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 * The entry for the new Japanese era has been added for ja_JP locale.
14 Deprecated and removed features, and other changes affecting compatibility:
16 * The copy_file_range function fails with ENOSYS if the kernel does not
17 support the system call of the same name. Previously, user space
18 emulation was performed, but its behavior did not match the kernel
19 behavior, which was deemed too confusing. Applications which use the
20 copy_file_range function will have to be run on kernels which implement
21 the copy_file_range system call. Support for most architectures was added
22 in version 4.5 of the mainline Linux kernel.
24 The following bugs are resolved with this release:
26 [16573] malloc: Set and reset all hooks for tracing
28 [20568] Fix crash in _IO_wfile_sync
29 [24155] x32 memcmp can treat positive length as 0 (if sign bit in RDX is set) (CVE-2019-7309)
30 [24164] Systemtap probes need to use "nr" constraint on 32-bit Arm
31 [24161] __run_fork_handlers self-deadlocks in malloc/tst-mallocfork2
32 [24211] Use-after-free in Systemtap probe in pthread_join
33 [24228] old x86 applications that use legacy libio crash on exit
34 [24476] dlfcn: Guard __dlerror_main_freeres with __libc_once_get (once)
35 [24744] io: Remove the copy_file_range emulation
36 [24986] alpha: new getegid, geteuid and getppid syscalls used
38 [25203] libio: Disable vtable validation for pre-2.1 interposed handles
39 [25204] Ignore LD_PREFER_MAP_32BIT_EXEC for SUID programs
40 [25225] ld.so fails to link on x86 if GCC defaults to -fcf-protection
41 [25232] No const correctness for strchr et al. for Clang++
42 [25933] Off by one error in __strncmp_avx2
43 [25976] nss_compat: internal_end*ent may clobber errno, hiding ERANGE
45 Security related changes:
47 CVE-2019-7309: x86-64 memcmp used signed Jcc instructions to check
48 size. For x86-64, memcmp on an object size larger than SSIZE_MAX
49 has undefined behavior. On x32, the size_t argument may be passed
50 in the lower 32 bits of the 64-bit RDX register with non-zero upper
51 32 bits. When it happened with the sign bit of RDX register set,
52 memcmp gave the wrong result since it treated the size argument as
53 zero. Reported by H.J. Lu.
55 CVE-2019-9169: Attempted case-insensitive regular-expression match
56 via proceed_next_node in posix/regexec.c leads to heap-based buffer
57 over-read. Reported by Hongxu Chen.
59 CVE-2019-19126: ld.so failed to ignore the LD_PREFER_MAP_32BIT_EXEC
60 environment variable during program execution after a security
61 transition, allowing local attackers to restrict the possible mapping
62 addresses for loaded libraries and thus bypass ASLR for a setuid
63 program. Reported by Marcin KoĆcielnicki.
70 * The getcpu wrapper function has been added, which returns the currently
71 used CPU and NUMA node. This function is Linux-specific.
73 * A new convenience target has been added for distribution maintainers
74 to build and install all locales as directories with files. The new
75 target is run by issuing the following command in your build tree:
76 'make localedata/install-locale-files', with an optional DESTDIR
77 to set the install root if you wish to install into a non-default
80 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
82 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
83 for _GNU_SOURCE, to match BSD environments.
85 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
86 indicates that it will abort the transaction prior to entering the kernel
87 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
88 suspended, and this caused some undefined side-effects issues by aborting
89 transactions manually. Glibc avoided it by abort transactions manually on
90 each syscall, but it lead to performance issues on newer kernels where the
91 HTM state is saved and restore lazily (the state being saved even when the
92 process actually does not use HTM).
94 * The functions posix_spawn_file_actions_addchdir_np and
95 posix_spawn_file_actions_addfchdir_np have been added, enabling
96 posix_spawn and posix_spawnp to run the new process in a different
97 directory. These functions are GNU extensions. The function
98 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
101 * The popen and system do not run atfork handlers anymore (BZ#17490).
102 Although it is a possible POSIX violation, the POSIX rationale in
103 pthread_atfork documentation regarding atfork handlers is to handle
104 inconsistent mutex state after a fork call in a multi-threaded process.
105 In both popen and system there is no direct access to user-defined mutexes.
107 * Support for the C-SKY ABIV2 running on Linux has been added. This port
108 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
110 - C-SKY ABIV2 soft-float little-endian
111 - C-SKY ABIV2 hard-float little-endian
113 * strftime's default formatting of a locale's alternative year (%Ey)
114 has been changed to zero-pad the year to a minimum of two digits,
115 like "%y". This improves the display of Japanese era years during
116 the first nine years of a new era, and is expected to be harmless
117 for all other locales (only Japanese locales regularly have
118 alternative year numbers less than 10). Zero-padding can be
119 overridden with the '_' or '-' flags (which are GNU extensions).
121 * As a GNU extension, the '_' and '-' flags can now be applied to
122 "%EY" to control how the year number is formatted; they have the
123 same effect that they would on "%Ey".
125 Deprecated and removed features, and other changes affecting compatibility:
127 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
128 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
130 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
131 in <sys/procfs.h>, has been corrected to match the type actually used by
132 the Linux kernel. This affects the size and layout of that structure on
133 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
135 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
136 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
137 defined in <sys/procfs.h>, has been corrected to match the type actually
138 used by the Linux kernel. This affects the size and layout of those
141 * An archaic GNU extension to scanf, under which '%as', '%aS', and '%a[...]'
142 meant to scan a string and allocate space for it with malloc, is now
143 restricted to programs compiled in C89 or C++98 mode with _GNU_SOURCE
144 defined. This extension conflicts with C99's use of '%a' to scan a
145 hexadecimal floating-point number, which is now available to programs
146 compiled as C99 or C++11 or higher, regardless of _GNU_SOURCE.
148 POSIX.1-2008 includes the feature of allocating a buffer for string input
149 with malloc, using the modifier letter 'm' instead. Programs using
150 '%as', '%aS', or '%a[...]' with the old GNU meaning should change to
151 '%ms', '%mS', or '%m[...]' respectively. Programs that wish to use the
152 C99 '%a' no longer need to avoid _GNU_SOURCE.
154 GCC's -Wformat warnings can detect most uses of this extension, as long
155 as all functions that call vscanf, vfscanf, or vsscanf are annotated with
156 __attribute__ ((format (scanf, ...))).
158 Changes to build and runtime requirements:
160 * Python 3.4 or later is required to build the GNU C Library.
162 * On most architectures, GCC 5 or later is required to build the GNU C
163 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
165 Older GCC versions and non-GNU compilers are still supported when
166 compiling programs that use the GNU C Library.
168 Security related changes:
170 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
171 denial of service due to resource exhaustion when processing getaddrinfo
172 calls with crafted host names. Reported by Guido Vranken.
174 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
175 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
176 happened, accessing the 32-bit size_t value as the full 64-bit register
177 in the assembly string/memory functions would cause a buffer overflow.
180 CVE-2016-10739: The getaddrinfo function could successfully parse IPv4
181 addresses with arbitrary trailing characters, potentially leading to data
182 or command injection issues in applications.
184 The following bugs are resolved with this release:
186 [10425] localedata: it_IT/it_CH: LC_TIME format is wrong
187 [10496] localedata: 12h time representation in multiple locales faulty
188 [10797] localedata: it_IT locale numeric does not have a separator for
190 [11319] libc: dprintf doesn't handle errors properly
191 [16346] time: mktime: potentially unsafe use of localtime_offset
192 [17248] build: glibc should not sort CFLAGS (support gcc plugins and
194 [17405] libc: Implement posix_spawn_file_actions_addchdir_np,
195 posix_spawn_file_actions_addfchdir_np
196 [17426] localedata: Indian locales: set the correct date format
197 [17490] stdio: popen should not invoke atfork handlers
198 [17783] libc: TIOCSER_TEMT conditions inconsistent
199 [18040] regex: use-after-free in regexec/get_subexp
200 [18093] libc: Corrupted aux-cache causes ldconfig to segfault
201 [20018] network: getaddrinfo should reject IP addresses with trailing
202 characters (CVE-2016-10739)
203 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
204 [20271] libc: Missing "\n" in __libc_fatal calls
205 [20480] dynamic-link: Patch: ifunc not executable, crashes sudo qemu
206 [20544] libc: RFE: atexit, __cxa_atexit, on_exit should assert function
207 pointer argument is non-NULL
208 [21037] stdio: open_memstream and freopen
209 [21286] libc: bits/siginfo.h is missing enum definition for TRAP_HWBKPT
210 [21716] time: Crash in glibc's mktime in low-memory situations
211 [22834] stdio: Subprocess forked by popen may crash in Linux when
212 multithreads call popen
213 [22927] network: crash in vn_gai_enqueue_request if requests_tail was NULL
214 and pthread_create fails.
215 [23032] hurd: sysdeps/htl/pt-barrier-init.c:39: bad call to memcmp ?
216 [23125] libc: riscv64: endless loop when throwing an exception from a
218 [23275] nptl: Race in pthread_mutex_lock while promoting to
219 PTHREAD_MUTEX_ELISION_NP.
220 [23400] libc: stdlib/test-bz22786.c creates temporary files in glibc
222 [23479] math: [mips] bits/fenv.h should not define some macros for soft-
224 [23490] libc: sysdeps/unix/sysv/linux/x86/tst-cet-property-2.c:49: off by
226 [23497] libc: readdir64@GLIBC_2.1 cannot parse the kernel directory stream
227 [23509] dynamic-link: CET enabled glibc is incompatible with the older
229 [23520] nscd: nscd: Use-after-free in addgetnetgrentX and its callers
230 [23521] nss: get_next_alias nss_files file stream leak
231 [23538] nptl: Hang in pthread_cond_broadcast
232 [23562] libc: Wrong type for si_band in Linux-specific siginfo_t
233 [23578] regex: Invalid memory access if regex pattern contains NUL byte
234 [23579] libc: Errors misreported in preadv2
235 [23597] build: support/test-container.c doesn't work with different
237 [23603] time: mktime signed integer overflow on large timestamps
238 [23606] libc: Missing ENDBR32 in sysdeps/i386/start.S
239 [23614] libc: powerpc: missing CFI register information in __mpn_*
241 [23637] string: Generic strstr/strcasestr fails with huge needles
242 [23640] libc: no way to easily clear FD_CLOEXEC in
243 posix_spawn_file_actions_adddup2()
244 [23649] libc: [microblaze/mips/nios2/riscv] sys/procfs.h pr_uid, pr_gid
246 [23656] libc: [mips n32] sys/procfs.h pr_sigpend, pr_sighold, pr_flag have
248 [23679] libc: gethostid: Missing NULL check for gethostbyname_r result
249 [23689] libc: Bug in documentation for rusage.ru_ixrss in
250 bits/types/struct_rusage.h
251 [23690] dynamic-link: Segfault in _dl_profile_fixup with a high number of
253 [23707] dynamic-link: Missing unwind info in sysdeps/powerpc/powerpc32/dl-
255 [23709] string: glibc 2.25 lacks sse2 optimized strstr()
256 [23716] dynamic-link: _dl_runtime_resolve_shstk isn't selected properly
257 [23717] libc: glibc: stdlib/tst-setcontext9 test suite failure on
259 [23724] localedata: Albanian date formats are incorrect
260 [23735] math: libnldbl_nonshared.a references internal libm symbols
261 [23740] localedata: kl_GL: Month names and date formats need update
262 [23744] regex: regex refactorings to remove BE, avoid duplication
263 [23745] time: mktime fix for Gnulib + coreutils
264 [23758] time: Improve the width of alternate representation for year in
266 [23783] libc: [mips] Missing CMSPAR bits/termios.h
267 [23789] time: mktime does not set errno on failure
268 [23791] localedata: Wrong monetary format for ca_ES locale
269 [23793] locale: c32rtomb and mbrtoc32 should not alias wcrtomb and mbrtowc
270 [23794] locale: c16rtomb does not handle surrogate pairs
271 [23821] libc: si_band in siginfo_t has wrong type long int on sparc64
272 [23822] math: ia64 static libm.a is missing exp2f, log2f and powf symbols
273 [23836] time: time/tst-mktime2 test failure on Arm (32-bit)
274 [23848] libc: [sparc] Some socket syscalls wrongly assumed to be present
275 [23861] nptl: rdlock stalls indefinitely on an unlocked pthread rwlock
276 [23862] libc: [sh] missing kernel-features.h undefines
277 [23864] libc: [riscv] missing kernel-features.h undefines
278 [23867] libc: [arm/microblaze] __ASSUME_MLOCK2 incorrect
279 [23907] malloc: Incorrect double-free malloc tcache check disregards
281 [23913] libc: off-by-one in function maybe_script_execute in
282 sysdeps/posix/spawni.c
283 [23915] libc: [arm] __ASSUME_COPY_FILE_RANGE incorrect
284 [23923] locale: Add --no-hard-links option to localedef
285 [23927] network: Linux if_nametoindex() does not close descriptor
287 [23961] math: powf can overflow to inf without setting errno in non-
288 nearest rounding mode
289 [23967] libc: [2.28 Regression]: New sigaction implementation breaks m68k
290 [23972] libc: __old_getdents64 uses wrong d_off value on overflow
291 [23993] libc: glibc 2.29 doesn't build with gcc 4.9
292 [23995] localedata: Remove execution flags from localedata/locales/bi_VU
293 [24011] localedata: Fixed small type in comment for locale bs_BA
294 [24018] libc: gettext() may return NULL
295 [24022] build: riscv build failure with Linux kernel 4.20-rc7
296 [24023] build: [2.29 Regression] FAIL: elf/check-localplt
297 [24024] string: strerror() might set errno to ENOMEM due to -fno-math-
299 [24027] malloc: glibc: realloc() ncopies 32-bit integer overflow
300 [24034] libc: tst-cancel21-static fails with SIGBUS on pre-ARMv7 when
302 [24046] localedata: en_US locale doesn't define date_fmt
303 [24063] manual: @var{errno} should be @code{errno}
304 [24066] soft-fp: Inconsistent _FP_W_TYPE_SIZE check
305 [24088] libc: VSCR field is not being correctly read in ucontext_t on
307 [24096] time: Specifying '_' or '-' flag for "%EY" does not produce the
309 [24097] string: Can't use 64-bit register for size_t in assembly codes for
311 [24110] hurd: SS_DISABLE never set in stack_t value returned by
313 [24112] network: Do not send DNS queries for non-host names (where all
314 answers will be rejected)
315 [24130] libc: alpha __remqu corrupts $f3 register
322 * The localization data for ISO 14651 is updated to match the 2016
323 Edition 4 release of the standard, this matches data provided by
324 Unicode 9.0.0. This update introduces significant improvements to the
325 collation of Unicode characters. This release deviates slightly from
326 the standard in that the collation element ordering for lowercase and
327 uppercase LATIN script characters is adjusted to ensure that regular
328 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
329 is not matched by [a-z]. With the update many locales have been
330 updated to take advantage of the new collation information. The new
331 collation information has increased the size of the compiled locale
332 archive or binary locales.
334 * The GNU C Library can now be compiled with support for Intel CET, AKA
335 Intel Control-flow Enforcement Technology. When the library is built
336 with --enable-cet, the resulting glibc is protected with indirect
337 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
338 compatible with all existing executables and shared libraries. This
339 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
340 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
341 capable of multi-byte NOPs, like x86-64 processors as well as Intel
342 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
343 x86_64 and x32 on non-CET processors. --enable-cet has been tested
344 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
347 * The GNU C Library now has correct support for ABSOLUTE symbols
348 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
349 relocated incorrectly or in some cases discarded. The GNU linker can
350 make use of the newer semantics, but it must communicate it to the
351 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
352 field) to indicate such support is required.
354 * Unicode 11.0.0 Support: Character encoding, character type info, and
355 transliteration tables are all updated to Unicode 11.0.0, using
356 generator scripts contributed by Mike FABIAN (Red Hat).
358 * <math.h> functions that round their results to a narrower type are added
359 from TS 18661-1:2014 and TS 18661-3:2015:
361 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
364 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
367 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
370 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
373 * Two grammatical forms of month names are now supported for the following
374 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
375 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
376 support two grammatical forms in abbreviated month names: Catalan, Greek,
379 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
380 include the support for two grammatical forms of month names.
382 * Building and running on GNU/Hurd systems now works without out-of-tree
385 * The renameat2 function has been added, a variant of the renameat function
386 which has a flags argument. If the flags are zero, the renameat2 function
387 acts like renameat. If the flag is not zero and there is no kernel
388 support for renameat2, the function will fail with an errno value of
389 EINVAL. This is different from the existing gnulib function renameatu,
390 which performs a plain rename operation in case of a RENAME_NOREPLACE
391 flags and a non-existing destination (and therefore has a race condition
392 that can clobber the destination inadvertently).
394 * The statx function has been added, a variant of the fstatat64
395 function with an additional flags argument. If there is no direct
396 kernel support for statx, glibc provides basic stat support based on
397 the fstatat64 function.
399 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
400 library if installed. libidn2 version 2.0.5 or later is recommended. If
401 libidn2 is not available, internationalized domain names are not encoded
402 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
403 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
404 with an encoding error.) Flags which used to change the IDN encoding and
405 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
406 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
407 deprecated. They no longer have any effect.
409 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
410 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
411 range of ELF gABI expressions including such constructs as
412 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
413 have been further restricted, and where in the past a dynamic string
414 token sequence may have been interpreted as a literal string it will
415 now cause a load failure. These load failures were always considered
416 unspecified behaviour from the perspective of the dynamic loader, and
417 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
418 results in a load failure now.
420 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
421 implementation includes all the standard functions provided by
424 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
425 thrd_detach, thrd_exit, and thrd_join for thread management.
427 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
428 mtx_destroy for mutual exclusion.
430 - call_once for function call synchronization.
432 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
433 cnd_wait for conditional variables.
435 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
437 Application developers must link against libpthread to use ISO C threads.
439 Deprecated and removed features, and other changes affecting compatibility:
441 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
442 installed. Software that was using either header should be updated to
443 use standard <stdio.h> interfaces instead.
445 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
446 This was never required by the C standard, and the macros just expanded
447 to call alternative names for the same functions. If you hoped getc and
448 putc would provide performance improvements over fgetc and fputc, instead
449 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
450 necessary, flockfile and funlockfile.
452 * All stdio functions now treat end-of-file as a sticky condition. If you
453 read from a file until EOF, and then the file is enlarged by another
454 process, you must call clearerr or another function with the same effect
455 (e.g. fseek, rewind) before you can read the additional data. This
456 corrects a longstanding C99 conformance bug. It is most likely to affect
457 programs that use stdio to read interactive input from a terminal.
460 * The macros 'major', 'minor', and 'makedev' are now only available from
461 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
462 headers that happen to include <sys/types.h>. These macros are rarely
463 used, not part of POSIX nor XSI, and their names frequently collide with
464 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
467 <sys/sysmacros.h> is a GNU extension. Portable programs that require
468 these macros should first include <sys/types.h>, and then include
469 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
471 * The tilegx*-*-linux-gnu configurations are no longer supported.
473 * The obsolete function ustat is no longer available to newly linked
474 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
475 function has been deprecated in favor of fstatfs and statfs.
477 * The obsolete function nfsservctl is no longer available to newly linked
478 binaries. This function was specific to systems using the Linux kernel
479 and could not usefully be used with the GNU C Library on systems with
480 version 3.1 or later of the Linux kernel.
482 * The obsolete function name llseek is no longer available to newly linked
483 binaries. This function was specific to systems using the Linux kernel
484 and was not declared in a header. Programs should use the lseek64 name
485 for this function instead.
487 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
488 getaddrinfo and getnameinfo functions have been deprecated. The behavior
489 previously selected by them is now always enabled.
491 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
492 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
493 restriction (rejecting '_' in host names, among other things) has been
494 removed, for increased compatibility with non-IDN name resolution.
496 * The fcntl function now have a Long File Support variant named fcntl64. It
497 is added to fix some Linux Open File Description (OFD) locks usage on non
498 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
499 fcntl and LFS support is handled transparently. Also for Linux, the OFD
500 locks act as a cancellation entrypoint.
502 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
503 ecb_crypt, and des_setparity are no longer available to newly linked
504 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
505 longer installed. These functions encrypted and decrypted data with the
506 DES block cipher, which is no longer considered secure. Software that
507 still uses these functions should switch to a modern cryptography library,
510 * Reflecting the removal of the encrypt and setkey functions above, the
511 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
512 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
515 * The obsolete function fcrypt is no longer available to newly linked
516 binaries. It was just another name for the standard function crypt,
517 and it has not appeared in any header file in many years.
519 * We have tentative plans to hand off maintenance of the passphrase-hashing
520 library, libcrypt, to a separate development project that will, we hope,
521 keep up better with new passphrase-hashing algorithms. We will continue
522 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
523 'crypt_r' should not need to change at all; however, distributions will
524 need to install <crypt.h> and libcrypt from a separate project.
526 In this release, if the configure option --disable-crypt is used, glibc
527 will not install <crypt.h> or libcrypt, making room for the separate
528 project's versions of these files. The plan is to make this the default
529 behavior in a future release.
531 Changes to build and runtime requirements:
533 GNU make 4.0 or later is now required to build glibc.
535 Security related changes:
537 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
538 been fixed by removing the glibc-internal IDNA implementation and using
539 the system-provided libidn2 library instead. Originally reported by Hanno
540 Böck and Christian Weisgerber.
542 CVE-2017-18269: An SSE2-based memmove implementation for the i386
543 architecture could corrupt memory. Reported by Max Horn.
545 CVE-2018-11236: Very long pathname arguments to realpath function could
546 result in an integer overflow and buffer overflow. Reported by Alexey
549 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
550 architecture could write beyond the target buffer, resulting in a buffer
551 overflow. Reported by Andreas Schwab.
553 The following bugs are resolved with this release:
555 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
556 [6889] manual: 'PWD' mentioned but not specified
557 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
558 when __WORDSIZE != 64
559 [13762] regex: re_search etc. should return -2 on memory exhaustion
560 [13888] build: /tmp usage during testing
561 [13932] math: dbl-64 pow unexpectedly slow for some inputs
562 [14092] nptl: Support C11 threads
563 [14095] localedata: Review / update collation data from Unicode / ISO
565 [14508] libc: -Wformat warnings
566 [14553] libc: Namespace pollution loff_t in sys/types.h
567 [14890] libc: Make NT_PRFPREG canonical.
568 [15105] libc: Extra PLT references with -Os
569 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
571 [16335] manual: Feature test macro documentation incomplete and out of
573 [16552] libc: Unify umount implementations in terms of umount2
574 [17082] libc: htons et al.: statement-expressions prevent use on global
575 scope with -O1 and higher
576 [17343] libc: Signed integer overflow in /stdlib/random_r.c
577 [17438] localedata: pt_BR: wrong d_fmt delimiter
578 [17662] libc: please implement binding for the new renameat2 syscall
579 [17721] libc: __restrict defined as /* Ignore */ even in c11
580 [17979] libc: inconsistency between uchar.h and stdint.h
581 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
582 [18023] libc: extend_alloca is broken (questionable pointer comparison,
583 horrible machine code)
584 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
586 [18471] libc: llseek should be a compat symbol
587 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
589 [18991] nss: nss_files skips large entry in database
590 [19239] libc: Including stdlib.h ends up with macros major and minor being
592 [19463] libc: linknamespace failures when compiled with -Os
593 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
594 [19527] locale: Normalized charset name not recognized by setlocale
595 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
596 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
598 [19728] network: out of bounds stack read in libidn function
599 idna_to_ascii_4i (CVE-2016-6261)
600 [19729] network: out of bounds heap read on invalid utf-8 inputs in
601 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
602 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
604 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
605 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
606 [20419] dynamic-link: files with large allocated notes crash in
608 [20530] libc: bswap_16 should use __builtin_bswap16() when available
609 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
610 [20980] manual: CFLAGS environment variable replaces vital options
611 [21163] regex: Assertion failure in pop_fail_stack when executing a
612 malformed regexp (CVE-2015-8985)
613 [21234] manual: use of CFLAGS makes glibc detect no optimization
614 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
615 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
616 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
617 [21508] locale: intl/tst-gettext failure with latest msgfmt
618 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
619 [21812] network: getifaddrs() returns entries with ifa_name == NULL
620 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
622 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
624 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
625 [22247] network: Integer overflow in the decode_digit function in
626 puny_decode.c in libidn (CVE-2017-14062)
627 [22342] nscd: NSCD not properly caching netgroup
628 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
629 [22550] localedata: es_ES locale (and other es_* locales): collation
630 should treat ñ as a primary different character, sync the collation
631 for Spanish with CLDR
632 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
633 by gcc configured with --enable-default-pie
634 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
635 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
636 crossing 2GB threshold (CVE-2017-18269)
637 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
639 [22735] time: Misleading typo in time.h source comment regarding
641 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
642 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
643 `__assert_perror_fail ()' to `__assert_fail_base ()'
644 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
646 [22786] libc: Stack buffer overflow in realpath() if input size is close
647 to SSIZE_MAX (CVE-2018-11236)
648 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
649 through an absolute DT_NEEDED path
650 [22792] build: tcb-offsets.h dependency dropped
651 [22797] libc: pkey_get() uses non-reserved name of argument
652 [22807] libc: PTRACE_* constants missing for powerpc
653 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
654 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
655 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
656 [22848] localedata: ca_ES: update date definitions from CLDR
657 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
658 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
659 [22896] localedata: Update locale data for an_ES
660 [22902] math: float128 test failures with GCC 8
661 [22918] libc: multiple common of `__nss_shadow_database'
662 [22919] libc: sparc32: backtrace yields infinite backtrace with
664 [22926] libc: FTBFS on powerpcspe
665 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
667 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
668 [22947] libc: FAIL: misc/tst-preadvwritev2
669 [22963] localedata: cs_CZ: Add alternative month names
670 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
671 [22996] localedata: change LC_PAPER to en_US in es_BO locale
672 [22998] dynamic-link: execstack tests are disabled when SELinux is
674 [23005] network: Crash in __res_context_send after memory allocation
676 [23007] math: strtod cannot handle -nan
677 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
678 [23036] regex: regex equivalence class regression
679 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
680 [23069] libc: sigaction broken on riscv64-linux-gnu
681 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
682 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
683 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
684 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
685 and libc build with -Os)
686 [23140] localedata: More languages need two forms of month names
687 [23145] libc: _init/_fini aren't marked as hidden
688 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
689 [23171] math: C++ iseqsig for long double converts arguments to double
690 [23178] nscd: sudo will fail when it is run in concurrent with commands
691 that changes /etc/passwd
692 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
694 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
695 [23208] localedata: New locale - Lower Sorbian (dsb)
696 [23233] regex: Memory leak in build_charclass_op function in file
698 [23236] stdio: Harden function pointers in _IO_str_fields
699 [23250] nptl: Offset of __private_ss differs from GCC
700 [23253] math: tgamma test suite failures on i686 with -march=x86-64
701 -mtune=generic -mfpmath=sse
702 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
704 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
705 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
707 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
708 [23277] math: nan function should not have const attribute
709 [23279] math: scanf and strtod wrong for some hex floating-point
710 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
712 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
713 [23303] build: undefined reference to symbol
714 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
715 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
717 [23313] stdio: libio vtables validation and standard file object
719 [23329] libc: The __libc_freeres infrastructure is not properly run across
721 [23349] libc: Various glibc headers no longer compatible with
723 [23351] malloc: Remove unused code related to heap dumps and malloc
725 [23363] stdio: stdio-common/tst-printf.c has non-free license
726 [23396] regex: Regex equivalence regression in single-byte locales
727 [23422] localedata: oc_FR: More updates of locale data
728 [23442] build: New warning with GCC 8
729 [23448] libc: Out of bounds access in IBM-1390 converter
730 [23456] libc: Wrong index_cpu_LZCNT
731 [23458] build: tst-get-cpu-features-static isn't added to tests
732 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
734 [23467] dynamic-link: x86/CET: A property note parser bug
741 * The GNU C Library can now be compiled with support for building static
742 PIE executables (See --enable-static-pie in INSTALL). These static PIE
743 executables are like static executables but can be loaded at any address
744 and provide additional security hardening benefits at the cost of some
745 memory and performance. When the library is built with --enable-static-pie
746 the resulting libc.a is usable with GCC 8 and above to create static PIE
747 executables using the GCC option '-static-pie'. This feature is currently
748 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
749 aarch64 with binutils 2.30 or later.
751 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
752 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
755 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
757 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
759 * In order to support faster and safer process termination the malloc API
760 family of functions will no longer print a failure address and stack
761 backtrace after detecting heap corruption. The goal is to minimize the
762 amount of work done after corruption is detected and to avoid potential
763 security issues in continued process execution. Reducing shutdown time
764 leads to lower overall process restart latency, so there is benefit both
765 from a security and performance perspective.
767 * The abort function terminates the process immediately, without flushing
768 stdio streams. Previous glibc versions used to flush streams, resulting
769 in deadlocks and further data corruption. This change also affects
770 process aborts as the result of assertion failures.
772 * On platforms where long double has the IEEE binary128 format (aarch64,
773 alpha, mips64, riscv, s390 and sparc), the math library now implements
774 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
775 These are the same interfaces added in version 2.26 for some platforms where
776 this format is supported but is not the format of long double.
778 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
779 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
780 implements interfaces for that type, as defined by ISO/IEC TS
781 18661-3:2015. These are corresponding interfaces to those supported for
784 * The math library now implements interfaces for the _Float32, _Float64 and
785 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
786 corresponding interfaces to those supported for _Float128.
788 * glibc now implements the memfd_create and mlock2 functions on Linux.
790 * Support for memory protection keys was added. The <sys/mman.h> header now
791 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
794 * The copy_file_range function was added.
796 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
798 * The ldconfig utility now processes `include' directives using the C/POSIX
799 collation ordering. Previous glibc versions used locale-specific
800 ordering, the change might break systems that relied on that.
802 * Support for two grammatical forms of month names has been added.
803 In a call to strftime, the "%B" and "%b" format specifiers will now
804 produce the grammatical form required when the month is used as part
805 of a complete date. New "%OB" and "%Ob" specifiers produce the form
806 required when the month is named by itself. For instance, in Greek
807 and in many Slavic and Baltic languages, "%B" will produce the month
808 in genitive case, and "%OB" will produce the month in nominative case.
810 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
811 are all valid and will all accept any known form of month
812 name---standalone or complete, abbreviated or full. In a call to
813 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
814 the strings used by "%B" and "%b", respectively. New query
815 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
816 used by "%OB" and "%Ob", respectively.
818 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
819 define the strings for %OB and %Ob, respectively; these have the
820 same syntax as "mon" and "abmon". These arrays are optional; if they
821 are not provided then they have the same content as "mon" and "abmon",
824 These features are provided for locales which define "alt_mon" and/or
825 "ab_alt_mon" in their locale source data. This release includes such
826 alternative month name data for the following languages: Belarusian,
827 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
829 This feature is currently a GNU extension, but it is expected to
830 be added to the next revision of POSIX, and it is also already
831 available on some BSD-derived operating systems.
833 This feature will cause existing statically compiled applications
834 to fail to load locales and fall back to the builtin C/POSIX locales.
835 See notes below for other changes affecting compatibility.
837 * Support for the RISC-V ISA running on Linux has been added. This port
838 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
839 for the following ISA and ABI pairs:
845 Deprecated and removed features, and other changes affecting compatibility:
847 * Statically compiled applications attempting to load locales compiled for the
848 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
849 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
850 support for two grammatical forms of the month names, also extends the locale
851 data binary format. Static applications needing locale support must be
852 recompiled to match the runtime and data they are deployed with. In some
853 distributions there is an upgrade window where dynamically linked applications
854 may use a new library but the old locale data and also fall back to the
855 builtin C/POSIX locales; restarting the application process is sufficient to
858 * Support for statically linked applications which call dlopen is deprecated
859 and will be removed in a future version of glibc. Applications which call
860 dlopen need to be linked dynamically instead.
862 * Support for old programs which use internal stdio data structures and
863 functions is deprecated. This includes programs which use the C++ streams
864 provided by libstdc++ in GCC 2.95. Programs which use the internal
865 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
866 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
867 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
868 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
869 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
870 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
871 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
872 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
873 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
874 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
875 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
876 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
877 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
878 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
879 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
880 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
881 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
882 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
883 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
884 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
885 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
886 future version of glibc. Unlike other symbol removals, these old
887 applications will not be supported using compatibility symbols.
889 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
890 defined by <sys/ptrace.h>.
892 * libm no longer supports SVID error handling (calling a user-provided
893 matherr function on error) or the _LIB_VERSION variable to control error
894 handling. (SVID error handling and the _LIB_VERSION variable still work
895 for binaries linked against older versions of the GNU C Library.) The
896 libieee.a library is no longer provided. math.h no longer defines struct
897 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
898 TLOSS, PLOSS and HUGE.
900 * The libm functions pow10, pow10f and pow10l are no longer supported for
901 new programs. Programs should use the standard names exp10, exp10f and
902 exp10l for these functions instead.
904 * The mcontext_t type is no longer the same as struct sigcontext. On
905 platforms where it was previously the same, this changes the C++ name
906 mangling for interfaces involving this type.
908 * The add-ons mechanism for building additional packages at the same time as
909 glibc has been removed. The --enable-add-ons configure option is now
912 * The --without-fp configure option is now ignored. Whether hardware
913 floating-point instructions are used is now configured based on whether
914 the compiler used at configure time (without any options implied by a
915 --with-cpu= configure option) uses such instructions.
917 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
918 the specified string can be parsed as a domain name.
920 * In the malloc_info output, the <heap> element may contain another <aspace>
921 element, "subheaps", which contains the number of sub-heaps.
923 * The libresolv function p_secstodate is no longer supported for new
926 * The tilepro-*-linux-gnu configuration is no longer supported.
928 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
929 and will be removed in a future release. Software that is still using
930 either header should be updated to use standard <stdio.h> interfaces
933 libio.h was originally the header for a set of supported GNU extensions,
934 but they have not been maintained as such in many years, they are now
935 standing in the way of improvements to stdio, and we don't think there are
936 any remaining external users. _G_config.h was never intended for public
937 use, but predates the bits convention.
939 Changes to build and runtime requirements:
941 * bison version 2.7 or later is required to generate code in the 'intl'
944 Security related changes:
946 CVE-2009-5064: The ldd script would sometimes run the program under
947 examination directly, without preventing code execution through the
948 dynamic linker. (The glibc project disputes that this is a security
949 vulnerability; only trusted binaries must be examined using the ldd
952 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
953 suffered from a one-byte overflow during ~ operator processing (either
954 on the stack or the heap, depending on the length of the user name).
955 Reported by Tim RĂŒhsen.
957 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
958 would sometimes fail to free memory allocated during ~ operator
959 processing, leading to a memory leak and, potentially, to a denial
962 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
963 without GLOB_NOESCAPE, could write past the end of a buffer while
964 unescaping user names. Reported by Tim RĂŒhsen.
966 CVE-2017-17426: The malloc function, when called with an object size near
967 the value SIZE_MAX, would return a pointer to a buffer which is too small,
968 instead of NULL. This was a regression introduced with the new malloc
969 thread cache in glibc 2.26. Reported by Iain Buclaw.
971 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
972 to the allocation of too much memory. (This is not a security bug per se,
973 it is mentioned here only because of the CVE assignment.) Reported by
976 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
977 of the number of search path components. (This is not a security
978 vulnerability per se because no trust boundary is crossed if the fix for
979 CVE-2017-1000366 has been applied, but it is mentioned here only because
980 of the CVE assignment.) Reported by Qualys.
982 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
983 for AT_SECURE or SUID binaries could be used to load libraries from the
986 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
987 succeeds without returning an absolute path due to unexpected behaviour
988 of the Linux kernel getcwd syscall. Reported by halfdog.
990 CVE-2018-6485: The posix_memalign and memalign functions, when called with
991 an object size near the value of SIZE_MAX, would return a pointer to a
992 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
994 CVE-2018-6551: The malloc function, when called with an object size near
995 the value of SIZE_MAX, would return a pointer to a buffer which is too
996 small, instead of NULL.
998 The following bugs are resolved with this release:
1000 [866] glob: glob should match dangling symlinks
1001 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
1002 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
1004 [5997] math: Very slow execution of sinf function
1005 [10580] localedata: hr_HR: updated locale
1006 [10871] locale: 'mon' array should contain both nominative and genitive
1008 [12349] localedata: eu_ES: incorrect thousands separator
1009 [13605] localedata: shn_MM: new Shan locale
1010 [13805] localedata: ru_RU: currency should use ',' as radix point
1011 [13953] localedata: km_KH: locale update
1012 [13994] localedata: mjw_IN: new locale
1013 [14121] build: make writes .mo files in po directory
1014 [14333] libc: Fix the race between atexit() and exit()
1015 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
1016 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
1018 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
1019 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
1020 width Latin characters
1021 [15332] localedata: es_CU: locale update
1022 [15436] stdio: Don't close or flush stdio streams on abort
1023 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
1025 [16148] localedata: ca_ES: incorrect thousands separator
1026 [16750] dynamic-link: ldd should not try to execute the binaries
1028 [16777] localedata: pl_PL: incorrect thousands separator in locale
1029 [16905] localedata: hanzi: new collation
1030 [17563] localedata: cmn_TW: add hanzi collation
1031 [17750] localedata: wrong collation order of diacritics in most locales
1032 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
1034 [17956] build: Build fails on missing definitions from header file
1035 nss/nss.h when Mozilla NSS is used for cryptography
1036 [18203] libc: realpath() does not handle unreachable paths correctly
1037 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
1038 [18812] localedata: kab_DZ: new Kabyle Algeria locale
1039 [18822] libc: Internal functions are called via PLT
1040 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
1041 [19170] libc: __gmon_start__ defined in hppa in crtn.S
1042 [19574] libc: glibc should support building static PIE binaries
1043 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
1045 [19971] glob: glob: Do not skip entries with zero d_ino values
1046 [19982] localedata: fr.po: spelling mistake for error code EXDEV
1047 [20008] localedata: km_KH: convert to translit_neutral
1048 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
1049 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
1050 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
1051 [20482] localedata: de_CH: abbreviated weekdays should be two letters
1052 [20498] localedata: miq_NI: new MĂskitu / Miskito (miq) language locale
1054 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
1055 they're set, wrong errors returned by gaih_inet when lookup functions
1057 [20756] localedata: [PATCH] Use Unicode wise thousands separator
1058 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
1060 [20952] localedata: yuw_PG: new locale
1061 [21084] localedata: charmaps/IBM858: new codepage
1062 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
1063 [21242] libc: assert gives pedantic warning in old gcc versions
1064 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
1065 __regcall calling convention
1066 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
1067 [21326] libc: C99 functions are not declared for C++11 and later if
1068 _GNU_SOURCE is not predefined by g++
1069 [21457] libc: sys/ucontext.h namespace
1070 [21530] libc: tmpfile() should be implemented using O_TMPFILE
1071 [21660] math: GCC fails to compile a formula with tgmath.h
1072 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
1073 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
1074 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
1075 [21684] math: tgmath.h handling of complex integers
1076 [21685] math: tgmath.h handling of bit-fields
1077 [21686] math: tgmath.h handling of __int128
1078 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1080 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
1082 [21750] localedata: column width of characters incompatible with classical
1084 [21754] malloc: malloc: Perform as little work as possible after heap
1085 consistency check failures
1086 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
1087 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
1088 [21791] string: Unused XXX_chk_XXX functions in libc.a
1089 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
1091 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
1093 [21845] localedata: Added new Locale bho_NP
1094 [21853] localedata: Fix abday Which looks same as day in zh_SG
1095 [21854] localedata: Added New Locale en_SC
1096 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
1097 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
1098 _dl_runtime_resolve_avx_slow
1099 [21885] network: getaddrinfo: gethosts does not release resolver context
1100 on memory allocation failure
1101 [21899] libc: XPG4.2 sigaction namespace
1102 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
1104 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
1106 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
1107 errno value, causing getaddrinfo to fail
1108 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
1109 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
1110 instead of EAI_NODATA
1111 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
1112 constant PTRACE_SEIZE_DEVEL
1113 [21930] math: C-only gcc builtins used in <math.h> isinf
1114 [21932] network: Unpaired __resolv_context_get in generic get*_r
1116 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
1117 [21944] libc: sigval namespace
1118 [21951] localedata: Update hanzi collation by stroke
1119 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
1120 sysdeps/x86_64/fpu/e_expf.S
1121 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
1122 allocation in syscall loops)
1123 [21959] localedata: Fix Country name for xh_ZA
1124 [21960] localedata: Fix abmon for bem_ZM
1125 [21966] math: AVX2 mathvec functions use FMA without checking
1126 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
1127 [21971] localedata: Added New Locale for mfe_MU
1128 [21972] libc: assert macro requires operator== (int) for its argument type
1129 [21973] math: [sparc] libm missing sqrtl compat symbol
1130 [21974] libc: Remove __bb_init_func and __bb_exit_func
1131 [21982] string: stratcliff.c: error: assuming signed overflow does not
1133 [21986] stdio: __guess_grouping is called incorrectly
1134 [21987] math: [sparc32] wrong bits/long-double.h installed
1135 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
1137 [22022] localedata: Missing country_name for mni_IN
1138 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
1140 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
1142 [22026] locale: iconv_open: heap overflow on gconv_init failure
1143 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
1144 [22035] math: [m68k] bits/math-inline.h macro namespace
1145 [22038] localedata: Fix abbreviated weeks and months for Somali
1146 [22044] localedata: Remove redundant data for Limburgish Language
1147 [22050] malloc: Linking with -lmcheck does not hook
1148 __malloc_initialize_hook correctly
1149 [22051] libc: zero terminator in the middle of glibc's .eh_frame
1150 [22052] malloc: malloc failed to compile with GCC 7 and -O3
1151 [22070] localedata: charmaps/UTF-8: wcwidth for
1152 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
1153 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
1154 Jungseong and Jongseong) should be 0
1155 [22078] nss: nss_files performance issue in multi mode
1156 [22082] math: bits/math-finite.h exp10 condition
1157 [22086] libc: pcprofiledump incorrect cross-endian condition
1158 [22093] dynamic-link: ld.so no longer searches in .../x86_64
1159 [22095] network: Name server address allocation memory leak in resolv.conf
1161 [22096] network: __resolv_conf_attach can incorrectly free passed conf
1163 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
1164 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
1166 [22111] malloc: malloc: per thread cache is not returned when thread exits
1167 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
1168 [22134] libc: [linux] implement fexecve with execveat
1169 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
1171 [22145] libc: ttyname() gives up too early in the face of namespaces
1172 [22146] math: C++ build issue with float128 on x86_64
1173 [22153] nptl: nptl: save error code before process termination
1174 [22156] libc: [hppa,ia64,microblaze] Executable stack default
1175 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
1176 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
1177 [22165] libc: [hppa] Text relocations in libc.so
1178 [22180] libc: destructor registered via __cxa_atexit is called twice
1179 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
1180 symlinks") cause "make" segfaults
1181 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
1183 [22207] libc: FAIL: stdlib/test-atexit-race
1184 [22225] math: nearbyint arithmetic moved before feholdexcept
1185 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
1186 [22235] math: iscanonical in C++ and float128
1187 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
1188 the svid compat wrapper
1189 [22244] math: ynf and yn are wrong without the svid compat wrapper
1190 [22273] libc: Improper assert in Linux posix_spawn implementation
1191 [22284] libc: -pg -pie doesn't work
1192 [22292] locale: localedef exits with error 4 when it should be error 1
1193 [22294] locale: Allow "" for int_currency_symbol definition in locales.
1194 [22295] locale: Don't warn on non-symbolic characters in locale sources in
1196 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
1197 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
1199 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
1200 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
1201 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
1202 [22322] libc: [mips64] wrong bits/long-double.h installed
1203 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
1204 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
1206 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
1207 [22343] malloc: Integer overflow in posix_memalign
1208 [22347] libc: getrandom() returns the number of bytes that were copied to
1209 the buffer even though the comments say "Return 0 on success and -1 on
1211 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1212 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1214 [22370] dynamic-link: Incorrect note padding check
1215 [22375] libc: malloc returns pointer from tcache_get when should return
1216 NULL (CVE-2017-17426)
1217 [22377] math: iseqsig, float128 and C++
1218 [22382] localedata: Error in tpi_PG locale
1219 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1220 inside the ASCII printable range
1221 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1223 [22403] localedata: Slash needs escaping in some locales
1224 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1225 [22409] network: res_hnok does not accept some host names used on the
1227 [22412] network: res_dnok, res_hnok should perform syntax checks
1228 [22413] network: ns_name_pton ignores syntactically invalid trailing
1230 [22415] stdio: setvbuf can lead to invalid free/segfault
1231 [22432] build: Non-deterministic build
1232 [22439] malloc: malloc_info should compute summary statistics for all sub-
1234 [22442] network: if_nametoindex could report index for the wrong
1235 networking interface
1236 [22446] build: aliasing violation calling readlink in handle_request
1237 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1238 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1239 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1240 __nscd_hash/__nss_hash
1241 [22463] network: p_secstodate overflow handling
1242 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1243 [22478] libc: sigwait can fail with EINTR
1244 [22505] libc: ldconfig processes include directive in locale-specific
1246 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1247 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1248 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1249 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1250 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1251 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1252 same as for Croatian
1253 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1254 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1255 [22577] libc: missing newline after "cannot allocate TLS data structures
1257 [22588] manual: manual/conf.texi: missing underscore in front of
1259 [22593] math: nextafter and nexttoward are declared with const attribute
1260 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1261 [22603] string: ia64 memchr overflows internal pointer check
1262 [22605] libc: SH clone does not set the exit code correctly
1263 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1265 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1266 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1267 modified in case of success
1268 [22614] build: gcc: error: unrecognized command line option â-no-pieâ
1269 [22615] manual: manual: ambiguous wording about errno value in case of
1271 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1273 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1274 binaries or if /proc is not mounted (CVE-2017-16997)
1275 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1276 [22630] build: $(no-pie-ldflag) is no longer effective
1277 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1278 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1279 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1280 [22637] nptl: guard size is subtracted from thread stack size instead of
1282 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1283 [22657] localedata: hu_HU: Avoid double space in date
1284 [22660] math: fmax, fmin sNaN handling on alpha
1285 [22664] libc: New warning of GCC8
1286 [22665] math: alpha: ceil and floor raise inexact exceptions
1287 [22666] math: alpha: trunc raise inexact exceptions
1288 [22667] libc: makecontext lacks stack alignment on i386
1289 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1290 [22679] libc: getcwd(3) can succeed without returning an absolute path
1292 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1293 elision and tunables
1294 [22687] math: [powerpc-nofpu] complex long double functions spurious
1296 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1297 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1298 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1300 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1301 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1303 [22701] nis: Incomplete removal of libnsl
1304 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1305 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1306 [22715] dynamic-link: FAIL: elf/tst-audit10
1307 [22719] libc: Backtrace tests fail on hppa
1308 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1309 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1310 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1311 before the first call to crypt_r () in crypt/badsalttest.c
1318 * A per-thread cache has been added to malloc. Access to the cache requires
1319 no locks and therefore significantly accelerates the fast path to allocate
1320 and free small amounts of memory. Refilling an empty cache requires locking
1321 the underlying arena. Performance measurements show significant gains in a
1322 wide variety of user workloads. Workloads were captured using a special
1323 instrumented malloc and analyzed with a malloc simulator. Contributed by
1324 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1326 * Unicode 10.0.0 Support: Character encoding, character type info, and
1327 transliteration tables are all updated to Unicode 10.0.0, using
1328 generator scripts contributed by Mike FABIAN (Red Hat).
1329 These updates cause user visible changes, especially the changes in
1330 wcwidth for many emoji characters cause problems when emoji sequences
1331 are rendered with pango, see for example:
1332 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1334 * Collation of Hungarian has been overhauled and is now consistent with "The
1335 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1338 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1340 - The GNU C Library will now detect when /etc/resolv.conf has been
1341 modified and reload the changed configuration. The new resolver option
1342 âno-reloadâ (RES_NORELOAD) disables this behavior.
1344 - The GNU C Library now supports an arbitrary number of search domains
1345 (configured using the âsearchâ directive in /etc/resolv.conf);
1346 previously, there was a hard limit of six domains. For backward
1347 compatibility, applications that directly modify the â_resâ global
1348 object are still limited to six search domains.
1350 - When the ârotateâ (RES_ROTATE) resolver option is active, the GNU C
1351 Library will now randomly pick a name server from the configuration as a
1352 starting point. (Previously, the second name server was always used.)
1354 * The tunables feature is now enabled by default. This allows users to tweak
1355 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1357 * New function reallocarray, which resizes an allocated block (like realloc)
1358 to the product of two sizes, with a guaranteed clean failure upon integer
1359 overflow in the multiplication. Originally from OpenBSD, contributed by
1360 Dennis Wölfing and RĂŒdiger Sonderfeld.
1362 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1363 These are extended versions of preadv and pwritev, respectively, taking an
1364 additional flags argument. The set of supported flags depends on the
1365 running kernel; full support currently requires kernel 4.7 or later.
1367 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1368 create a new session ID for the spawned process. This feature is
1369 scheduled to be added to the next major revision of POSIX; for the time
1370 being, it is available under _GNU_SOURCE.
1372 * errno.h is now safe to use from C-preprocessed assembly language on all
1373 supported operating systems. In this context, it will only define the
1374 Exxxx constants, as preprocessor macros expanding to integer literals.
1376 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1377 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1378 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1379 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1381 To compile programs that use this feature, the compiler must support
1382 128-bit floating point with the type name _Float128 (as defined by TS
1383 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1384 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1385 must be defined to make the new interfaces visible.
1387 The new functions and macros correspond to those present for other
1388 floating-point types (except for a few obsolescent interfaces not
1389 supported for the new type), with F128 or f128 suffixes; for example,
1390 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1391 printf or scanf formats for the new type; the strfromf128 and strtof128
1392 interfaces should be used instead.
1394 Deprecated and removed features, and other changes affecting compatibility:
1396 * The synchronization that pthread_spin_unlock performs has been changed to
1397 now be equivalent to a C11 atomic store with release memory order to the
1398 spin lock's memory location. Previously, several (but not all)
1399 architectures used stronger synchronization (e.g., containing what is
1400 often called a full barrier). This change can improve performance, but
1401 may affect odd fringe uses of spin locks that depend on the previous
1402 behavior (e.g., using spin locks as atomic variables to try to implement
1403 Dekker's mutual exclusion algorithm).
1405 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1408 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1409 will only be built and installed when the GNU C Library is configured with
1410 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1411 as TIRPC or rpcsvc-proto, to be used.
1413 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1414 libnss_compat, are deprecated, and will not be built or installed by
1417 The NIS(+) support library, libnsl, is also deprecated. By default, a
1418 compatibility shared library will be built and installed, but not headers
1419 or development libraries. Only a few NIS-related programs require this
1420 library. (In particular, the GNU C Library has never required programs
1421 that use 'gethostbyname' to be linked with libnsl.)
1423 Replacement implementations based on TIRPC, which additionally support
1424 IPv6, are available from <https://github.com/thkukuk/>. The configure
1425 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1426 name service modules, to be built and installed.
1428 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1429 support is enabled, the configured recursive resolver must support EDNS.
1430 (Responding to EDNS-enabled queries with responses which are not
1431 EDNS-enabled is fine, but FORMERR responses are not.)
1433 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1434 servers have not supported this opcode for a long time.
1436 * The _res_opcodes variable has been removed from libresolv. It had been
1437 exported by accident.
1439 * <string.h> no longer includes inline versions of any string functions,
1440 as this kind of optimization is better done by the compiler. The macros
1441 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1443 * The nonstandard header <xlocale.h> has been removed. Most programs should
1444 use <locale.h> instead. If you have a specific need for the definition of
1445 locale_t with no other declarations, please contact
1446 libc-alpha@sourceware.org and explain.
1448 * The obsolete header <sys/ultrasound.h> has been removed.
1450 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1452 * The obsolete function cfree has been removed. Applications should use
1455 * The stack_t type no longer has the name struct sigaltstack. This changes
1456 the C++ name mangling for interfaces involving this type.
1458 * The ucontext_t type no longer has the name struct ucontext. This changes
1459 the C++ name mangling for interfaces involving this type.
1461 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1462 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1463 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1464 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1465 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1466 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1467 fpregset_t type no longer has the name struct fpu. This changes the C++
1468 name mangling for interfaces involving those types.
1470 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1471 synced with the kernel:
1473 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1474 are not supported on this architecture and have been removed.
1476 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1477 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1478 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1480 Programs that assume the GET/SETREGS ptrace requests are universally
1481 available will now fail to build, instead of malfunctioning at runtime.
1483 Changes to build and runtime requirements:
1485 * Linux kernel 3.2 or later is required at runtime, on all architectures
1486 supported by that kernel. (This is a change from version 2.25 only for
1489 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1491 * On most architectures, GCC 4.9 or later is required to build the GNU C
1492 Library. On powerpc64le, GCC 6.2 or later is required.
1494 Older GCC versions and non-GNU compilers are still supported when
1495 compiling programs that use the GNU C Library. (We do not know exactly
1496 how old, and some GNU extensions to C may be _de facto_ required. If you
1497 are interested in helping us make this statement less vague, please
1498 contact libc-alpha@sourceware.org.)
1500 Security related changes:
1502 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1503 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1505 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1506 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1508 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1509 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1512 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1513 fixed (CVE-2017-12133).
1515 The following bugs are resolved with this release:
1517 [984] network: Respond to changed resolv.conf in gethostbyname
1518 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1519 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1521 [12189] libc: __stack_chk_fail should not attempt a backtrace
1523 [14096] time: Race condition on timezone/tst-timezone.out
1524 [14172] localedata: az_IR: new locale
1525 [14995] build: glibc fails to build if gold is the default linker, even if
1527 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1528 [16637] network: inet_pton function is accepting IPv6 with bad format
1529 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1530 x86 and other generic code
1531 [16875] localedata: ko_KR: fix lang_name
1532 [17225] localedata: ar_SY: localized month names for May and June are
1534 [17297] localedata: da_DK: wrong date_fmt string
1535 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1536 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1537 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1538 [19066] localedata: ar_SA abbreviated day and month names are in English
1539 [19569] network: resolv: Support an arbitrary number of search domains
1540 [19570] network: Implement random DNS server selection in the stub
1542 [19838] locale: localedef fails on PA-RISC
1543 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1544 order of 0D36 and 0D37
1545 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1547 [20098] libc: FAIL: debug/backtrace-tst on hppa
1548 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1550 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1552 [20313] localedata: Update locale data to Unicode 9.0
1553 [20424] manual: Document how to provide a malloc replacement
1554 [20496] localedata: agr_PE: new language locale AwajĂșn / Aguaruna (agr)
1556 [20686] locale: Add el_GR@euro to SUPPORTED.
1557 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1558 failures consistently
1559 [21015] dynamic-link: Document and fix --enable-bind-now
1560 [21016] nptl: pthread_cond support is broken on hppa
1561 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1562 frame-pointer on i386
1563 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1565 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1566 [21088] libc: Build fails with --enable-static-nss
1567 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1568 generic c code is used
1569 [21109] libc: Tunables broken on big-endian
1570 [21112] math: powf has large ulp errors with base close to 1 and exponent
1572 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1574 [21120] malloc: glibc malloc is incompatible with GCC 7
1575 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1577 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1578 (-0.0) when linking with -lieee
1579 [21171] math: log10, log2 and lgamma return incorrect results
1580 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1581 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1582 [21207] localedata: ce_RU: update weekdays from CLDR
1583 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1584 [21217] localedata: Update months from CLDR-31
1585 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1586 [21243] libc: support_delete_temp_file should issue warning for failed
1588 [21244] libc: support resolv_test_start() socket fd close should be
1590 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1591 new posix_spawn implementation
1592 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1593 leads to lower CPU frequency
1594 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1595 [21261] libc: [sparc64] bits/setjmp.h namespace
1596 [21267] network: [mips] bits/socket.h IOC* namespace
1597 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1598 [21270] libc: mmap64 silently truncates large offset values
1599 [21275] libc: posix_spawn always crashes on ia64 now
1600 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1601 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1602 [21289] libc: Incorrect declaration for 32-bit platforms with
1603 _FILE_OFFSET_BITS=64 causes build error
1604 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1605 not support gethostbyname4_r
1606 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1608 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1610 [21340] libc: Support POSIX_SPAWN_SETSID
1611 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1612 [21359] network: ns_name_pack needs additional byte in destination buffer
1613 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1614 against fragmentation attacks (CVE-2017-12132)
1615 [21369] network: resolv: Remove EDNS fallback
1616 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1618 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1619 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1620 [21393] stdio: Missing dup3 error check in freopen, freopen64
1621 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1622 [21399] localedata: Bad description for U00EC in
1623 localedata/charmaps/CP1254
1624 [21411] malloc: realloc documentation error
1625 [21426] network: sys/socket.h uio.h namespace
1626 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1627 [21445] libc: signal.h bsd_signal namespace
1628 [21455] network: Network headers stdint.h namespace
1629 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1631 [21475] network: resolv: Overlong search path is truncated mid-label
1632 [21511] libc: sigstack namespace
1633 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1634 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1635 bits/syscall.h: No such file or directory
1636 [21517] libc: struct sigaltstack namespace
1637 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1638 [21533] localedata: Update locale data to Unicode 10.0
1640 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1641 junk at end of line, first unrecognized character is `@'
1642 [21538] libc: SIG_HOLD missing for XPG4
1643 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1644 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1645 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1646 [21543] libc: sigevent namespace
1647 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1648 size for general purpose registers in mcontext_t structure
1649 [21550] libc: sigwait namespace
1650 [21552] libc: XPG4 bsd_signal namespace
1651 [21554] libc: sigpause namespace
1652 [21560] libc: sys/wait.h signal.h namespace
1653 [21561] libc: waitid namespace
1654 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1655 '/usr/include/stdlib.h': Permission denied
1656 [21575] libc: sys/wait.h missing struct rusage definition
1657 [21584] libc: sigaltstack etc namespace
1658 [21597] libc: siginterrupt namespace
1659 [21607] math: hppa: FAIL: math/test-tgmath
1660 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1662 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1663 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1664 stack and heap (CVE-2017-1000366)
1665 [21625] libc: wait3 namespace
1666 [21654] nss: Incorrect pointer alignment in NSS group merge result
1668 [21657] network: Parse interface zone id for node-local multicast
1669 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1671 [21666] libc: .symver is used on common symbol
1672 [21668] network: resolv: res_init cross-thread broadcast introduces race
1674 [21687] math: tgmath.h totalorder, totalordermag return type
1675 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1677 [21696] libc: Incorrect assumption of of __cpu_mask in
1678 posix/sched_cpucount.c
1679 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1680 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1682 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1683 leading to relocation crash
1684 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1685 `resolv_conf_matches (resp, conf)' failed.
1686 [21710] localedata: Added Samoan language locale for Samoa
1687 [21711] localedata: Pashto yesstr/nostr locale are missing
1688 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1689 __have_pthread_attr_t can cause redefinition of typedef âpthread_attr_tâ
1690 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1691 [21723] localedata: yesstr/nostr missing for Chinese language locale
1692 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1694 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1696 [21728] localedata: New Locale for Tongan language
1697 [21729] localedata: incorrect LC_NAME fields for hi_IN
1698 [21733] localedata: yesstr and nostr are missing for zh_HK
1699 [21734] localedata: Missing yesstr and nostr are for kw_GB
1700 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1701 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1702 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1703 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1705 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1707 [21749] localedata: Wrong abbreviated day name (âabdayâ) for
1709 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1710 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1711 [21759] localedata: missing yesstr and nostr for Tigrinya
1712 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1713 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1714 [21767] localedata: Missing Bislama locales
1715 [21768] localedata: Missing yesstr and nostr for aa_ET
1716 [21770] localedata: Missing Field in li_NL
1717 [21778] nptl: Robust mutex may deadlock
1718 [21779] libc: MicroBlaze segfaults when loading libpthread
1719 [21783] localedata: Fix int_select international_call_prefixes
1720 [21784] localedata: Inconsistency in country_isbn
1721 [21788] localedata: Missing Country Postal Abbreviations
1722 [21794] localedata: Added-country_isbn-for-Italy
1723 [21795] localedata: Add/Fix country_isbn for France
1724 [21796] localedata: Added country_isbn for Republic of Korea
1725 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1726 [21799] localedata: Added int_select international_call_prefixes
1727 [21801] localedata: Added int_select international_call_prefixes
1728 [21804] nptl: Double semicolon in thread-shared-types.h
1729 [21807] localedata: LC_ADDRESS fix for pap_CW
1730 [21808] localedata: Fix LC_ADDRESS for pap_AW
1731 [21821] localedata: Added country_name in mai_IN
1732 [21822] localedata: Fix LC_TIME for mai_IN
1733 [21823] localedata: missing yesstr, nostr for sa_IN
1734 [21825] localedata: Fix name_mrs for mag_IN
1735 [21828] localedata: 2.26 changelog should mention user visible changes
1737 [21835] localedata: Added Maithili language locale for Nepal
1738 [21838] localedata: Removed redundant data for the_NP
1739 [21839] localedata: Fix LC_MONETARY for ta_LK
1740 [21844] localedata: Fix Latin characters and Months Sequence.
1741 [21848] localedata: Fix mai_NP Title Name
1746 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1747 24731-2:2010, is supported to enable declarations of functions from that
1748 TR. Note that not all functions from that TR are supported by the GNU C
1751 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1752 18661-1:2014, is supported to enable declarations of functions and macros
1753 from that TS. Note that not all features from that TS are supported by
1756 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1757 18661-4:2015, is supported to enable declarations of functions and macros
1758 from that TS. Note that most features from that TS are not supported by
1761 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1762 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1763 Since the GNU C Library defaults to a much newer revision of POSIX, this
1764 will only affect programs that specifically request an old conformance
1765 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1766 see a change in the visible declarations, but a program compiled with
1767 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1770 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1771 defined by all multithreaded code, but glibc has not required this for
1774 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1775 means that in a future release, the macros âmajorâ, âminorâ, and âmakedevâ
1776 will only be available from <sys/sysmacros.h>.
1778 These macros are not part of POSIX nor XSI, and their names frequently
1779 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1780 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1781 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1784 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1785 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1786 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1788 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1789 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1790 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1791 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1792 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1793 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1794 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1795 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1796 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1797 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1798 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1801 * New <math.h> features are added from TS 18661-1:2014:
1803 - Signaling NaN macros: SNANF, SNAN, SNANL.
1805 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1806 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1807 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1809 - llogb functions: the llogb, llogbf and llogbl functions, and the
1810 FP_LLOGB0 and FP_LLOGBNAN macros.
1812 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1815 - Comparison macros: iseqsig.
1817 - Classification macros: iscanonical, issubnormal, iszero.
1819 - Total order functions: totalorder, totalorderf, totalorderl,
1820 totalordermag, totalordermagf, totalordermagl.
1822 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1824 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1825 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1827 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1828 are added to libc. They convert a floating-point number into string.
1830 * Most of glibc can now be built with the stack smashing protector enabled.
1831 It is recommended to build glibc with --enable-stack-protector=strong.
1832 Implemented by Nick Alcock (Oracle).
1834 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1835 intended to be used instead of memset() to erase sensitive data after use;
1836 the compiler will not optimize out calls to explicit_bzero even if they
1837 are "unnecessary" (in the sense that no _correct_ program can observe the
1838 effects of the memory clear).
1840 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1841 to float instead of double. This does not affect the ABI of any libraries
1842 that are part of the GNU C Library, but may affect the ABI of other
1843 libraries that use this type in their interfaces.
1845 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1846 float_t and double_t types are now defined to long double instead of float
1847 and double. These options are not the default, and this does not affect
1848 the ABI of any libraries that are part of the GNU C Library, but it may
1849 affect the ABI of other libraries that use this type in their interfaces,
1850 if they are compiled or used with those options.
1852 * The getentropy and getrandom functions, and the <sys/random.h> header file
1855 * The buffer size for byte-oriented stdio streams is now limited to 8192
1856 bytes by default. Previously, on Linux, the default buffer size on most
1857 file systems was 4096 bytes (and thus remains unchanged), except on
1858 network file systems, where the buffer size was unpredictable and could be
1859 as large as several megabytes.
1861 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1862 for the Linux quota interface which predates kernel version 2.4.22 has
1865 * The malloc_get_state and malloc_set_state functions have been removed.
1866 Already-existing binaries that dynamically link to these functions will
1867 get a hidden implementation in which malloc_get_state is a stub. As far
1868 as we know, these functions are used only by GNU Emacs and this change
1869 will not adversely affect already-built Emacs executables. Any undumped
1870 Emacs executables, which normally exist only during an Emacs build, should
1871 be rebuilt by re-running â./configure; makeâ in the Emacs build tree.
1873 * The âip6-dotintâ and âno-ip6-dotintâ resolver options, and the
1874 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1875 âno-ip6-dotintâ had already been the default, and support for the
1876 âip6-dotintâ option was removed from the Internet in 2006.
1878 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1879 flag from <resolv.h> have been removed. The option relied on a
1880 backwards-incompatible DNS extension which was never deployed on the
1883 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1884 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1885 They were already unimplemented.
1887 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1888 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1889 removed again from the IETF name lookup interface specification in RFC
1890 2553. Applications should use getaddrinfo instead.
1892 * DNSSEC-related declarations and definitions have been removed from the
1893 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1894 decode the data part of DNSSEC record types. Previous versions of glibc
1895 only implemented minimal support for the previous version of DNSSEC, which
1896 is incompatible with the currently deployed version.
1898 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1899 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1900 <arpa/nameser.h> header file because the distinction between RR types and
1901 meta-RR types is not officially standardized, subject to revision, and
1902 thus not suitable for encoding in a macro.
1904 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1905 and rhook members of the res_state type in <resolv.h> have been removed.
1906 The glibc stub resolver did not support these hooks, but the header file
1907 did not reflect that.
1909 * For multi-arch support it is recommended to use a GCC which has
1910 been built with support for GNU indirect functions. This ensures
1911 that correct debugging information is generated for functions
1912 selected by IFUNC resolvers. This support can either be enabled by
1913 configuring GCC with '--enable-gnu-indirect-function', or by
1914 enabling it by default by setting 'default_gnu_indirect_function'
1915 variable for a particular architecture in the GCC source file
1918 * GDB pretty printers have been added for mutex and condition variable
1919 structures in POSIX Threads. When installed and loaded in gdb these pretty
1920 printers show various pthread variables in human-readable form when read
1921 using the 'print' or 'display' commands in gdb.
1923 * Tunables feature added to allow tweaking of the runtime for an application
1924 program. This feature can be enabled with the '--enable-tunables' configure
1925 flag. The GNU C Library manual has details on usage and README.tunables has
1926 instructions on adding new tunables to the library.
1928 * A new version of condition variables functions have been implemented in
1929 the NPTL implementation of POSIX Threads to provide stronger ordering
1932 * A new version of pthread_rwlock functions have been implemented to use a more
1933 scalable algorithm primarily through not using a critical section anymore to
1936 Security related changes:
1938 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1939 have been created with makecontext could fail to terminate due to a
1940 missing .cantunwind annotation. This has been observed to lead to a hang
1941 (denial of service) in some Go applications compiled with gccgo. Reported
1942 by Andreas Schwab. (CVE-2016-6323)
1944 * The DNS stub resolver functions would crash due to a NULL pointer
1945 dereference when processing a query with a valid DNS question type which
1946 was used internally in the implementation. The stub resolver now uses a
1947 question type which is outside the range of valid question type values.
1950 The following bugs are resolved with this release:
1952 [4099] stdio: Overly agressive caching by stream i/o functions.
1953 [7065] build: Support building glibc with -fstack-protector or -fstack-
1955 [9842] localedata: en_CA: incorrect date format
1956 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1957 before it started waiting
1958 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1959 [15765] nptl: sem_open is wrongly a cancellation point
1960 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1961 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1963 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1964 library linked with pthread
1965 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1966 "FIXME: Ingo" issue)
1967 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1968 [17252] libc: getrandom and getentropy syscall
1969 [17730] malloc: thread-local storage is sometimes improperly free()'d
1970 after being __libc_memalign()'d
1971 [18241] stdio: failed fseek on memstream does not set errno and can fail
1973 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1975 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1976 PTHREAD_PRIO_INHERIT mutex on ARM
1977 [18784] network: res_query and related function crash for special record
1978 type queries (CVE-2015-5180)
1979 [19380] math: strtod does not raise "inexact"
1980 [19387] string: Integer overflow in memchr
1981 [19390] string: Integer overflow in strncat
1982 [19398] build: linknamespace tests fail with massively parallel build
1983 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1985 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1986 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1987 [19514] libc: [PATCH] Fix spelling errors in spelling
1988 "implement"/"implementation" in several places
1989 [19582] network: Deprecate RES_USE_INET6
1990 [19673] manual: clog10 docs appear to be erroneous
1991 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1992 causes a segmentation fault
1993 [19826] libc: invalid pointer returned from __tls_get_addr with static
1995 [20016] network: resolv: Remove hooks support from the API
1996 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1998 [20033] math: [x86_64] vectorized math function don't call the __finite
2000 [20116] nptl: use after free in pthread_create
2001 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
2002 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
2004 [20311] nptl: please install proc_service.h
2005 [20366] libc: Compilation errors in installed headers in strict-compliance
2007 [20370] malloc: malloc: Arena free list management is still racy
2008 (incorrect fix in bug 19243)
2009 [20386] libc: assert (X = 0) does not result in GCC warning
2010 [20432] malloc: malloc: Minimize interface required for interposition
2011 [20435] libc: Missing unwind info in __startcontext causes infinite loop
2012 in _Unwind_Backtrace (CVE-2016-6323)
2013 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
2014 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
2016 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
2017 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
2018 [20477] network: resolv: incorrect double-checked locking related to
2020 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
2022 [20495] math: x86_64 performance degradation due to AVX/SSE transition
2024 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
2025 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
2026 cause transition penalty
2027 [20517] math: sparcv9 missing fdiml compat symbol
2028 [20524] manual: strverscmp is inconsistent
2029 [20525] libc: <sys/quota.h> should be based on kernel headers
2030 [20539] math: GCC 7 -static -lm fails to link at -O3
2031 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
2033 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
2035 [20591] network: Remove obsolete DNSSEC support
2036 [20592] network: DNS resource record type classification macros in
2037 <arpa/nameser.h> are incorrect
2038 [20593] network: Update DNS RR type definitions
2039 [20611] network: getaddrinfo accepts invalid numeric scope IDs
2040 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
2042 [20629] network: libresolv: Remove support for bitstring labels
2044 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
2045 assertion: "maxidx >= 2"
2046 [20660] math: [arm] Use VSQRT
2047 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
2048 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
2049 [20689] libc: Test for FMA should also check for AVX.
2050 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
2051 [20715] math: iszero macro breaks existing code
2052 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
2053 [20728] libc: powerpc: Missing TOC stub in clone
2054 [20729] build: glibc-2.24 fails to build for i486 with -Os
2055 [20750] build: Build fails with default PIE enabled gcc-6.x
2056 [20768] math: [alpha] sqrt fegetenv namespace
2057 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
2059 [20787] math: float_t is defined as float incorrectly on x86_64 even if
2060 __FLT_EVAL_METHOD__ is 2
2061 [20790] network: rpcgen buffer overrun in get_prog_declaration
2062 [20822] nptl: race condition in __lll_unlock_elision on powerpc
2063 [20829] libc: crypt snprintf namespace
2064 [20847] libc: tst-vfork3 failure
2065 [20855] math: Default bits/mathdef.h has inappropriate float_t
2066 [20859] math: [sh4] FP_ILOGB0 invalid
2067 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
2068 U+20AC), not same as GBK
2069 [20915] dynamic-link: global-dynamic TLS broken on aarch64
2070 [20916] math: pow handling of sNaN arguments
2071 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
2072 [20919] math: Bad pow (qNaN, 0) result with -lieee
2073 [20940] math: hypot sNaN handling
2074 [20947] math: fmax, fmin sNaN handling
2075 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
2076 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
2078 [20971] string: powerpc64/power7 memchr overflows internal pointer check
2079 [20973] nptl: robust mutexes: Lost wake-ups
2080 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
2081 [20978] nis: strlen on null pointer in nss_nisplus
2082 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
2084 [21014] string: i686 memchr overflows internal pointer check
2085 [21019] libc: [mips] n32 lseek incorrect on overflow
2086 [21022] libc: [microblaze] __backtrace get_frame_size namespace
2087 [21026] libc: [MIPS] readahead syscall is broken on n64
2088 [21028] math: Fallback fesetexceptflag should always succeed
2089 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
2090 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
2092 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
2093 [21061] librt: [microblaze] librt lost clock_* exports
2094 [21073] libc: tunables: insecure environment variables passed to
2095 subprocesses with AT_SECURE
2096 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
2100 * The minimum Linux kernel version that this version of the GNU C Library
2101 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
2102 kernel version 2.6.32 or later suffices (on architectures that already
2103 required kernel versions more recent than 3.2, those requirements remain
2104 unchanged). Linux 3.2 or later kernel headers are required on all
2107 * The pap_AN locale has been deleted. This has been deprecated for a long
2108 time. It has been replaced by pap_AW & pap_CW, both of which have long
2109 been included in previous releases.
2111 * The readdir_r and readdir64_r functions have been deprecated. It is
2112 recommended to use readdir and readdir64 instead.
2114 * The type âunion waitâ has been removed. It was deprecated in the early
2115 1990s and never part of POSIX. Application code should use the int type
2116 instead of âunion waitâ.
2118 * A new NSS action is added to facilitate large distributed system
2119 administration. The action, MERGE, allows remote user stores like LDAP
2120 to be merged into local user stores like /etc/groups in order to provide
2121 easy to use, updated, and managed sets of merged credentials. The new
2122 action can be used by configuring it in /etc/nsswitch.conf:
2123 group: files [SUCCESS=merge] nis
2124 Implemented by Stephen Gallagher (Red Hat).
2126 * The deprecated __malloc_initialize_hook variable has been removed from the
2129 * The long unused localedef --old-style option has been removed. It hasn't
2130 done anything in over 16 years. Scripts using this option can safely
2133 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
2134 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
2135 return the next representable value in the direction of positive infinity
2136 and the nextdown functions return the next representable value in the
2137 direction of negative infinity. These are currently enabled as GNU
2140 Security related changes:
2142 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
2143 could result in a stack overflow when getnetbyname was called with an
2144 overly long name. (CVE-2016-3075)
2146 * Previously, getaddrinfo copied large amounts of address data to the stack,
2147 even after the fix for CVE-2013-4458 has been applied, potentially
2148 resulting in a stack overflow. getaddrinfo now uses a heap allocation
2149 instead. Reported by Michael Petlan. (CVE-2016-3706)
2151 * The glob function suffered from a stack-based buffer overflow when it was
2152 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
2153 Reported by Alexander Cherepanov. (CVE-2016-1234)
2155 * The Sun RPC UDP client could exhaust all available stack space when
2156 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
2157 alloca plugin for GCC. (CVE-2016-4429)
2159 * The IPv6 name server management code in libresolv could result in a memory
2160 leak for each thread which is created, performs a failing naming lookup,
2161 and exits. Over time, this could result in a denial of service due to
2162 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
2164 The following bugs are resolved with this release:
2166 [1170] localedata: ne_NP: update Nepali locale definition file
2167 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
2169 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
2170 [6796] math: fdim() does not set errno on overflow
2171 [10354] libc: posix_spawn should use vfork() in more cases than presently
2172 [11213] localedata: localedata: add copyright disclaimer to locale files
2173 [12143] localedata: chr_US: new Cherokee locale
2174 [12450] localedata: sgs_LT: new locale
2175 [12676] localedata: ln_CD: new locale
2176 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
2178 [13304] math: fma, fmaf, fmal produce wrong results
2179 [14259] build: --localedir arg to configure is ignored
2180 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
2181 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
2182 [14934] localedata: es_CL: wrong first weekday chilean locale
2183 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
2185 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
2187 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
2188 [15368] nptl: raise() is not async-signal-safe
2189 [15479] math: ceil, floor, round and trunc raise inexact exception
2190 [15578] localedata: kk_KZ: various updates
2191 [16003] localedata: pap_AN: punt old locale
2192 [16137] localedata: iw_IL: punt old locale
2193 [16190] localedata: eo: new esperanto locale
2194 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
2195 [16742] malloc: race condition: pthread_atfork() called before first
2196 malloc() results in unexpected locking behaviour/deadlocks
2197 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
2199 [16983] localedata: postal_fmt does not allow %l and %n modifiers
2200 [17565] localedata: pt_PT: wrong (work-)week start
2201 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
2202 [17950] build: Build fails with -msse
2203 [18205] localedata: be_BY*: wrong first_weekday and first_workday
2204 [18433] libc: posix_spawn does not return correctly upon failure to
2206 [18453] localedata: charmaps/IBM875: incorrect codes
2207 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2209 [18896] localedata: he_IL: improvements for currency
2210 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2211 Romanian locale data
2212 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2214 [19056] libc: Deprecate readdir_r
2215 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2217 [19198] localedata: nl_NL: small improvements for Dutch locales
2218 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2219 nameservers (CVE-2016-5417)
2220 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2222 [19400] locale: Language missing in "iso-639.def", trivial fix in
2224 [19431] malloc: Deadlock between fflush, getdelim, and fork
2225 [19505] libc: Incorrect file descriptor validity checks in
2226 posix_spawn_file_actions_add{open,close,dup2}
2227 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2228 when using RTLD_NEXT
2229 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2230 `intl/{gettextP,loadinfo}.h'
2231 [19534] libc: execle, execlp may use malloc
2232 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2233 [19573] network: res_nclose and __res_maybe_init disagree about name
2234 server initialization, breaking Hesiod
2235 [19575] localedata: Status of GB18030 tables
2236 [19581] localedata: sr_* date_fmt string contains additional newline
2237 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2239 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2240 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2241 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2242 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2244 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2246 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2248 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2249 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2250 [19633] locale: strfmon_l applies global locale to number formatting
2251 [19642] network: Memory leak in getnameinfo
2252 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2253 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2255 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2256 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2258 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2259 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2261 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2263 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2265 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2266 pointers and lengths in error-case.
2267 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2268 report errors on UTF-16 surrogates values.
2269 [19755] nscd: nscd assertion failure in gc
2270 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2271 [19759] libc: mempcpy shouldn't be inlined
2272 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2273 [19765] libc: s390 needs an optimized mempcpy
2274 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2275 NAME_MAX limit assumption (CVE-2016-1234)
2276 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2277 [19787] network: Missing and incorrect truncation checks in getnameinfo
2278 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2280 [19791] network: Assertion failure in res_query.c with un-connectable name
2282 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2283 [19822] math: libm.so install clobbers old version
2284 [19825] network: resolv: send_vc can return uninitialized data in second
2285 response to getaddrinfo
2286 [19830] network: nss_dns: should check RDATA length against buffer length
2287 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2288 confronted with A/AAAA records of invalid size
2289 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2290 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2291 [19853] stdio: Printing IBM long double in decimal with high precision is
2293 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2294 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2295 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2297 [19865] network: Assertion failure or memory leak in
2298 _nss_dns_getcanonname_r
2299 [19868] network: nss_dns: netent code does not skip over non-PTR records
2300 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2302 [19881] string: Improve x86-64 memset
2303 [19907] string: Incorrect memcpy tests
2304 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2306 [19925] libc: termios.h XCASE namespace
2307 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2308 [19929] libc: limits.h NL_NMAX namespace
2309 [19931] stdio: Memory leak in vfprintf
2310 [19957] libc: clone(CLONE_VM) access invalid parent memory
2311 [19963] localedata: en_IL: New locale
2312 [19989] stdio: stdio.h cuserid namespace
2313 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2315 [19996] locale: langinfo.h nl_langinfo_l namespace
2316 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2317 [20010] network: getaddrinfo: Stack overflow in hostent translation
2319 [20012] stdio: libio: fmemopen append mode failure
2320 [20014] stdio: stdio.h namespace for pre-threads POSIX
2321 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2322 [20023] libc: fcntl.h timespec namespace
2323 [20024] math: [x86_64] vectorized sincos trashes the stack
2324 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2325 [20041] time: sys/time.h timespec namespace
2326 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2327 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2328 [20051] libc: ttyslot in wrong header under wrong conditions
2329 [20054] libc: gethostname not declared for XPG4
2330 [20055] libc: termios.h missing tcgetsid for XPG4
2331 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2333 [20073] libc: sys/stat.h fchmod namespace
2334 [20074] libc: stdlib.h rand_r namespace
2335 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2336 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2338 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2339 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2341 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2342 [20119] libc: Wrong mask for processors level type from CPUID
2343 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2344 AS not supporting AVX512
2345 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2347 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2348 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2350 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2351 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2352 [20178] libc: posix_spawn{p} should not call exit
2353 [20191] stdio: libio: vtables hardening
2354 [20195] string: FMA4 detection requires CPUID execution with register
2356 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2357 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2359 [20212] math: acos (sNaN) returns sNaN
2360 [20213] math: asin (sNaN) returns sNaN
2361 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2362 [20218] math: [i386] asinhl (sNaN) returns sNaN
2363 [20219] math: [i386] atanhl (sNaN) returns sNaN
2364 [20222] stdio: fopencookie: Mangle function pointers
2365 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2366 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2367 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2369 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2370 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2371 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2372 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2373 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2374 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2375 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2376 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2377 [20240] math: modf (sNaN) returns sNaN
2378 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2379 [20250] math: frexp (sNaN) returns sNaN
2380 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2381 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2383 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2384 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2385 subscript is below array bounds [-Werror=array-bounds]
2386 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2387 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2389 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2390 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2392 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2393 "invalid" exceptions
2394 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2395 [20316] localedata: id_ID: Februari instead of Pebruari
2396 [20327] string: POWER8 strcasecmp returns incorrect result
2397 [20347] math: Failure: Test: j0_downward (0xap+0)
2398 [20348] libc: FAIL: misc/tst-preadvwritev64
2399 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2400 [20350] libc: There is no test for p{read,write}64
2401 [20357] math: Incorrect cos result for 1.5174239687223976
2402 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2406 * Unicode 8.0.0 Support: Character encoding, character type info, and
2407 transliteration tables are all updated to Unicode 8.0.0, using new
2408 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2409 These updates cause user visible changes, such as the fixes for bugs
2410 89, 16061, and 18568.
2412 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2413 kernel-internal CPU set size. This means that requests that change the
2414 CPU affinity which failed before (for example, an all-ones CPU mask) will
2415 now succeed. Applications that need to determine the effective CPU
2416 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2417 setting it because the kernel can adjust it (and the previous size check
2418 would not detect this in the majority of cases).
2420 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2421 the following new symbols are used: fts64_children, fts64_close,
2422 fts64_open, fts64_read and fts64_set.
2424 * getaddrinfo now detects certain invalid responses on an internal netlink
2425 socket. If such responses are received, an affected process will
2426 terminate with an error message of "Unexpected error <number> on netlink
2427 descriptor <number>" or "Unexpected netlink response of size <number> on
2428 descriptor <number>". The most likely cause for these errors is a
2429 multi-threaded application which erroneously closes and reuses the netlink
2430 file descriptor while it is used by getaddrinfo.
2432 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2433 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2434 unnecessary serialization of memory allocation requests across threads.
2435 The defect is now corrected. Users should see a substantial increase in
2436 the concurent throughput of allocation requests for applications which
2437 trigger this bug. Affected applications typically create and
2438 destroy threads frequently. (Bug 19048 was reported and analyzed by
2441 * There is now a --disable-timezone-tools configure option for disabling the
2442 building and installing of the timezone related utilities (zic, zdump, and
2443 tzselect). This is useful for people who build the timezone data and code
2444 independent of the GNU C Library.
2446 * The obsolete header <regexp.h> has been removed. Programs that require
2447 this header must be updated to use <regex.h> instead.
2449 * The obsolete functions bdflush, create_module, get_kernel_syms,
2450 query_module and uselib are no longer available to newly linked binaries;
2451 the header <sys/kdaemon.h> has been removed. These functions and header
2452 were specific to systems using the Linux kernel and could not usefully be
2453 used with the GNU C Library on systems with version 2.6 or later of the
2456 * Optimized string, wcsmbs and memory functions for IBM z13.
2457 Implemented by Stefan Liebler.
2459 * Newly linked programs that define a variable called signgam will no longer
2460 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2461 require signgam to be set by those functions must ensure that they use the
2462 variable provided by the GNU C Library and declared in <math.h>, without
2463 defining their own copy.
2465 * The minimum GCC version that can be used to build this version of the GNU
2466 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2467 still be used to compile programs using the GNU C Library.
2469 Security related changes:
2471 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2472 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2474 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2475 the pointer guard feature. It is always enabled. Previously,
2476 LD_POINTER_GUARD could be used to disable security hardening in binaries
2477 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2480 * An integer overflow in hcreate and hcreate_r could lead to an
2481 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2483 * The catopen function no longer has unbounded stack usage. Reported by
2484 Max. (CVE-2015-8779)
2486 * The nan, nanf and nanl functions no longer have unbounded stack usage
2487 depending on the length of the string passed as an argument to the
2488 functions. Reported by Joseph Myers. (CVE-2014-9761)
2490 * A stack-based buffer overflow was found in libresolv when invoked from
2491 libnss_dns, allowing specially crafted DNS responses to seize control
2492 of execution flow in the DNS client. The buffer overflow occurs in
2493 the functions send_dg (send datagram) and send_vc (send TCP) for the
2494 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2495 family. The use of AF_UNSPEC triggers the low-level resolver code to
2496 send out two parallel queries for A and AAAA. A mismanagement of the
2497 buffers used for those queries could result in the response of a query
2498 writing beyond the alloca allocated buffer created by
2499 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2500 the overflow. Thanks to the Google Security Team and Red Hat for
2501 reporting the security impact of this issue, and Robert Holiday of
2502 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2504 The following bugs are resolved with this release:
2506 [89] localedata: Locales nb_NO and nn_NO should transliterate ÊÞÄ
2507 [887] math: Math library function "logb" and "nextafter" inconsistent
2508 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2509 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2510 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2511 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2513 [4404] localedata: German translation of "Alarm clock" is misleading
2514 [6799] math: nextafter() and nexttoward() doen't set errno on
2515 overflow/underflow errors
2516 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2518 [10432] nis: _nss_nis_setnetgrent assertion failure
2519 [11460] libc: fts has no LFS support
2520 [12926] network: getaddrinfo()/make_request() may spin forever
2521 [13065] nptl: Race condition in pthread barriers
2522 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2523 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2525 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2526 [14912] libc: Rename non-installed bits/*.h headers
2527 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2528 [15367] math: Let gcc use __builtin_isinf
2529 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2530 [15421] math: lgamma wrongly sets signgam for ISO C
2531 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2532 FE_INVALID with argument out of range
2533 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2535 [15786] dynamic-link: ifunc resolver functions can smash function
2537 [15918] math: Unnecessary check for equality in hypotf()
2538 [16061] localedata: Review / update transliteration data
2539 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2541 [16141] time: strptime %z offset restriction
2542 [16171] math: drem should be alias of remainder
2543 [16296] math: fegetround is pure?
2544 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2545 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2547 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2548 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2550 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2551 "Exception "Inexact" set" on ppc32
2552 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2553 [16517] math: Missing underflow exception from tanf/tan/tanl
2554 [16519] math: Missing underflow exception from sinhf
2555 [16520] math: Missing underflow exception from tanhf
2556 [16521] math: Missing underflow exception from exp2
2557 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2559 [16734] stdio: fopen calls mmap to allocate its buffer
2560 [16961] math: nan function incorrect handling of bad sequences
2561 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2562 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2563 [16985] locale: localedef: confusing error message when opening output
2565 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2566 [17197] locale: Redundant shift character in iconv conversion output at
2568 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2569 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2570 [17250] dynamic-link: static linking breaks nss loading
2571 (getaddrinfo/getpwnam/etc...)
2572 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2574 [17441] math: isnan() should use __builtin_isnan() in GCC
2575 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2576 (related to lock elision)
2577 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2578 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2579 [17887] time: strptime should be able to parse "+01:00" style timezones
2580 [17905] libc: catopen() Multiple unbounded stack allocations
2582 [18084] libc: backtrace (..., 0) dumps core on x86
2583 [18086] libc: nice() sets errno to 0 on success
2584 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2585 is too large (CVE-2015-8778)
2586 [18251] dynamic-link: SONAME missing when audit modules provides path
2587 [18265] libc: add attributes for wchar string and memory functions
2588 [18370] math: csqrt missing underflows
2589 [18421] libc: [hppa] read-only segment has dynamic relocations
2590 [18472] libc: Obsolete syscall wrappers should be compat symbols
2591 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2592 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2593 [18525] localedata: Remove locale timezone information
2594 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2595 [18568] localedata: Update locale data to Unicode 8.0
2596 [18589] locale: sort-test.sh fails at random
2597 [18595] math: ctan, ctanh missing underflows
2598 [18604] libc: assert macro-expands its argument
2599 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2600 contains a vector instruction exception.
2601 [18611] math: j1, jn missing errno setting on underflow
2602 [18618] localedata: sync Chechen locale definitions with other *_RU
2604 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2605 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2606 [18665] network: In send_dg, the recvfrom function is NOT always using the
2607 buffer size of a newly created buffer (CVE-2015-7547)
2608 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2610 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2612 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2613 [18699] math: tilegx cproj() for various complex infinities does not yield
2615 [18724] libc: Harden put*ent functions against data injection
2616 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2617 [18755] build: build errors with -DNDEBUG
2618 [18757] stdio: fmemopen fails to set errno on failure
2619 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2621 [18781] libc: openat64 lacks O_LARGEFILE
2622 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2623 error: canât find a register in class âR1_REGSâ while reloading âasmâ
2624 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2625 [18790] math: [ldbl-128ibm] tanhl inaccurate
2626 [18795] libc: stpncpy fortification misses buffer lengths that are
2627 statically too large
2628 [18796] build: build fails for --disable-mathvec
2629 [18803] math: hypot missing underflows
2630 [18820] stdio: fmemopen may leak memory on failure
2631 [18823] math: csqrt spurious underflows
2632 [18824] math: fma spurious underflows
2633 [18825] math: pow missing underflows
2634 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2635 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2636 [18870] build: sem_open.c fails to compile with missing symbol
2638 [18872] stdio: Fix memory leak in printf_positional
2639 [18873] libc: posix_fallocate overflow check ineffective
2640 [18875] math: Excess precision leads incorrect libm
2641 [18877] libc: arm: mmap offset regression
2642 [18887] libc: memory corruption when using getmntent on blank lines
2643 [18918] localedata: hu_HU: change time to HH:MM:SS format
2644 [18921] libc: Regression: extraneous stat() and fstat() performed by
2646 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2647 binaries (CVE-2015-8777)
2648 [18951] math: tgamma missing underflows
2649 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2651 [18953] localedata: lt_LT: change currency symbol to the euro
2652 [18956] math: powf inaccuracy
2653 [18961] math: [i386] exp missing underflows
2654 [18966] math: [i386] exp10 missing underflows
2655 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2656 [18969] build: multiple string test failures due to missing locale
2658 [18970] libc: Reference of pthread_setcancelstate in libc.a
2659 [18977] math: float / long double Bessel functions not in XSI POSIX
2660 [18980] math: i386 libm functions return with excess range and precision
2661 [18981] math: i386 scalb*, ldexp return with excess range and precision
2662 [18982] stdio: va_list and vprintf
2663 [18985] time: Passing out of range data to strftime() causes a segfault
2665 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2666 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2667 [19012] locale: iconv_open leaks memory on error path
2668 [19016] math: clog, clog10 inaccuracy
2669 [19018] nptl: Mangle function pointers in tls_dtor_list
2670 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2671 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2672 [19048] malloc: malloc: arena free list can become cyclic, increasing
2674 [19049] math: [powerpc] erfc incorrect zero sign
2675 [19050] math: [powerpc] log* incorrect zero sign
2676 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2677 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2678 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2679 [19074] network: Data race in _res_hconf_reorder_addrs
2680 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2681 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2682 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2684 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2686 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2687 [19086] manual: posix_fallocate64 documented argument order is wrong.
2688 [19088] math: lround, llround missing exceptions close to overflow
2690 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2691 [19095] math: dbl-64 lrint incorrect for 64-bit long
2692 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2693 [19124] dynamic-link: ld.so failed to build with older assmebler
2694 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2695 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2696 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2697 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2698 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2699 pthread_setaffinity_np
2700 [19156] math: [ldbl-128] j0l spurious underflows
2701 [19164] nptl: tst-getcpu fails with many possible CPUs
2702 [19168] math: math/test-ildoubl and math/test-ldouble failure
2703 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2704 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2706 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2707 (FE_NOMASK_ENV) do not clear SSE exceptions
2708 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2709 _int_new_arena/reused_arena
2710 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2711 [19201] math: dbl-64 remainder incorrect sign of zero result
2712 [19205] math: bits/math-finite.h conditions do not match math.h and
2714 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2715 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2716 for C99-based standards
2717 [19212] libc: features.h not -Wundef clean
2718 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2720 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2721 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2722 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2724 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2725 "inexact" exceptions
2726 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2728 [19242] libc: strtol incorrect in Turkish locales
2729 [19243] malloc: reused_arena can pick an arena on the free list, leading
2730 to an assertion failure and reference count corruption
2731 [19253] time: tzset() ineffective when temporary TZ did not include DST
2733 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2734 [19270] math: [hppa] Shared libm missing __isnanl
2735 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2736 MAP_HUGETLB and MAP_STACK defines
2737 [19313] nptl: Wrong __cpu_mask for x32
2738 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2739 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2740 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2741 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2742 [19363] time: x32: times() return value wrongly truncates/sign extends
2744 [19367] dynamic-link: Improve branch prediction on Silvermont
2745 [19369] network: Default domain name not reset by res_ninit when "search"
2746 / "domain" entry is removed from resolv.conf
2747 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2749 [19385] localedata: bg_BG: time separator should be colon, not comma
2750 [19408] libc: linux personality syscall wrapper may erroneously return an
2751 error on 32-bit architectures
2752 [19415] libc: dladdr returns wrong names on hppa
2753 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2754 IBM905, IBM907, and IBM909
2755 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2756 [19443] build: build failures with -DDEBUG
2757 [19451] build: Make check fails on test-double-vlen2
2758 [19462] libc: Glibc failed to build with -Os
2759 [19465] math: Wrong code with -Os
2760 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2762 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2764 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2765 [19486] math: S390: Math tests fail with "Exception Inexact set".
2766 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2767 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2769 [19590] math: Fail to build shared objects that use libmvec.so functions.
2773 * The following bugs are resolved with this release:
2775 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2776 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2777 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2778 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2779 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2780 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2781 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2782 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2783 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2784 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2785 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2786 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2787 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2788 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2789 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2790 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2791 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2792 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2793 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2794 18657, 18676, 18694, 18696, 18887.
2796 * Cache information can be queried via sysconf() function on s390 e.g. with
2797 _SC_LEVEL1_ICACHE_SIZE as argument.
2799 * A buffer overflow in gethostbyname_r and related functions performing DNS
2800 requests has been fixed. If the NSS functions were called with a
2801 misaligned buffer, the buffer length change due to pointer alignment was
2802 not taken into account. This could result in application crashes or,
2803 potentially arbitrary code execution, using crafted, but syntactically
2804 valid DNS responses. (CVE-2015-1781)
2806 * The time zone file parser has been made more robust against crafted time
2807 zone files, avoiding heap buffer overflows related to the processing of
2808 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2809 large time zone data files. Overly long time zone specifiers in the TZ
2810 variable no longer result in stack overflows and crashes.
2812 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2813 for LD and GD on x86 and x86-64, has been implemented. You will need
2814 binutils-2.24 or later to enable this optimization.
2816 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2817 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2818 Hat). These updates cause user visible changes, such as the fix for bug
2821 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2822 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2823 condition in some applications.
2825 * Added vector math library named libmvec with the following vectorized x86_64
2826 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2828 The library can be disabled with --disable-mathvec. Use of the functions is
2829 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2830 Shared library libmvec.so is linked in as needed when using -lm (no need to
2831 specify -lmvec explicitly for not static builds).
2832 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2834 * A new fmemopen implementation has been added with the goal of POSIX
2835 compliance. The new implementation fixes the following long-standing
2836 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2837 old implementation is still present for use be by existing binaries.
2839 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2840 releases. It has been fixed to match 2.19 and older, but binaries built
2841 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2843 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2844 Contributed by Roland McGrath (Google).
2846 * The header <regexp.h> is deprecated, and will be removed in a future
2847 release. Use of this header will trigger a deprecation warning.
2848 Application developers should update their code to use <regex.h> instead.
2850 This header was formerly part of SUS, but was deprecated in 1994 and
2851 removed from the standard in 2001. Also, the glibc implementation
2852 leaks memory. See BZ#18681 for more details.
2856 * The following bugs are resolved with this release:
2858 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2859 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2860 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2861 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2862 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2863 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2864 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2865 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2866 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2867 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2870 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2871 too little memory for the to-be-scanned arguments and overflow the
2872 allocated buffer. The implementation now correctly computes the required
2873 buffer size when using malloc, and switches to malloc from alloca as
2876 * A new semaphore algorithm has been implemented in generic C code for all
2877 machines. Previous custom assembly implementations of semaphore were
2878 difficult to reason about or ensure that they were safe. The new version
2879 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2880 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2881 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2883 * Port to Altera Nios II has been contributed by Mentor Graphics.
2885 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2886 implementations for powerpc64/powerpc64le.
2887 Implemented by Adhemerval Zanella (IBM).
2889 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2890 and powerpc64le. This may improve lock scaling of existing programs on
2891 HTM capable systems. The lock elision code is only enabled with
2892 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2893 powerpc will issue a transaction abort on every syscall to avoid side
2894 effects being visible outside transactions.
2896 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2897 AArch64. Contributed by ARM Ltd.
2899 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2901 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2902 under certain input conditions resulting in the execution of a shell for
2903 command substitution when the application did not request it. The
2904 implementation now checks WRDE_NOCMD immediately before executing the
2905 shell and returns the error WRDE_CMDSUB as expected.
2907 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2908 processing format strings with a large number of format specifiers.
2910 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2911 infinite loop if the DNS response contained a PTR record of an unexpected
2914 * The minimum GCC version that can be used to build this version of the GNU
2915 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2916 still be used to compile programs using the GNU C Library.
2918 * The GNU C Library is now built with -Werror by default. This can be
2919 disabled by configuring with --disable-werror.
2921 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2923 * The obsolete sigvec function has been removed. This was the original
2924 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2925 programs have been using instead for about 25 years. Of course, ABI
2926 compatibility for old binaries using sigvec remains intact.
2928 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2929 with newer versions of bison.
2931 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2932 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2933 registers overlay two consecutive single-precision registers. MIPS32R2
2934 introduced a new FPU mode (FR=1) where double-precision registers extend the
2935 corresponding single-precision registers which is incompatible with the
2936 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2937 require the use of FR=1 making a transition necessary. New o32 ABI
2938 extensions enable users to migrate over time from the original o32 ABI
2939 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2940 tracks the ABI of any loaded object and verifies that new objects are
2941 compatible. Mode transitions will also be requested as required and
2942 unsupportable objects will be rejected. The ABI checks include both soft and
2943 hard float ABIs for o32, n32 and n64.
2945 GCC 5 with GNU binutils 2.25 onwards:
2946 It is strongly recommended that all o32 system libraries are built using the
2947 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2948 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2949 compiler using --with-fp-32=xx to set this by default.
2953 * The following bugs are resolved with this release:
2955 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2956 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2957 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2958 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2959 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2960 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2961 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2962 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2963 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2964 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2965 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2966 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2967 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2968 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2969 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2971 * Reverted change of ABI data structures for s390 and s390x:
2972 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2973 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2974 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2975 are preserved pointing straight to the same implementation as the old ones.
2976 Given that, new callers will simply provide a too-big buffer to these
2977 functions. Any applications/libraries out there that embed jmp_buf or
2978 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2979 against 2.19 headers will have to rebuilt again. This is necessary in any
2980 case to revert the breakage in their ABI caused by the glibc change.
2982 * Support for file description locks is added to systems running the
2983 Linux kernel. The standard file locking interfaces are extended to
2984 operate on file descriptions, not file descriptors, via the use of
2985 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2986 are associated with an open file instead of a process.
2988 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2990 * The minimum Linux kernel version that this version of the GNU C Library
2991 can be used with is 2.6.32.
2993 * Running the testsuite no longer terminates as soon as a test fails.
2994 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2995 with PASS or FAIL lines for individual tests. A summary of the results is
2996 printed, including a list of failing lists, and "make check" exits with
2997 error status if there were any unexpected failures. "make check
2998 stop-on-test-failure=y" may be used to keep the old behavior.
3000 * The am33 port, which had not worked for several years, has been removed
3003 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
3004 supported; they now act the same as _DEFAULT_SOURCE (but generate a
3005 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
3006 conflicted with POSIX (support for which was removed in 2.19), the
3007 interfaces those macros enabled remain available when compiling with
3008 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
3009 test macros defined.
3011 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
3013 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
3014 This may improve lock scaling of existing programs on TX capable systems.
3015 The lock elision code is only built with --enable-lock-elision=yes and
3016 then requires a GCC version supporting the TX builtins. With lock elision
3017 default mutexes are elided via __builtin_tbegin, if the cpu supports
3018 transactions. By default lock elision is not enabled and the elision code
3021 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
3022 copy the path argument. This allowed programs to cause posix_spawn to
3023 deference a dangling pointer, or use an unexpected pathname argument if
3024 the string was modified after the posix_spawn_file_actions_addopen
3027 * All supported architectures now use the main glibc sysdeps directory
3028 instead of some being in a separate "ports" directory (which was
3029 distributed separately before glibc 2.17).
3031 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
3032 On configurations that support it (all Linux configurations), it's now
3033 used regardless of the --enable-add-ons switch to configure. It is no
3034 longer possible to build such configurations without pthreads support.
3036 * Locale names, including those obtained from environment variables (LANG
3037 and the LC_* variables), are more tightly checked for proper syntax.
3038 setlocale will now fail (with EINVAL) for locale names that are overly
3039 long, contain slashes without starting with a slash, or contain ".." path
3040 components. (CVE-2014-0475) Previously, some valid locale names were
3041 silently replaced with the "C" locale when running in AT_SECURE mode
3042 (e.g., in a SUID program). This is no longer necessary because of the
3045 * On x86-64, the dynamic linker's lazy-binding support is now compatible
3046 with application code using Intel MPX instructions. (With all previous
3047 versions, the MPX register state could be clobbered when making calls
3048 into or out of a shared library.) Note that while the new dynamic
3049 linker is compatible with all known x86 hardware whether or not it
3050 supports Intel MPX, some x86 instruction-set emulators might fail to
3051 handle the new instruction encodings. This is known to affect Valgrind
3052 versions up through 3.9 (but will be fixed in the forthcoming 3.10
3053 release), and might affect other tools that do instruction emulation.
3055 * Support for loadable gconv transliteration modules has been removed.
3056 The support for transliteration modules has been non-functional for
3057 over a decade, and the removal is prompted by security defects. The
3058 normal gconv conversion modules are still supported. Transliteration
3059 with //TRANSLIT is still possible, and the //IGNORE specifier
3060 continues to be supported. (CVE-2014-5119)
3062 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
3063 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
3064 resulting a denial-of-service security vulnerability in applications which
3065 use functions related to iconv. (CVE-2014-6040)
3069 * The following bugs are resolved with this release:
3071 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
3072 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
3073 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
3074 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
3075 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
3076 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
3077 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
3078 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
3079 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
3080 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
3081 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
3082 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
3083 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
3084 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
3085 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
3086 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
3087 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
3088 16431, 16453, 16474, 16506, 16510, 16529
3090 * Slovenian translations for glibc messages have been contributed by the
3091 Translation Project's Slovenian team of translators.
3093 * The public headers no longer use __unused nor __block. This change is to
3094 support compiling programs that are derived from BSD sources and use
3095 __unused internally, and to support compiling with Clang's -fblock
3096 extension which uses __block.
3098 * CVE-2012-4412 The strcoll implementation caches indices and rules for
3099 large collation sequences to optimize multiple passes. This cache
3100 computation may overflow for large collation sequences and may cause a
3101 stack or buffer overflow. This is now fixed to use a slower algorithm
3102 which does not use a cache if there is an integer overflow.
3104 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
3105 rules for large collation sequences to optimize multiple passes and falls
3106 back to alloca if malloc fails, resulting in a possible stack overflow.
3107 The implementation now falls back to an uncached collation sequence lookup
3110 * CVE-2013-4788 The pointer guard used for pointer mangling was not
3111 initialized for static applications resulting in the security feature
3112 being disabled. The pointer guard is now correctly initialized to a
3113 random value for static applications. Existing static applications need
3114 to be recompiled to take advantage of the fix (bug 15754).
3116 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
3117 to the d_name member of struct dirent, or omit the terminating NUL
3118 character. (Bugzilla #14699).
3120 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
3121 aligned_alloc functions could allocate too few bytes or corrupt the
3122 heap when passed very large allocation size values (Bugzilla #15855,
3125 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
3126 for AF_INET6 has been fixed (Bugzilla #16072).
3128 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
3129 pap_CW, quz_PE, the_NP.
3131 * Substantially revised locales: gd_GB, ht_HT
3133 * The LC_ADDRESS field was updated to support country_car for almost all
3136 * ISO 1427 definitions were updated.
3138 * ISO 3166 definitions were updated.
3140 * The localedef utility now supports --big-endian and --little-endian
3141 command-line options to generate locales for a different system from that
3142 for which the C library was built.
3144 * Binary locale files now only depend on the endianness of the system for
3145 which they are generated and not on other properties of that system. As a
3146 consequence, binary files generated with new localedef may be incompatible
3147 with old versions of the GNU C Library, and binary files generated with
3148 old localedef may be incompatible with this version of the GNU C Library,
3149 in the following circumstances:
3151 + Locale files may be incompatible on m68k systems.
3153 + Locale archive files (but not separate files for individual locales) may
3154 be incompatible on systems where plain "char" is signed.
3156 * The configure option --disable-versioning has been removed. Builds with
3157 --disable-versioning had not worked for several years.
3159 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
3160 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
3162 * SystemTap probes for malloc have been introduced.
3164 * SystemTap probes for slow multiple precision fallback paths of
3165 transcendental functions have been introduced.
3167 * Support for powerpc64le has been added.
3169 * The soft-float powerpc port now supports e500 processors.
3171 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
3173 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
3174 set of header declarations that are enabled by default, even when other
3175 feature test macros or compiler options such as -std=c99 would otherwise
3176 disable some of those declarations.
3178 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
3179 conflict with POSIX. The libbsd-compat library (which was a dummy library
3180 that did nothing) has also been removed.
3182 * Preliminary documentation about Multi-Thread, Async-Signal and
3183 Async-Cancel Safety has been added.
3185 * Change of ABI data structures for s390 and s390x:
3186 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
3187 allow for future hardware extensions. All library functions that accept or
3188 return these structures were versioned in order to provide backwards
3189 compatibility with existing code. However, not all cases can be handled
3190 with interface versioning. If an application embeds either structure into
3191 another structure and that structure is passed to another compilation unit
3192 which expects the newer larger-sized object then any access to the new
3193 fields will result in undefined behaviour. Similarly any access to fields
3194 that were moved by the enlarging of the structures will cause undefined
3195 behaviour. To fix the undefined behaviour all objects that are part of the
3196 final in-memory image must agree on the size of structures, and this may
3197 require recompilation.
3201 * The following bugs are resolved with this release:
3203 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
3204 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
3205 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
3206 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
3207 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3208 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3209 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3210 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3211 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3212 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3213 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3214 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3215 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3218 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3219 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3220 Distributions can re-enable building and using pt_chown via the new configure
3221 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3222 considerable security risks and should only be used if the distribution
3223 understands and accepts the risks.
3225 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3228 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3229 fixed (Bugzilla #15330).
3231 * Add support for calling C++11 thread_local object destructors on thread
3232 and program exit. This needs compiler support for offloading C++11
3233 destructor calls to glibc.
3235 * Improved worst case performance of libm functions with double inputs and
3238 * Support for priority inherited mutexes in pthread condition variables on
3239 non-x86 architectures.
3241 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3243 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3245 * Optimized string functions for ARM. Implemented by Will Newton and
3248 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3250 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3253 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3254 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3256 * Added a benchmark framework to track performance of functions in glibc.
3258 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3259 It is based on draft TS 18661 and currently enabled as a GNU extension.
3261 * On Linux, the clock function now uses the clock_gettime system call
3262 for improved precision, rather than old times system call.
3264 * Added support for version-3 tz format files. This is needed when using
3265 the tz database release 2013e or later, and affects a few unusual cases --
3266 currently only TZ='America/Godthab' for time stamps after 2037.
3268 * Added new API functions pthread_getattr_default_np and
3269 pthread_setattr_default_np to get and set the default pthread
3270 attributes of a process.
3272 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3273 This may improve lock scaling of existing programs on TSX capable systems.
3274 When the --enable-lock-elision=yes parameter is specified at configure
3275 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3278 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3279 availability). Implemented by Ryan S. Arnold.
3281 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3285 * The following bugs are resolved with this release:
3287 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3288 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3289 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3290 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3291 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3292 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3293 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3294 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3295 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3296 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3297 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3298 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3299 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3300 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3301 14879, 14889, 14893, 14898, 14914.
3303 * Optimization of memcpy for MIPS.
3305 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3306 EMFILE has been fixed (Bugzilla #14889).
3308 * The add-on ports collection is now distributed in the "ports" subdirectory
3309 of the main GNU C Library distribution, rather than separately.
3311 * Port to ARM AArch64 contributed by Linaro.
3313 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3314 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3316 Implemented by Andreas Krebbel.
3318 * The new function secure_getenv allows secure access to the environment,
3319 returning NULL if running in a SUID/SGID process. This function replaces
3320 the internal function __secure_getenv.
3322 * SystemTap static probes have been added into the dynamic linker.
3323 Implemented by Gary Benson.
3325 * Optimizations of string functions strstr, strcasestr and memmem.
3326 Implemented by Maxim Kuvyrkov.
3328 * The minimum Linux kernel version that this version of the GNU C Library
3329 can be used with is 2.6.16.
3331 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3332 powerpc POWER7. Implemented by Will Schmidt.
3334 * New configure option --disable-nscd builds the C library such that it
3335 never attempts to contact the Name Service Caching Daemon (nscd).
3336 New configure option --disable-build-nscd avoids building nscd itself;
3337 this is the default if --disable-nscd is used.
3339 * Improved support for cross-compilation, including cross-testing and
3340 bootstrap builds without a previously built glibc.
3342 * Several testsuite tests are now able to test multiple IFUNC variants of an
3343 interface, rather than just testing the one that would be chooen by
3346 * New configure options --with-bugurl and --with-pkgversion, for
3347 distributors to use to embed their bug-reporting and package version
3348 information in --help and --version output.
3350 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3351 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3352 allows creation of chroots without the procfs mounted on /proc.
3354 * The `crypt' function now fails if passed salt bytes that violate the
3355 specification for those values. On Linux, the `crypt' function will
3356 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3357 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3358 when the mode is enabled.
3360 * The `clock_*' suite of functions (declared in <time.h>) is now available
3361 directly in the main C library. Previously it was necessary to link with
3362 -lrt to use these functions. This change has the effect that a
3363 single-threaded program that uses a function such as `clock_gettime' (and
3364 is not linked with -lrt) will no longer implicitly load the pthreads
3365 library at runtime and so will not suffer the overheads associated with
3366 multi-thread support in other code such as the C++ runtime library.
3368 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3373 * The following bugs are resolved with this release:
3375 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3376 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3377 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3378 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3379 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3380 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3381 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3382 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3383 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3384 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3385 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3386 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3387 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3388 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3389 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3390 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3391 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3392 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3393 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3394 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3395 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3396 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3399 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3400 configuring glibc with:
3401 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3402 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3403 Implemented by H.J. Lu.
3407 + define static_assert
3409 + do not declare gets
3411 + declare at_quick_exit and quick_exit also for ISO C11
3413 + aligned_alloc. NB: The code is deliberately allows the size parameter
3414 to not be a multiple of the alignment. This is a moronic requirement
3415 in the standard but it is only a requirement on the caller, not the
3418 + timespec_get added
3420 + uchar.h support added
3422 + CMPLX, CMPLXF, CMPLXL added
3424 Implemented by Ulrich Drepper.
3426 * Support for the IA-64 has been moved to ports.
3428 * Remove support for anything but ELF binary format
3430 * Checking versions of poll, ppoll added.
3431 Implemented by Ulrich Drepper.
3433 * More generic and 64-bit performance optimizations to math functions.
3434 Implemented by Ulrich Drepper.
3436 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3437 headers and functions available at compile time as they were before
3438 version 2.14. This option will be removed at some time in the future
3439 after the TI-RPC library becomes fully sufficient for the needs of
3440 existing applications.
3442 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3443 Note that glibc is not expected to work with any Linux kernel version
3446 * New header <sys/auxv.h> and function getauxval allowing easy access to
3447 the AT_* key-value pairs passed from the Linux kernel. The header also
3448 defines the HWCAP_* bits associated with the AT_HWCAP key.
3450 * New locales: mag_IN
3452 * New configure option --enable-systemtap builds SystemTap static probes
3453 into libc for setjmp and longjmp and into libpthread for various operations.
3454 So far the setjmp/longjmp probes and some of the libpthread probes are
3455 provided only for i*86 and x86_64.
3456 Implemented by Roland McGrath and Rayson Ho.
3458 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3460 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3463 * More optimized functions for SPARC. Implemented by David S. Miller.
3465 * Improved support for cross-compilation, especially bootstrap builds
3466 without a previously built glibc.
3468 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3469 Chris Metcalf from Tilera.
3471 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3472 now supported for ARM processors.
3474 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3475 as the name of the dynamic linker, to distinguish it from the
3476 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3478 * Support for CRIS has been removed from ports.
3480 * A new class of installed header has been documented for low-level
3481 platform-specific functionality. PowerPC added the first instance with a
3482 function to provide time base register access. Contributed by Tulio
3483 Magno Quites Machado Filho.
3485 * ix86 configurations now install header files that are consistent with
3486 what x86-64 configurations install. These same header files can be used
3487 for -m32, -m64, or -mx32 builds.
3488 Contributed by H.J. Lu.
3490 * Math library bug fixes. A thorough audit of all open math library bugs was
3491 conducted by Joseph Myers. Significant progress was made on many math
3492 library bugs resulting in more accurate exceptions and function results.
3493 Many thanks to all those that contributed including Andreas Jaeger for his
3494 patch review and work on the x87 trigonometric instruction issues.
3496 * Timezone data is no longer installed. Timezone-related binaries and scripts
3497 will continue to be installed. Users should obtain their timezone data from
3498 their distribution provider or from the tzdata package at
3499 <ftp://munnari.oz.au/pub/>.
3503 * The following bugs are resolved with this release:
3505 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3506 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3507 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3508 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3509 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3510 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3511 13484, 13506, 13515, 13523, 13524, 13538, 13540
3513 * New program pldd to list loaded object of a process
3514 Implemented by Ulrich Drepper.
3516 * Add nss_db support back to glibc. No more dependency on Berkeley db
3517 and support for initgroups lookups.
3518 Implemented by Ulrich Drepper.
3520 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3521 Contributed by HJ Lu.
3523 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3524 Contributed by HJ Lu.
3526 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3527 on x86-32 and x86-64.
3528 Contributed by Liubov Dmitrieva.
3530 * Optimized strchr and strrchr for SSE on x86-32.
3531 Contributed by Liubov Dmitrieva.
3533 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3534 for x86-64 and x86-32.
3535 Contributed by Liubov Dmitrieva.
3537 * New interfaces: scandirat, scandirat64
3538 Implemented by Ulrich Drepper.
3540 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3541 Implemented by Ulrich Drepper.
3543 * nscd now also caches the netgroup database.
3544 Implemented by Ulrich Drepper.
3546 * Integrate libm with gcc's -ffinite-math-only option.
3547 Implemented by Ulrich Drepper.
3549 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3550 to math functions. Implemented by Ulrich Drepper.
3552 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3553 Implemented by Ulrich Drepper.
3555 * New Linux interfaces: process_vm_readv, process_vm_writev
3557 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3558 Implemented by Ulrich Drepper.
3560 * Optimized nearbyint and strcasecmp for PPC.
3561 Implemented by Adhemerval Zanella.
3563 * New locales: bho_IN, unm_US, es_CU, ta_LK
3567 * The following bugs are resolved with this release:
3569 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3570 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3571 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3572 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3573 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3574 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3575 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3576 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3577 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3578 12795, 12811, 12813, 12814, 12841
3580 * The RPC implementation in libc is obsoleted. Old programs keep working
3581 but new programs cannot be linked with the routines in libc anymore.
3582 Programs in need of RPC functionality must be linked against TI-RPC.
3583 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3585 Visible changes of this change include (obviously) the inability to link
3586 programs using RPC functions without referencing the TI-RPC library and the
3587 removal of the RPC headers from the glibc headers.
3588 Implemented by Ulrich Drepper.
3590 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3591 syncfs, setns, sendmmsg
3593 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3594 yue_HK, lij_IT, mhr_RU
3596 * New encodings: CP770, CP771, CP772, CP773, CP774
3598 * New program sotruss to trace calls through PLTs
3599 Implemented by Ulrich Drepper.
3601 * The malloc hook implementation is marked deprecated and will be removed
3602 from the default implementation in the next version. The design never
3603 worked ever since the introduction of threads. Even programs which do
3604 not create threads themselves can use multiple threads created internally.
3608 * The following bugs are resolved with this release:
3610 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3611 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3612 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3613 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3616 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3618 * POWER7 optimizations: memset, memcmp, strncmp
3620 * New optimized string functions for x86-64: strnlen (SSE2),
3621 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3622 Implemented by Ulrich Drepper.
3626 * The following bugs are resolved with this release:
3628 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3629 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3630 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3631 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3632 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3633 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3634 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3635 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3637 * New interfaces: pthread_getname_np, pthread_setname_np
3639 * New Linux interface: recvmmsg
3641 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3643 * The dynamic linker now recognizes supported ABI versions from the
3644 EI_ABIVERSION field in the ELF header.
3645 Implemented by Ulrich Drepper.
3647 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3648 will not be used to fill in password fields in the passwd.byname replies.
3649 Instead it is used to synthesize the shadow.byname table, should it be
3650 missing. This is a useful mode in some installations involving Solaris.
3651 Implemented by Ulrich Drepper.
3653 * New locales: kok_IN, sq_MK, cv_RU
3657 * The following bugs are resolved with this release:
3659 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3660 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3661 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3662 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3663 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3664 10789, 10817, 10823, 10840, 10847
3666 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3668 Implemented by Ulrich Drepper.
3670 * Checking version of longjmp added that fails if an uninitialized stack
3671 frame would be created. Implemented by Ulrich Drepper.
3673 * STT_GNU_IFUNC is now supported in static executables.
3674 Implemented by H.J. Lu.
3676 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3678 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3679 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3680 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3681 Contributed by H.J. Lu.
3683 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3684 Implemented by Ulrich Drepper.
3686 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3688 Contributed by H.J. Lu.
3690 * Support for fma instruction in AVX on x86-64.
3691 Implemented by H.J. Lu and Ulrich Drepper.
3693 * AVX support in x86-64 auditing support in ld.so.
3694 Implemented by H.J. Lu.
3696 * STB_GNU_UNIQUE support added.
3697 Implemented by Ulrich Drepper.
3699 * Implement second fallback mode for optimized DNS lookup for even more
3700 broken environments. If two requests from the same port are not
3701 handled correctly close the socket and open a new one before sending
3702 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3703 can be used to select this mode right away, instead of rediscovering the
3704 necessity is every process again.
3705 Implemented by Ulrich Drepper.
3707 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3708 Implemented by Adam Tkac.
3710 * Optimized iconv conversions for S390x.
3711 Implemented by Andreas Krebbel.
3713 * Using condvars with PI mutexes is now more efficient due to kernel
3714 support for requeueing to PI futexes. NPTL support added for x86-64.
3715 Implemented by Ulrich Drepper.
3717 * New locales: ps_AF, my_MM
3721 * The following bugs are resolved with this release:
3723 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3724 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3725 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3726 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3727 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3729 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3730 Implemented by Ulrich Drepper.
3732 * New Linux interfaces: accept4, fallocate, fallocate64.
3733 Implemented by Ulrich Drepper.
3735 * Correct declarations of string function when used in C++ code. This
3736 could lead to compile errors for invalid C++ code.
3738 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3741 * New POSIX 2008 interface: psiginfo
3742 Implemented by Ulrich Drepper.
3744 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3745 Implemented by Ulrich Drepper.
3747 * Support for selecting between multiple function definitions at runtime
3748 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3750 * The libcrypt library can now use the hash function implementations in
3751 NSS. Implemented by Ulrich Drepper.
3753 * The malloc implementation can be compiled to be less memory efficient
3754 but higher performing in multi-threaded programs.
3755 Implemented by Ulrich Drepper.
3757 * New locales: nan_TW@latin, ks_IN
3759 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3760 Implemented by Ulrich Drepper.
3762 * Extended printf hook support. It is possible to use user-defined types
3763 and extend existing format specifiers.
3764 Implemented by Ulrich Drepper.
3766 * Handling for group shadow files has been added.
3767 Implemented by Ulrich Drepper.
3769 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3770 servers (the case, e.g., for some people using the built-in DNS
3771 server in ADSL modems/routers). There is a once-per-process timeout
3772 in case of a broken server. To avoid it, users can run nscd or put
3773 'options single-request' in /etc/resolv.conf.
3774 Implemented by Ulrich Drepper.
3778 * The following bugs are resolved with this release:
3780 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3781 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3782 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3783 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3784 6968, 6974, 6980, 6995, 7008, 7009, 7029
3786 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3787 up at the same time. Implemented by Ulrich Drepper.
3789 * TLS descriptors for LD and GD on x86 and x86-64.
3790 Implemented by Alexandre Oliva.
3792 * getaddrinfo now handles DCCP and UDPlite.
3793 Implemented by Ulrich Drepper.
3795 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3796 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3797 Implemented by Ulrich Drepper.
3799 * New implementation of memmem, strstr, and strcasestr which is O(n).
3800 Implemented by Eric Blake.
3802 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3804 * Implement "e" option for popen to open file descriptor with the
3805 close-on-exec flag set. Implemented by Ulrich Drepper.
3807 * Implement "b" mode for fmemopen. In this mode writes writes don't
3808 implicitly add a NUL byte and seeks from the end of the buffer really
3809 use the buffer end, not the string length as the basis.
3810 Implemented by Ulrich Drepper.
3812 * Many functions, exported and internal, now atomically set the close-on-exec
3813 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3815 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3817 Implemented by Pravin Satpute.
3819 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3823 * The following bugs are resolved with this release:
3825 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3826 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3827 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3828 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3829 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3830 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3831 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3833 * New locales: bo_CN, bo_IN, shs_CA.
3835 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3837 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3838 Implemented by Pravin Satpute.
3840 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3842 * nscd caches now all timeouts for DNS entries
3843 Implemented by Ulrich Drepper.
3845 * nscd is more efficient and wakes up less often.
3846 Implemented by Ulrich Drepper.
3848 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3849 vdprintf, and obstack_vprintf.
3850 Implemented by Jakub Jelinek.
3852 * Faster memset for x86-64.
3853 Implemented by Harsha Jagasia and H.J. Lu.
3855 * Faster memcpy on x86.
3856 Implemented by Ulrich Drepper.
3858 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3859 Implemented by Ulrich Drepper.
3861 * Faster sqrt and sqrtf implemention for some PPC variants.
3862 Implemented by Steven Munroe.
3866 * The following bugs are resolved with this release:
3868 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3869 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3870 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3871 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3872 5071, 5103, 5104, 5112, 5113, 5184, 5186
3874 * More checking functions: fread, fread_unlocked, open*, mq_open.
3875 Implemented by Jakub Jelinek and Ulrich Drepper.
3877 * Extend fortification to C++. Implemented by Jakub Jelinek.
3879 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3880 by not recognizing 'a' as a modifier when those specs are requested.
3881 Implemented by Jakub Jelinek.
3883 * PPC optimizations to math and string functions.
3884 Implemented by Steven Munroe.
3886 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3887 options to be passed. Implemented by Ulrich Drepper.
3889 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3891 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3892 Implemented by Ulrich Drepper.
3894 * Handle private futexes in the NPTL implementation.
3895 Implemented by Jakub Jelinek and Ulrich Drepper.
3897 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3898 Implemented by Roland McGrath and Ulrich Drepper.
3900 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3902 * SHA-256 and SHA-512 based password encryption.
3903 Implemented by Ulrich Drepper.
3905 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3906 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3909 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3910 Implemented by Ulrich Drepper.
3914 * The following bugs are resolved with this release:
3916 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3917 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3918 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3919 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3920 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3921 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3922 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3925 * New Linux interfaces: epoll_pwait, sched_getcpu.
3927 * New generic interfaces: strerror_l.
3929 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3933 * The following bugs are resolved with this release:
3935 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3936 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3937 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3938 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3939 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3940 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3941 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3942 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3943 3225, 3251, 3252, 3253, 3273
3945 * For Linux, the sorting of addresses returned by getaddrinfo now also
3946 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3947 Implemented by Ulrich Drepper.
3949 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3950 Implemented by Ulrich Drepper.
3952 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3954 * New iconv module for MIK. Contributed by Alexander Shopov.
3956 * For sites with broken group and/or passwd database, the auto-propagate
3957 option of nscd can prevent creating ID lookup entries from the results
3958 of a name lookup and vice versa. This usually is no problem but some
3959 site might have problems with the default behavior.
3960 Implemented by Ulrich Drepper.
3962 * Iterating over entire database in NIS can be slow. With the
3963 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3964 to trade time for memory. The entire database will be read at once.
3965 Implemented by Ulrich Drepper.
3967 * The interfaces introduced in RFC 3542 have been implemented by
3970 * Support for the new ELF hash table format was added by Ulrich Drepper.
3972 * Support for priority inheritance mutexes added by Jakub Jelinek and
3975 * Support for priority protected mutexes added by Jakub Jelinek.
3977 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3978 or_IN, csb_PL, fy_NL, sr_ME.
3982 * More overflow detection functions.
3984 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3985 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3987 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3988 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3989 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3990 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3991 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3992 by Masahide Washizawa.
3994 * It is now possible to install audit libraries for the dynamic linker using
3995 LD_AUDIT. Implemented by Ulrich Drepper.
3997 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3998 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3999 For a libc and libpthread that works well on Linux 2.4 kernels, we
4000 recommend using the stable 2.3 branch.
4002 * The new function open_wmemstream acts like open_memstream,
4003 but uses a wchar_t wide character buffer.
4005 * The new function ppoll is an improved version of poll, similar to pselect.
4007 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
4008 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
4009 renameat, unlinkat, linkat, symlinkat, readlinkat.
4011 * New Linux kernel system calls: unshare,
4012 inotify_init, inotify_add_watch, inotify_rm_watch.
4014 * The euidaccess function is now also known by the alias eaccess,
4015 for compatibility with some other systems.
4017 * Timezone data updated to 2006b version.
4021 * The following bugs are resolved with this release:
4023 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
4024 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
4025 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
4026 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
4027 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
4028 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
4030 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4032 * As of this release, GCC 4 can be used to compile the C Library.
4034 * Timezone data updated to 2005m version.
4038 * The following bugs are resolved with this release:
4040 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
4041 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
4042 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
4043 777, 787, 821, 822, 823, 825
4045 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4049 * Support for RFC 3678. Real implementations exist only for Linux so far.
4050 Implemented by Ulrich Drepper.
4052 * nscd can now cache entries persistently. Expiring entries are reloaded.
4053 For speedups the cache can be shared in memory with client processes.
4054 Implemented by Ulrich Drepper.
4056 * nscd can now perform SELinux checks.
4057 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
4059 * getaddrinfo queries are now cached. Canonical name lookup is performed
4061 Implemented by Ulrich Drepper.
4063 * The nothrow function attribute is used when headers are used by gcc when
4064 compiling C code. This can avoid significant amounts of exception
4067 * The malloc functions perform more error checking and are stricter when
4068 it comes to reacting on errors. The default action is to terminate
4069 the process after showing an error message. Implemented by Ulrich Drepper.
4071 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
4072 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
4074 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
4075 namespaces using the new function dlmopen(). This feature is of course,
4076 like most other dynamic loading functionality, not available in statically
4077 linked applications. Implemented by Ulrich Drepper.
4079 * Low-overhead boundary checking variants of string and some stdio functions
4080 were added. These are to be used in conjunction with a gcc patch by
4081 Jakub Jelinek which adds calls to these functions if possible.
4082 Implemented by Jakub Jelinek and Ulrich Drepper.
4084 * Old code for several operating systems and machine architectures that
4085 have not been in working condition in a long time have been removed from
4086 the main source tree maintained by the GNU C Library's maintainers.
4087 These files are now reside in the separate `ports' source module
4088 that is usable as an add-on when building the library.
4092 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
4093 interrogate the dynamic linker, compatible with the Solaris interface.
4095 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
4096 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
4098 * getifaddrs now uses the netlink interface on Linux to get its information.
4099 Implemented by Thorsten Kukuk.
4101 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
4102 Implemented by Ulrich Drepper.
4104 * support for non-executable stacks on x86 has been added. Changes mostly
4107 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
4110 * getaddrinfo now performs destination address selection according to
4115 * Thread-safe interfaces for many functions that access locale data
4116 were added in version 2.3, but these features were omitted from NEWS.
4117 Many functions have variants with an `_l' suffix that take a `locale_t'
4118 object as a parameter rather than consulting the current locale.
4119 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
4120 create and maintain `locale_t' objects. Additionally, the new function
4121 `uselocale' sets "the current locale" (as used by functions not so
4122 parameterized) set for an individual thread. These features were added
4123 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
4125 * The functions getresuid, getresgid, setresuid, and setresgid, which
4126 have long been available on Linux, are now declared in <unistd.h>
4127 and are now also available on the Hurd.
4129 * ELF thread-local storage support (TLS) now works on x86-64.
4131 * The new dynamic string token $LIB is expanded in shared library names.
4132 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
4134 * Aldy Hernandez contributed complete software floating point support for
4135 PowerPC machines with no FPU.
4137 * fexecve is implemented on Linux.
4139 * The `btowc' function should work at least twice as fast due to
4140 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
4142 * With approriate thread add-ons cancelable functions are now implemented
4143 in libc.so as well. No need to call the function in libpthread. This
4144 change allowed to finally disable the incorrect and expensive handling
4145 of weak definition in ld.so.
4147 * Yet more PLT entries in libc.so have been removed. We finally arrived
4148 at the bare minimum. Startup times improved appropriately.
4150 * Support for the new Linux/x86 system call interface was added. The
4151 AT_SYSINFO auxiliary vector entry is recognized and handled.
4155 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
4158 * iconv (the program and the interface) now accepts empty names (excluding
4159 options like //TRANSLIT) to mean "use charset of current locale".
4161 * localedef can now transliterate characters in strings which are not in
4162 the provided charmap. The information from the input locale is used.
4164 * Prelinking support was added for ELF targets. This requires additional
4165 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
4167 * Read-only stdio streams now use mmap to speed up operation by eliminating
4168 copying and buffer underflows. To use add 'm' to the mode string of
4169 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
4171 * The malloc functions were completely rewritten by Wolfram Gloger based
4172 on Doug Lea's malloc-2.7.0.c.
4174 * Isamu Hasegawa contributed a completely new and POSIX-conformant
4175 implementation of regex.
4177 * Bruno Haible upgraded the iconv and locale implementation to support
4180 * Contents of the LC_* and LANG environment variables in the CEN style are
4181 not recognized anymore. It never was used. Change by Ulrich Drepper.
4183 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
4184 thread-local storage (TLS) ABI on some platforms.
4185 Changes by Ulrich Drepper. SH support by Kaz Kojima.
4187 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
4188 EUC-JISX0213, and TSCII.
4190 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
4191 BSD-compatible interface for getting all network interface addresses.
4192 Implementation for IPv4 by Roland McGrath.
4194 * Loading of locale data is faster due to the introduction of a locale
4195 archive. Implemented by Roland McGrath and Ulrich Drepper.
4197 * Startup times are significantly reduced by not using exported functions
4198 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
4201 * Steven Munroe contributed a port to PowerPC64/Linux.
4205 * The Hurd now uses the GNU libio implementation of stdio rather than the
4206 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
4208 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4209 with the same behavior as the Linux system call of the same name.
4213 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4214 128-bit long double format.
4216 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4217 IBM1161, and IBM1162 charsets.
4219 * Andreas Jaeger contributed a port to x86-64/Linux.
4221 * Peter Bruin contributed a port to PowerPC/Hurd.
4223 * libc functions using I/O streams now can handle wide-oriented streams
4226 * optimizations in the dynamic linker. Binaries created by recent binutils
4227 versions start up quicker due to reduced time spent on relocations.
4229 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4233 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4234 asinh, atanh, j0 for the 128-bit long double format.
4236 * Bruno Haible updated all the code handling Unicode in some form to
4237 support Unicode 3.1.
4239 * Speed of regex for single-byte locales is back to previous levels.
4240 Patch by Isamu Hasegawa.
4242 * Alpha, SPARC, and IA-64 now also using floating stacks.
4244 * Startup time of internationalized applications greatly improved through
4245 iconv cache. Use iconvconfig to generate the cache file.
4246 Contributed by Ulrich Drepper.
4248 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4249 severe performance problems. Patch by David Mosberger.
4251 * The Hurd port got a lot more functionality like AIO, various stdio
4252 extensions, etc. Mainly done by Roland McGrath.
4254 * mtrace can now lookup symbols in shared libraries.
4258 * Intel's IA-64 math library is largely integrated. It provides fast and
4259 accurate implementatations for most basic and standard math functions
4260 in float, double, and long double format.
4262 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4263 and asin for the 96-bit long double format and asin, log, tan for the
4264 128-bit long double format.
4266 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4267 The basic double functions exist today. Contributed by Abraham Ziv
4268 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4269 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4271 * An asynchronous name lookup library was added. The interface is designed
4272 after POSIX AIO. The proposal was circulated beforehand to get comments.
4273 No negative ones came in. Implemented by Ulrich Drepper.
4275 * Port to S390/64bit contributed by Martin Schwidefsky
4276 <schwidefsky@de.ibm.com>.
4278 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4279 of functions for Linux/IA-64.
4281 * The RPC code is now thread safe. Threads can now use the same service
4282 of different services at the same time. Patch by Eric Norum
4283 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4285 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4286 family of functions for Linux/S390.
4288 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4289 of functions for Linux/x86.
4291 * Port to Linux/CRIS contributed by Axis Communications.
4295 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4296 we can now check for namespace violations and missing declarations. The
4297 result is that almost all headers are now Unix-compliant (as defined in
4298 the upcoming XPG6). The negative side is that some programs might need
4299 corrections, too, if they depend on the incorrect form of the headers in
4300 previous versions which defined too many symbols and included too many
4303 * regex now handles multibyte character sets correctly.
4304 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4306 * iconv (the program) does now conform to the upcoming XPG6 and handles
4307 charmaps. Instead of the charset names the path of charmaps can be
4308 provided and the conversion happens based on this data.
4309 Contributed by Ulrich Drepper.
4311 * The locale program now provides detailed information about the installed
4316 only lists the names of the supported locales
4320 provides details such as country, language, and codeset name.
4321 Contributed by Ulrich Drepper.
4325 * The gencat program now parses the input file according to the charset
4326 selected by the LC_CTYPE category. This is important for stateful
4327 character sets. To make generating catalogs easier there is a way
4328 to overwrite the charset selected by the locale: before the first
4329 message or $ quote line the catalog can contain a line like
4331 $ codeset=ISO-8859-2
4333 to select the charset (ISO-8859-2 in this case).
4335 Implemented by Shinya Hanataka and Ulrich Drepper.
4337 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4338 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4339 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4341 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4342 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4344 * The LANGUAGE environment variable is now ignored unless the locale is
4345 changed from the default "C" locale.
4347 * The usual bug fixes.
4351 * Greg McGary added runtime support for bounds checking using gcc's
4352 new -fbounded-pointers option. ix86 target is complete. PowerPC
4355 * Thorsten Kukuk added secure mode support to nscd.
4357 * The Berkeley db libraries have been removed.
4359 Related, the nss_db module is now in a separate package since it
4360 obviously requires a database library being available.
4362 * Wide character I/O streams implemented by Ulrich Drepper.
4364 * Functions from the extended socket API added by Ulrich Drepper.
4366 * Functions feenableexcept and fedisableexcept to control the
4367 behaviour of individual exceptions have been added by Andreas Jaeger.
4369 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4371 * The resolver code has been updated from bind 8.2.3-T5B which supports
4372 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4375 This change could in some situations effect backward compatibility. Since
4376 now `_res' is a thread-local instead of a global variable, modifying it
4377 in one thread does not have any effect in other threads.
4379 The resolver library was also extended to allow IPv6 as the transport
4380 protocol for the requests. This work was done by Stig Venaas.
4382 * Compatibility code for K&R C compilers has been removed from the
4383 header files. A ISO C compiler is needed to use the library
4384 (conforming to either C89 or C99 standard).
4386 * Complete rewrite of the localedef program to support multibyte character
4387 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4388 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4389 Implemented by Ulrich Drepper.
4391 Bruno Haible significantly improved the generation and use of the data
4392 structures for the wide character tables.
4394 * Plural handling in gettext implemented by Ulrich Drepper.
4396 * The utmp daemon has been removed.
4398 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4400 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4403 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4405 * POSIX barriers implemented by Kaz Kylheku.
4407 * POSIX spawn function family implemented by Ulrich Drepper.
4409 * POSIX spinlocks are now available.
4411 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4413 * the configure option --enable-kernel=X.Y.Z allows to strip out
4414 compatibility for kernel versions before X.Y.Z. This is currently only
4415 implemented for Linux.
4417 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4418 field sin6_scope_id. This means that all programs using IPv6 should be
4419 recompiled. Don't expect binary compatibility with previous glibc
4422 * various conversion modules for IBM character sets contributed by
4425 * IA-64 port by Jes Sorensen and HJ Lu.
4439 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4441 * New xdr functions are added; some rpc functions are now 64bit clean.
4443 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4445 * Fixed known incompatibilities with glibc 2.0.
4447 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4449 * Optimized a number of functions (especially the ELF dynamic loader).
4451 * Update timezone data files.
4453 * lots of charmaps corrections
4455 * some new locale definitions and charmaps
4460 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4461 conform to POSIX member type requirements. Symbol versions have been
4462 adjusted as needed within the library, and for direct use by applications,
4463 but there is potential for problems if third-party libraries use
4464 struct timeval as part of their interface. This does not present
4465 a problem for X and other "essential" system libraries.
4467 * An additional locale model to support C++ Standard Library locale
4468 model and probably more was implemented by Ulrich Drepper.
4470 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4473 * Miles Bader provided the `argp' function family to support hierachical
4474 command line argument parsing, layered on top of getopt.
4476 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4478 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4481 * scanf recognizes the %a and %A format for scanning floating point numbers.
4483 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4484 information and interfaces for the available integer types.
4486 * about 130 new math functions were added to implement the ISO C9x math
4489 * the new header <complex.h> contains definitions of the complex math
4490 functions from ISO C 9X.
4492 * the new header <tgmath.h> defines generic macros to use complex or
4493 real valued functions.
4495 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4497 * Andreas Jaeger provided a test suite for the math library.
4499 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4501 * Ulrich Drepper added character set conversion functions (iconv).
4503 * Optimized string functions have been added.
4505 * The localedata addon is now part of glibc.
4507 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4509 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4510 daemon for NSS (nscd).
4512 Missing a better place here are some numbers on improvements. Under
4513 Linux 2.1.125 un-tar-ing the kernel sources takes
4517 using local files 12.19s 6.88s 22.91s
4519 using NIS 13.92s 8.91s 26.34s
4521 using NIS & nscd 10.37s 7.34s 25.30s
4523 using NIS+ 27.57s 30.37s 640.46s
4525 using NIS+ & nscd 10.25s 7.83s 26.51s
4527 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4529 Keep in mind that non-namelookup related operations dominate above times.
4530 It was just a common complain that using NIS+ unpacking the kernel is
4533 [1] The old nscd implementation is not available anymore since it was
4534 distributed with glibc up to version 2.0.98 and thus is now replaced.
4536 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4538 * Mark Kettenis provided a Hesiod NSS module.
4540 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4541 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4543 * rcmd can now handle netgroups (Dick Streefland).
4545 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4548 * Support for the IPv6 protocol has been added to the socket API, as per the
4549 latest draft standards.
4551 * Support for Linux 2.2 has been added.
4553 * Interface changes relative to the latest 2.0.x release:
4554 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4555 addseverity NEW: Unix98
4556 alphasort64 NEW: LFS
4557 argp_err_exit_status NEW: argp, GNU ext
4558 argp_error NEW: argp, GNU ext
4559 argp_failure NEW: argp, GNU ext
4560 argp_help NEW: argp, GNU ext
4561 argp_parse NEW: argp, GNU ext
4562 argp_program_bug_address NEW: argp, GNU ext
4563 argp_program_version NEW: argp, GNU ext
4564 argp_program_version_hook NEW: argp, GNU ext
4565 argp_state_help NEW: argp, GNU ext
4566 argp_usage NEW: argp, GNU ext
4567 authdes_create NEW: Secure RPC
4568 authdes_getucred NEW: Secure RPC
4569 authdes_pk_create NEW: Secure RPC
4570 backtrace NEW: GNU ext.
4571 backtrace_symbols NEW: GNU ext.
4572 backtrace_symbols_fd NEW: GNU ext.
4574 cacosf NEW: ISO C 9x
4575 cacosh NEW: ISO C 9x
4576 cacoshf NEW: ISO C 9x
4577 cacoshl NEW: ISO C 9x
4578 cacosl NEW: ISO C 9x
4585 casinf NEW: ISO C 9x
4586 casinh NEW: ISO C 9x
4587 casinhf NEW: ISO C 9x
4588 casinhl NEW: ISO C 9x
4589 casinl NEW: ISO C 9x
4591 catanf NEW: ISO C 9x
4592 catanh NEW: ISO C 9x
4593 catanhf NEW: ISO C 9x
4594 catanhl NEW: ISO C 9x
4595 catanl NEW: ISO C 9x
4596 cbc_crypt NEW: Secure RPC
4600 ccoshf NEW: ISO C 9x
4601 ccoshl NEW: ISO C 9x
4607 cimagf NEW: ISO C 9x
4608 cimagl NEW: ISO C 9x
4609 clearerr_locked REMOVED
4610 clntunix_create NEW: sunrpc ext
4612 clog10 NEW: ISO C 9x
4613 clog10f NEW: ISO C 9x
4614 clog10l NEW: ISO C 9x
4624 cprojf NEW: ISO C 9x
4625 cprojl NEW: ISO C 9x
4627 crealf NEW: ISO C 9x
4628 creall NEW: ISO C 9x
4633 csinhf NEW: ISO C 9x
4634 csinhl NEW: ISO C 9x
4637 csqrtf NEW: ISO C 9x
4638 csqrtl NEW: ISO C 9x
4642 ctanhf NEW: ISO C 9x
4643 ctanhl NEW: ISO C 9x
4645 des_setparity NEW: Secure RPC
4646 ecb_crypt NEW: Secure RPC
4647 endutxent NEW: Unix98
4649 exp10f NEW: ISO C 9x
4650 exp10l NEW: ISO C 9x
4654 fattach NEW: STREAMS
4655 fdetach NEW: STREAMS
4659 feclearexcept NEW: ISO C 9x
4660 fegetenv NEW: ISO C 9x
4661 fegetexceptflag NEW: ISO C 9x
4662 fegetround NEW: ISO C 9x
4663 feholdexcept NEW: ISO C 9x
4665 feraiseexcept NEW: ISO C 9x
4666 ferror_locked REMOVED
4667 fesetenv NEW: ISO C 9x
4668 fesetexceptflag NEW: ISO C 9x
4669 fesetround NEW: ISO C 9x
4670 fetestexcept NEW: ISO C 9x
4671 feupdateenv NEW: ISO C 9x
4672 fflush_locked REMOVED
4676 fgets_unlocked NEW: GNU ext.
4677 fileno_locked REMOVED
4689 fputc_locked REMOVED
4690 fputs_unlocked NEW: GNU ext.
4691 fread_unlocked NEW: GNU ext.
4696 fstatvfs NEW: Unix98
4700 ftruncate64 NEW: LFS
4702 fwrite_unlocked NEW: GNU ext.
4703 gai_strerror NEW: IPv6
4707 getchar_locked REMOVED
4709 getdate_err NEW: Unix98
4710 getdate_r NEW: GNU ext.
4712 getnameinfo NEW: IPv6
4713 getnetname NEW: Secure RPC
4714 getpmsg NEW: STREAMS
4715 getpt NEW: Unix98 PTY
4716 getrlimit64 NEW: LFS
4717 getutxent NEW: Unix98
4718 getutxid NEW: Unix98
4719 getutxline NEW: Unix98
4720 glob64 NEW: GNU ext.
4721 globfree64 NEW: GNU ext.
4722 gnu_get_libc_release NEW: GNU ext.
4723 gnu_get_libc_version NEW: GNU ext.
4724 grantpt NEW: Unix98 PTY
4725 host2netname NEW: Secure RPC
4727 iconv_close NEW: iconv
4728 iconv_open NEW: iconv
4729 if_freenameindex NEW: IPv6
4730 if_indextoname NEW: IPv6
4731 if_nameindex NEW: IPv6
4732 if_nametoindex NEW: IPv6
4733 in6addr_any NEW: IPv6
4734 in6addr_loopback NEW: IPv6
4735 inet6_isipv4mapped NEW: IPv6
4736 isastream NEW: STREAMS
4737 iswblank NEW: GNU ext.
4738 key_decryptsession NEW: Secure RPC
4739 key_decryptsession_pk NEW: Secure RPC
4740 key_encryptsession NEW: Secure RPC
4741 key_encryptsession_pk NEW: Secure RPC
4742 key_gendes NEW: Secure RPC
4743 key_get_conv NEW: Secure RPC
4744 key_secretkey_is_set NEW: Secure RPC
4745 key_setnet NEW: Secure RPC
4746 key_setsecret NEW: Secure RPC
4747 llrint NEW: ISO C 9x
4748 llrintf NEW: ISO C 9x
4749 llrintl NEW: ISO C 9x
4750 llround NEW: ISO C 9x
4751 llroundf NEW: ISO C 9x
4752 llroundl NEW: ISO C 9x
4757 lrintf NEW: ISO C 9x
4758 lrintl NEW: ISO C 9x
4759 lround NEW: ISO C 9x
4760 lroundf NEW: ISO C 9x
4761 lroundl NEW: ISO C 9x
4763 makecontext NEW: Unix98
4764 mempcpy NEW: GNU ext.
4767 modify_ldt NEW: kernel
4771 nearbyint NEW: ISO C 9x
4772 nearbyintf NEW: ISO C 9x
4773 nearbyintl NEW: ISO C 9x
4774 netname2host NEW: Secure RPC
4775 netname2user NEW: Secure RPC
4776 nexttoward NEW: ISO C 9x
4777 nexttowardf NEW: ISO C 9x
4778 nexttowardl NEW: ISO C 9x
4782 passwd2des NEW: Secure RPC
4784 pow10f NEW: GNU ext.
4785 pow10l NEW: GNU ext.
4788 printf_size NEW: GNU ext.
4789 printf_size_info NEW: GNU ext.
4790 profil_counter REMOVED
4791 pthread_mutexattr_getkind_np REPLACED
4792 pthread_mutexattr_setkind_np REPLACED
4793 ptsname NEW: Unix98 PTY
4794 ptsname_r NEW: Unix98 PTY
4796 putchar_locked REMOVED
4797 putgrent NEW: GNU ext.
4799 putpmsg NEW: STREAMS
4800 pututxline NEW: Unix98
4804 readdir64_r NEW: LFS
4805 remquo NEW: ISO C 9x
4806 remquof NEW: ISO C 9x
4807 remquol NEW: ISO C 9x
4809 roundf NEW: ISO C 9x
4810 roundl NEW: ISO C 9x
4812 scalbln NEW: ISO C 9x
4813 scalblnf NEW: ISO C 9x
4814 scalblnl NEW: ISO C 9x
4816 sendfile NEW: kernel
4817 setcontext NEW: Unix98
4818 setrlimit64 NEW: LFS
4819 setutxent NEW: Unix98
4821 sigignore NEW: Unix98
4822 sigqueue NEW: POSIX.1b
4823 sigrelse NEW: Unix98
4824 sigset NEW: POSIX.1b
4825 sigtimedwait NEW: POSIX.1b
4826 sigwaitinfo NEW: POSIX.1b
4827 sincos NEW: GNU ext.
4828 sincosf NEW: GNU ext.
4829 sincosl NEW: GNU ext.
4833 strcasestr NEW: GNU ext.
4834 strtoimax NEW: ISO C 9x
4835 strtoumax NEW: ISO C 9x
4836 strverscmp NEW: GNU ext.
4837 svcauthdes_stats NEW: Secure RPC
4838 svcunix_create NEW: sunrpc etx
4839 svcunixfd_create NEW: sunrpc ext
4840 swapcontext NEW: Unix98
4841 tcgetsid NEW: Unix98 PTY
4842 tdestroy NEW: GNU ext.
4843 tgamma NEW: ISO C 9x
4844 tgammaf NEW: ISO C 9x
4845 tgammal NEW: ISO C 9x
4849 truncf NEW: ISO C 9x
4850 truncl NEW: ISO C 9x
4852 unlockpt NEW: Unix98 PTY
4853 updwtmpx NEW: Unix98
4854 user2netname NEW: Secure RPC
4855 utmpxname NEW: Unix98
4856 versionsort NEW: GNU ext.
4857 versionsort64 NEW: GNU ext.
4859 wcscasecmp NEW: GNU ext.
4860 wcsncasecmp NEW: GNU ext.
4861 wcsnlen NEW: GNU ext.
4862 wcstoimax NEW: ISO C 9x
4863 wcstoll NEW: ISO C 9x
4864 wcstoull NEW: ISO C 9x
4865 wcstoumax NEW: ISO C 9x
4867 wordexp NEW: POSIX.2
4868 wordfree NEW: POSIX.2
4869 write_profiling REMOVED
4870 xdecrypt NEW: Secure RPC
4871 xdr_authdes_cred NEW: Secure RPC
4872 xdr_authdes_verf NEW: Secure RPC
4873 xdr_cryptkeyarg NEW: Secure RPC
4874 xdr_cryptkeyarg2 NEW: Secure RPC
4875 xdr_cryptkeyres NEW: Secure RPC
4876 xdr_getcredres NEW: Secure RPC
4877 xdr_key_netstarg NEW: Secure RPC
4878 xdr_key_netstres NEW: Secure RPC
4879 xdr_keybuf NEW: Secure RPC
4880 xdr_keystatus NEW: Secure RPC
4881 xdr_netnamestr NEW: Secure RPC
4882 xdr_sizeof NEW: Secure RPC
4883 xdr_unixcred NEW: sunrpc ext
4884 xencrypt NEW: Secure RPC
4885 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4896 * inet_ntoa is thread-safe
4898 * updwtmp is moved from libutil to libc
4900 * rewrite of cbrt function
4902 * update of timezone data
4916 * add atoll function
4918 * fix complex problems in Berkeley DB code
4920 * fix math functions
4924 * fixed lots of header problems (especially Linux/GNU specific)
4926 * dynamic loader preserves all registers
4928 * Roland McGrath provided support for handling of auxiliary objects in
4929 the ELF dynamic loader.
4931 * support for parallel builds is improved
4935 * GNU extensions are no longer declared by default. To enable them you
4936 must define the macro `_GNU_SOURCE' in your program or compile with
4939 * The library has changed from using GNU ld symbol aliases to using weak
4940 symbols where available. The ELF object file format supports weak
4941 symbols; GNU ld also supports weak symbols in the a.out format. (There
4942 is also now support for other GNU ld extensions in ELF. Use the
4943 `--with-elf' option to configure to indicate you have ELF, and
4944 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4945 of many files which contained only symbol aliases, reducing the size of
4946 the source and the compiled library; many other files were renamed to
4947 less cryptic names previously occupied by the symbol alias files.
4948 There is a new header file <elf.h> for programs which operate on
4949 files in the ELF format.
4951 * Converted to Autoconf version 2, so `configure' has more options.
4952 Run `configure --help' to see the details.
4954 * The library can now be configured to build profiling, highly-optimized
4955 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4956 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4957 `configure' enable building these extra libraries. The shared library is
4958 built by default when using both ELF and GNU ld. When shared libraries
4959 are enabled, the new library `-ldl' is available for arbitrary run-time
4960 loading of shared objects; its interface is defined in <dlfcn.h>. The
4961 new header file <link.h> gives access to the internals of the run-time
4962 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4963 application of same name on other systems and it provides information
4964 about dynamically linked binaries.
4966 * The C library now provides the run-time support code for profiling
4967 executables compiled with `-pg'. Programs can control the profiling code
4968 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4969 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4970 the support for this file format was contributed by David Mosberger-Tang.
4972 * The math code has been replaced with a math library based on fdlibm from
4973 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4974 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4975 functions. The math functions now reside in a separate library, so
4976 programs using them will need to use `-lm' their linking commands.
4978 * John C. Bowman contributed optimized ix87 assembler inline functions.
4980 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4981 mechanism similar to that found in Solaris 2. This is now used for the
4982 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4983 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4984 file controls what services are used for each individual database. This
4985 works by loading shared libraries with names specified in `nsswitch.conf',
4986 so service modules can be changed or added at any time without even
4987 relinking any program. Currently there are the file, db, and NIS based
4988 NSS services available.
4990 * The new functions `strtoq' and `strtouq' parse integer values from
4991 strings, like `strtol' and `strtoul', but they return `long long int' and
4992 `unsigned long long int' values, respectively (64-bit quantities).
4994 * The new functions `strtof' and `strtold' parse floating-point values from
4995 strings, like `strtod', but they return `float' and `long double' values,
4996 respectively (on some machines `double' and `long double' are the same).
4998 * Ulrich Drepper has contributed new implementations of the floating-point
4999 printing and reading code used in the `printf' family of functions and
5000 `strtod', `strtof', and `strtold'. These new functions are perfectly
5001 accurate, and much faster than the old ones.
5003 * The implementation of the POSIX locale model was completely rewritten by
5004 Ulrich Drepper. This includes the new programs `localedef' and `locale'
5005 to compile the POSIX locale definition.
5007 * The former dummy implementations of the strcoll and strxfrm function are
5008 now replaced by fully functional code contributed by Ulrich Drepper. The
5009 collation information comes from the POSIX locale definitions.
5011 * The new header <langinfo.h> defines an interface for accessing
5012 various locale-dependent data (using the locale chosen with `setlocale').
5014 * Ulrich Drepper has contributed a new suite of functions for operation on
5015 wide-character and multibyte-character strings, in <wchar.h>;
5016 and classification and case conversion of wide characters, in <wctype.h>.
5017 These new functions are conforming to the ISO C, Amendement 1 specification.
5019 * There is now a second implementation of the standard I/O library available.
5020 It comes from GNU libg++ as was written by Per Bothner, heavily modified
5021 by Hongjiu Lu and made thread safe by Ulrich Drepper.
5023 * You can now use positional parameter specifications in format strings
5024 for the `printf' and `scanf' families of functions. For example,
5025 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
5026 ``Number 6, Mr Jones''. This is mainly useful when providing different
5027 format strings for different languages, whose grammars may dictate
5028 different orderings of the values being printed. To support this
5029 feature, the interface for `register_printf_handler' has changed; see
5030 the header file <printf.h> for details.
5032 * The `printf' and `scanf' families of functions now understand a new
5033 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
5034 to group numbers as indicated by the locale; for `scanf' and friends, this
5035 says to accept as valid only a number with all the proper grouping
5036 separators in the right places. In the default "C" locale, numbers are
5037 not grouped; but locales for specific countries will define the usual
5038 conventions (i.e. separate thousands with `,' in the US locale).
5040 * The pgrp functions have been regularized, slightly incompatibly but much
5041 less confusingly. The core functions are now `getpgid' and `setpgid',
5042 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
5043 argument) and BSD `setpgrp' (identical to `setpgid') functions are
5044 provided for compatibility. There is no longer an incompatible `getpgrp'
5045 with an argument declared under _BSD_SOURCE; no BSD code uses it.
5047 * The new header file <fts.h> and suite of functions simplify programs that
5048 operate on directory trees. This code comes from 4.4 BSD.
5050 * The resolver code has been updated from the BIND 4.9.5-P1 release.
5051 Parts of the code were heavily modified by Ulrich Drepper to fit in the
5052 NSS scheme used in glibc.
5054 * There is a new malloc debugging hook `__memalign_hook'.
5056 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
5057 `unsigned int' in <sys/types.h>. These are for compatibility only and
5058 their use is discouraged.
5060 * The `-lmcheck' library to enable standard malloc debugging hooks is now
5061 done differently, so that it works even without GNU ld.
5063 * New function `euidaccess' checks allowed access to a file like `access',
5064 but using the effective IDs instead of the real IDs.
5066 * The time zone data files have been updated for the latest and greatest
5067 local time conventions of the countries of the world.
5069 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
5072 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
5073 for formatting floating-point numbers. They are provided only for
5074 compatibility; new programs should use `sprintf' instead. There are
5075 also equivalent function for the `long double' floating-point type and
5076 all functions also exist in a reentrant form.
5078 * The new auxiliary library `-lutil' from 4.4 BSD contains various
5079 functions for maintaining the login-record files (primarily of use to
5080 system programs such as `login'), and convenient functions for
5081 allocating and initializing a pseudo-terminal (pty) device.
5083 * Ulrich Drepper has contributed new support for System V style
5084 shared memory and IPC on systems that support it.
5086 * Ulrich Drepper has contributed several miscellaneous new functions found
5087 in System V: The `hsearch' family of functions provide an effective
5088 implementation of hash tables; `a64l' and `l64a' provide a very simple
5089 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
5092 * Ulrich Drepper has contributed new reentrant counterparts for the
5093 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
5095 * Ulrich Drepper has contributed new, highly-optimized versions of several
5096 string functions for the i486/Pentium family of processors.
5098 * Ulrich Drepper has updated the Linux-specific code, based largely
5099 on work done in Hongjiu Lu's version of GNU libc for Linux.
5100 The GNU library now supports Linux versions 2.0.10 and later,
5101 using the ELF object file format (i[3456]86-*-linux).
5103 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
5105 * David Mosberger-Tang and Richard Henderson have ported the C library
5106 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
5107 dynamic linking support for ELF/Alpha.
5109 * Richard Henderson contributed several Alpha optimized assembler function
5110 for arithmetic and string handling.
5112 * Ulrich Drepper has contributed a new set of message catalog functions to
5113 support multiple languages using the <libintl.h> interface, for use with
5114 his new package GNU gettext. Translation volunteers have contributed
5115 catalogs of the library's messages in Spanish, German, and Korean.
5117 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
5118 program and the `catgets' function for reading the catalog files it
5119 creates. (The <libintl.h> interface is preferred; we include the
5120 <nl_types.h> interface using `catgets' only for source compatibility with
5121 programs already written to use it.)
5123 * New header file <values.h> gives SVID-compatible names for <limits.h>
5126 * Various new macros, declarations, and small header files for compatibility
5129 * New function `group_member' is a convenient way to check if a process has
5130 a given effective group ID.
5132 * When using GCC 2.7 and later, the socket functions are now declared in a
5133 special way so that passing an argument of type `struct sockaddr_in *',
5134 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
5135 `struct sockaddr *' type, does not generate a type-clash warning.
5137 * New function `error' declared in header file <error.h> is a convenient
5138 function for printing error messages and optionally exiting; this is the
5139 canonical function used in GNU programs. The new functions `err', `warn',
5140 and friends in header file <err.h> are the canonical 4.4 BSD interface for
5141 doing the same thing.
5143 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
5144 POSIX.2 `glob' function to do ~ and {...} expansion.
5146 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
5147 `clearenv' which is used in POSIX.9 is also available.
5149 * New function `getsid' returns session ID number on systems that support it.
5151 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
5152 files <db.h> and <mpool.h> provide a rich set of functions for several
5153 types of simple databases stored in memory and in files, and <ndbm.h> is
5154 an old `ndbm'-compatible interface using the `db' functions. Link with
5155 `-ldb' to get these functions.
5157 * New macro `strdupa' copies a string like `strdup', but uses local stack
5158 space from `alloca' instead of dynamic heap space from `malloc'.
5160 * New function `strnlen' is like `strlen' but searches only a given maximum
5161 number of characters for the null terminator. `stpncpy', `strndup' and
5162 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
5165 * New function `statfs' in header <sys/statfs.h>.
5167 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
5168 provide convenient functions for operating on blocks of null-terminated
5171 * A new suite of functions in <utmp.h> handle all the details of reading
5172 and writing the utmp file.
5174 * An implementation of the NIS/YP(tm) based NSS service was contributed by
5177 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
5178 completely POSIX compliant and also implemented the extended functionality
5179 to handle alternate digit representation and alternate era date formats.
5181 * Ulrich Drepper provided an implementation of the `strptime' function
5182 defined in XPG4.2 which transforms a string into a `struct tm' value.
5184 * Paul Eggert provided the tzselect shell script as part of the timezone
5185 code. The shell script makes it easy to select the correct timezone
5188 * The implementation of the malloc family of functions is completely replaced
5189 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
5190 The implementation uses the mmap function (if available) and it is
5191 optimized for the use in multi threaded programs.
5193 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
5194 functions. This new functionality is usable by specifying a special
5195 salt string and it is compatible with implementation on *BSD systems.
5197 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
5198 `getsubopt' to handle second level command line options, `bsd_signal'
5199 to access BSD style `signal' functionality, the obsolete `regexp' style
5202 * the `lchown' function is available on system which support this
5205 * The implementation of the shadow password handling function was contributed
5208 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5210 * POSIX.1g support was added. The <sys/select.h> header is available,
5211 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5212 implementation of `getaddrinfo'.
5216 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5218 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5219 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5221 * The new function `daemon' from 4.4 BSD is useful for server programs that
5222 want to put themselves in the background.
5224 * Joel Sherrill has contributed support for several standalone boards that
5225 run without an operating system.
5227 * `printf', `scanf' and friends now accept a `q' type modifier for long
5228 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5230 * All of the code taken from BSD (notably most of the math and networking
5231 routines) has been updated from the BSD 4.4-Lite release.
5233 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5235 * The new functions `getdomainname' and `setdomainname' fetch or change the
5236 YP/NIS domain name. These are system calls which exist on systems which
5239 * The time zone data files have been updated for the latest international
5242 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5243 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5247 * The C library now includes support for Sun RPC, from Sun's free
5248 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5249 are included. (There is still no support for YP.)
5251 * Tom Quinn has contributed a port of the C library to SGI machines running
5252 Irix 4 (mips-sgi-irix4).
5254 * The new `lockf' function is a simplified interface to the locking
5255 facilities of `fcntl', included for compatibility.
5257 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5259 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5261 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5264 * The error message printed by `assert' for a failed assertion now includes
5265 the name of the program (if using GNU ld) and the name of the calling
5266 function (with versions of GCC that support this).
5268 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5270 * The library now includes the <sys/mman.h> header file and memory
5271 management functions `mmap', `munmap', `mprotect', `msync', and
5272 `madvise', on systems that support those facilities.
5274 * The interface for `mcheck' has changed slightly: the function called to
5275 abort the program when an allocation inconsistency is detected now takes
5276 an argument that indicates the type of failure. The new function
5277 `mprobe' lets you request a consistency check for a particular block at
5278 any time (checks are normally done only when you call `free' or `realloc'
5281 * It is now possible to easily cross-compile the C library, building on one
5282 system a library to run on another machine and/or operating system. All
5283 you need to do is set the variable `HOST_CC' in `configparms' to the
5284 native compiler for programs to run on the machine you are building on (a
5285 few generator programs are used on Unix systems); set `CC' to the
5288 * The new function `fexecve' (only implemented on the GNU system) executes
5289 a program file given a file descriptor already open on the file.
5293 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5294 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5296 * You can set the variable `libprefix' in `configparms' to specify a prefix
5297 to be prepended to installed library files; this makes it easy to install
5298 the GNU C library to be linked as `-lgnuc' or whatever.
5300 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5301 copies a limited number of characters from a string, and returns the
5302 address of the last character written.
5304 * You no longer need to check for whether the installed `stddef.h' is
5305 compatible with the GNU C library. configure now checks for you.
5307 * You can now define a per-stream `fileno' function to convert the
5308 stream's cookie into an integral file descriptor.
5310 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5311 allocates zero bytes of storage, and returns a unique pointer which
5312 you can pass to `realloc' or `free'. The behavior is undefined if
5313 you dereference this pointer.
5315 * The C library now runs on Sony NEWS m68k machines running either
5316 NewsOS 3 or NewsOS 4.
5318 * The new `syscall' function is a system-dependent primitive function
5319 for invoking system calls. It has the canonical behavior on Unix
5320 systems, including unreliable return values for some calls (such as
5321 `pipe', `fork' and `getppid').
5323 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5324 to `EAGAIN', which is the preferred name. On systems whose kernels
5325 use two distinct codes, the C library now translates EWOULDBLOCK to
5326 EAGAIN in every system call function.
5330 * The GNU C Library Reference Manual is now distributed with the library.
5331 `make dvi' will produce a DVI file of the printed manual.
5332 `make info' will produce Info files that you can read on line using C-h i
5333 in Emacs or the `info' program.
5334 Please send comments on the manual to bug-glibc-manual@gnu.org.
5336 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5338 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5340 * Jason Merrill has contributed a port to the Sequent Symmetry running
5341 Dynix version 3 (i386-sequent-dynix).
5343 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5344 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5346 * New function `memory_warnings' lets you arrange to get warnings when
5347 malloc is running out of memory to allocate, like Emacs gives you.
5349 * The C library now contains the relocating allocator used in Emacs 19 for
5350 its editing buffers. This allocator (ralloc) minimizes allocation
5351 overhead and fragmentation by moving allocated regions around whenever it
5352 needs to. You always refer to a ralloc'd region with a "handle" (a
5353 pointer to a pointer--an object of type `void **').
5355 * There is a new `printf' format: `%m' gives you the string corresponding
5356 to the error code in `errno'.
5358 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5359 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5360 pass, the `a' modifier says to fill in a `char **' you pass with a
5363 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5364 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5365 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5367 * `mkstemp' is a traditional Unix function to atomically create and open a
5368 uniquely-named temporary file.
5372 * The standard location for the file that says what the local timezone is
5373 has changed again. It is now `/usr/local/etc/localtime' (or more
5374 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5376 * The distribution no longer contains any files with names longer than 14
5379 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5380 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5382 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5386 * The configuration process has changed quite a bit. The `configure'
5387 script is now used just like the configuration scripts for other GNU
5388 packages. The `sysdeps' directory hierarchy is much rearranged.
5389 The file `INSTALL' explains the new scheme in detail.
5391 * The header files no longer need to be processed into ANSI C and
5392 traditional C versions. There is just one set of files to install, and
5393 it will work with ANSI or old C compilers (including `gcc -traditional').
5395 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5396 MIPS DECStation running Ultrix 4.
5398 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5399 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5400 made itself into a shared library.
5402 * Yet further improved support for the i386, running 4.3 BSD-like systems
5403 (such as Mach 3 with the Unix single-server), or System V.
5405 * New function `strncasecmp' to do case-insensitive string comparison
5406 with limited length.
5408 * New function `strsep' is a reentrant alternative to `strtok'.
5410 * New functions `scandir' and `alphasort' for searching directories.
5412 * New function `setenv' is a better interface to `putenv'.
5414 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5415 function for traversing a directory tree.
5417 * The GNU obstack package is now also part of the C library.
5418 The new function `open_obstack_stream' creates a stdio stream that
5419 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5420 formatted output directly to an obstack.
5422 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5423 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5425 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5427 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5428 things to your strings.
5430 * There are some new test programs: test-fseek, testmb, and testrand.
5432 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5433 These ports are not finished, but are a good starting place for really
5434 supporting those systems.
5436 * `/etc/localtime' is now the standard location for the file that says what
5437 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5438 This follows the general principle that `/etc' is the place for all local
5439 configuration files.
5441 * The C library header files now use `extern "C"' when used by the C++
5442 compiler, so the C library should now work with C++ code.
5444 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5445 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5448 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5449 function which uses the `merge sort' algorithm, and is said to be
5450 significantly faster than the old GNU `qsort' function. Merge sort is
5451 now the standard `qsort' function. The new algorithm can require a lot
5452 of temporary storage; so, the old sorting function is called when the
5453 required storage is not available.
5455 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5456 provides the Unix `crypt' function, plus some other entry points.
5458 * The code and header files taken from 4.4 BSD have been updated with the
5459 latest files released from Berkeley.
5461 ----------------------------------------------------------------------
5462 Copyright information:
5464 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5466 Permission is granted to anyone to make or distribute verbatim copies
5467 of this document as received, in any medium, provided that the
5468 copyright notice and this permission notice are preserved,
5469 thus giving the recipient permission to redistribute in turn.
5471 Permission is granted to distribute modified versions
5472 of this document, or of portions of it,
5473 under the above conditions, provided also that they
5474 carry prominent notices stating who last changed them.
5477 version-control: never