There is a new malloc debugging hook `__memalign_hook'.
5078 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
5079 `unsigned int' in <sys/types.h>. These are for compatibility only and
5080 their use is discouraged.
5082 * The `-lmcheck' library to enable standard malloc debugging hooks is now
5083 done differently, so that it works even without GNU ld.
5085 * New function `euidaccess' checks allowed access to a file like `access',
5086 but using the effective IDs instead of the real IDs.
5088 * The time zone data files have been updated for the latest and greatest
5089 local time conventions of the countries of the world.
5091 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
5094 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
5095 for formatting floating-point numbers. They are provided only for
5096 compatibility; new programs should use `sprintf' instead. There are
5097 also equivalent function for the `long double' floating-point type and
5098 all functions also exist in a reentrant form.
5100 * The new auxiliary library `-lutil' from 4.4 BSD contains various
5101 functions for maintaining the login-record files (primarily of use to
5102 system programs such as `login'), and convenient functions for
5103 allocating and initializing a pseudo-terminal (pty) device.
5105 * Ulrich Drepper has contributed new support for System V style
5106 shared memory and IPC on systems that support it.
5108 * Ulrich Drepper has contributed several miscellaneous new functions found
5109 in System V: The `hsearch' family of functions provide an effective
5110 implementation of hash tables; `a64l' and `l64a' provide a very simple
5111 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
5114 * Ulrich Drepper has contributed new reentrant counterparts for the
5115 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
5117 * Ulrich Drepper has contributed new, highly-optimized versions of several
5118 string functions for the i486/Pentium family of processors.
5120 * Ulrich Drepper has updated the Linux-specific code, based largely
5121 on work done in Hongjiu Lu's version of GNU libc for Linux.
5122 The GNU library now supports Linux versions 2.0.10 and later,
5123 using the ELF object file format (i[3456]86-*-linux).
5125 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
5127 * David Mosberger-Tang and Richard Henderson have ported the C library
5128 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
5129 dynamic linking support for ELF/Alpha.
5131 * Richard Henderson contributed several Alpha optimized assembler function
5132 for arithmetic and string handling.
5134 * Ulrich Drepper has contributed a new set of message catalog functions to
5135 support multiple languages using the <libintl.h> interface, for use with
5136 his new package GNU gettext. Translation volunteers have contributed
5137 catalogs of the library's messages in Spanish, German, and Korean.
5139 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
5140 program and the `catgets' function for reading the catalog files it
5141 creates. (The <libintl.h> interface is preferred; we include the
5142 <nl_types.h> interface using `catgets' only for source compatibility with
5143 programs already written to use it.)
5145 * New header file <values.h> gives SVID-compatible names for <limits.h>
5148 * Various new macros, declarations, and small header files for compatibility
5151 * New function `group_member' is a convenient way to check if a process has
5152 a given effective group ID.
5154 * When using GCC 2.7 and later, the socket functions are now declared in a
5155 special way so that passing an argument of type `struct sockaddr_in *',
5156 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
5157 `struct sockaddr *' type, does not generate a type-clash warning.
5159 * New function `error' declared in header file <error.h> is a convenient
5160 function for printing error messages and optionally exiting; this is the
5161 canonical function used in GNU programs. The new functions `err', `warn',
5162 and friends in header file <err.h> are the canonical 4.4 BSD interface for
5163 doing the same thing.
5165 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
5166 POSIX.2 `glob' function to do ~ and {...} expansion.
5168 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
5169 `clearenv' which is used in POSIX.9 is also available.
5171 * New function `getsid' returns session ID number on systems that support it.
5173 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
5174 files <db.h> and <mpool.h> provide a rich set of functions for several
5175 types of simple databases stored in memory and in files, and <ndbm.h> is
5176 an old `ndbm'-compatible interface using the `db' functions. Link with
5177 `-ldb' to get these functions.
5179 * New macro `strdupa' copies a string like `strdup', but uses local stack
5180 space from `alloca' instead of dynamic heap space from `malloc'.
5182 * New function `strnlen' is like `strlen' but searches only a given maximum
5183 number of characters for the null terminator. `stpncpy', `strndup' and
5184 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
5187 * New function `statfs' in header <sys/statfs.h>.
5189 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
5190 provide convenient functions for operating on blocks of null-terminated
5193 * A new suite of functions in <utmp.h> handle all the details of reading
5194 and writing the utmp file.
5196 * An implementation of the NIS/YP(tm) based NSS service was contributed by
5199 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
5200 completely POSIX compliant and also implemented the extended functionality
5201 to handle alternate digit representation and alternate era date formats.
5203 * Ulrich Drepper provided an implementation of the `strptime' function
5204 defined in XPG4.2 which transforms a string into a `struct tm' value.
5206 * Paul Eggert provided the tzselect shell script as part of the timezone
5207 code. The shell script makes it easy to select the correct timezone
5210 * The implementation of the malloc family of functions is completely replaced
5211 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
5212 The implementation uses the mmap function (if available) and it is
5213 optimized for the use in multi threaded programs.
5215 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
5216 functions. This new functionality is usable by specifying a special
5217 salt string and it is compatible with implementation on *BSD systems.
5219 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
5220 `getsubopt' to handle second level command line options, `bsd_signal'
5221 to access BSD style `signal' functionality, the obsolete `regexp' style
5224 * the `lchown' function is available on system which support this
5227 * The implementation of the shadow password handling function was contributed
5230 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5232 * POSIX.1g support was added. The <sys/select.h> header is available,
5233 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5234 implementation of `getaddrinfo'.
5238 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5240 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5241 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5243 * The new function `daemon' from 4.4 BSD is useful for server programs that
5244 want to put themselves in the background.
5246 * Joel Sherrill has contributed support for several standalone boards that
5247 run without an operating system.
5249 * `printf', `scanf' and friends now accept a `q' type modifier for long
5250 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5252 * All of the code taken from BSD (notably most of the math and networking
5253 routines) has been updated from the BSD 4.4-Lite release.
5255 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5257 * The new functions `getdomainname' and `setdomainname' fetch or change the
5258 YP/NIS domain name. These are system calls which exist on systems which
5261 * The time zone data files have been updated for the latest international
5264 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5265 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5269 * The C library now includes support for Sun RPC, from Sun's free
5270 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5271 are included. (There is still no support for YP.)
5273 * Tom Quinn has contributed a port of the C library to SGI machines running
5274 Irix 4 (mips-sgi-irix4).
5276 * The new `lockf' function is a simplified interface to the locking
5277 facilities of `fcntl', included for compatibility.
5279 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5281 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5283 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5286 * The error message printed by `assert' for a failed assertion now includes
5287 the name of the program (if using GNU ld) and the name of the calling
5288 function (with versions of GCC that support this).
5290 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5292 * The library now includes the <sys/mman.h> header file and memory
5293 management functions `mmap', `munmap', `mprotect', `msync', and
5294 `madvise', on systems that support those facilities.
5296 * The interface for `mcheck' has changed slightly: the function called to
5297 abort the program when an allocation inconsistency is detected now takes
5298 an argument that indicates the type of failure. The new function
5299 `mprobe' lets you request a consistency check for a particular block at
5300 any time (checks are normally done only when you call `free' or `realloc'
5303 * It is now possible to easily cross-compile the C library, building on one
5304 system a library to run on another machine and/or operating system. All
5305 you need to do is set the variable `HOST_CC' in `configparms' to the
5306 native compiler for programs to run on the machine you are building on (a
5307 few generator programs are used on Unix systems); set `CC' to the
5310 * The new function `fexecve' (only implemented on the GNU system) executes
5311 a program file given a file descriptor already open on the file.
5315 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5316 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5318 * You can set the variable `libprefix' in `configparms' to specify a prefix
5319 to be prepended to installed library files; this makes it easy to install
5320 the GNU C library to be linked as `-lgnuc' or whatever.
5322 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5323 copies a limited number of characters from a string, and returns the
5324 address of the last character written.
5326 * You no longer need to check for whether the installed `stddef.h' is
5327 compatible with the GNU C library. configure now checks for you.
5329 * You can now define a per-stream `fileno' function to convert the
5330 stream's cookie into an integral file descriptor.
5332 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5333 allocates zero bytes of storage, and returns a unique pointer which
5334 you can pass to `realloc' or `free'. The behavior is undefined if
5335 you dereference this pointer.
5337 * The C library now runs on Sony NEWS m68k machines running either
5338 NewsOS 3 or NewsOS 4.
5340 * The new `syscall' function is a system-dependent primitive function
5341 for invoking system calls. It has the canonical behavior on Unix
5342 systems, including unreliable return values for some calls (such as
5343 `pipe', `fork' and `getppid').
5345 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5346 to `EAGAIN', which is the preferred name. On systems whose kernels
5347 use two distinct codes, the C library now translates EWOULDBLOCK to
5348 EAGAIN in every system call function.
5352 * The GNU C Library Reference Manual is now distributed with the library.
5353 `make dvi' will produce a DVI file of the printed manual.
5354 `make info' will produce Info files that you can read on line using C-h i
5355 in Emacs or the `info' program.
5356 Please send comments on the manual to bug-glibc-manual@gnu.org.
5358 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5360 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5362 * Jason Merrill has contributed a port to the Sequent Symmetry running
5363 Dynix version 3 (i386-sequent-dynix).
5365 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5366 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5368 * New function `memory_warnings' lets you arrange to get warnings when
5369 malloc is running out of memory to allocate, like Emacs gives you.
5371 * The C library now contains the relocating allocator used in Emacs 19 for
5372 its editing buffers. This allocator (ralloc) minimizes allocation
5373 overhead and fragmentation by moving allocated regions around whenever it
5374 needs to. You always refer to a ralloc'd region with a "handle" (a
5375 pointer to a pointer--an object of type `void **').
5377 * There is a new `printf' format: `%m' gives you the string corresponding
5378 to the error code in `errno'.
5380 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5381 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5382 pass, the `a' modifier says to fill in a `char **' you pass with a
5385 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5386 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5387 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5389 * `mkstemp' is a traditional Unix function to atomically create and open a
5390 uniquely-named temporary file.
5394 * The standard location for the file that says what the local timezone is
5395 has changed again. It is now `/usr/local/etc/localtime' (or more
5396 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5398 * The distribution no longer contains any files with names longer than 14
5401 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5402 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5404 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5408 * The configuration process has changed quite a bit. The `configure'
5409 script is now used just like the configuration scripts for other GNU
5410 packages. The `sysdeps' directory hierarchy is much rearranged.
5411 The file `INSTALL' explains the new scheme in detail.
5413 * The header files no longer need to be processed into ANSI C and
5414 traditional C versions. There is just one set of files to install, and
5415 it will work with ANSI or old C compilers (including `gcc -traditional').
5417 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5418 MIPS DECStation running Ultrix 4.
5420 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5421 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5422 made itself into a shared library.
5424 * Yet further improved support for the i386, running 4.3 BSD-like systems
5425 (such as Mach 3 with the Unix single-server), or System V.
5427 * New function `strncasecmp' to do case-insensitive string comparison
5428 with limited length.
5430 * New function `strsep' is a reentrant alternative to `strtok'.
5432 * New functions `scandir' and `alphasort' for searching directories.
5434 * New function `setenv' is a better interface to `putenv'.
5436 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5437 function for traversing a directory tree.
5439 * The GNU obstack package is now also part of the C library.
5440 The new function `open_obstack_stream' creates a stdio stream that
5441 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5442 formatted output directly to an obstack.
5444 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5445 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5447 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5449 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5450 things to your strings.
5452 * There are some new test programs: test-fseek, testmb, and testrand.
5454 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5455 These ports are not finished, but are a good starting place for really
5456 supporting those systems.
5458 * `/etc/localtime' is now the standard location for the file that says what
5459 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5460 This follows the general principle that `/etc' is the place for all local
5461 configuration files.
5463 * The C library header files now use `extern "C"' when used by the C++
5464 compiler, so the C library should now work with C++ code.
5466 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5467 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5470 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5471 function which uses the `merge sort' algorithm, and is said to be
5472 significantly faster than the old GNU `qsort' function. Merge sort is
5473 now the standard `qsort' function. The new algorithm can require a lot
5474 of temporary storage; so, the old sorting function is called when the
5475 required storage is not available.
5477 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5478 provides the Unix `crypt' function, plus some other entry points.
5480 * The code and header files taken from 4.4 BSD have been updated with the
5481 latest files released from Berkeley.
5483 ----------------------------------------------------------------------
5484 Copyright information:
5486 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5488 Permission is granted to anyone to make or distribute verbatim copies
5489 of this document as received, in any medium, provided that the
5490 copyright notice and this permission notice are preserved,
5491 thus giving the recipient permission to redistribute in turn.
5493 Permission is granted to distribute modified versions
5494 of this document, or of portions of it,
5495 under the above conditions, provided also that they
5496 carry prominent notices stating who last changed them.
5499 version-control: never