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 [Add new features here]
14 Deprecated and removed features, and other changes affecting compatibility:
16 [Add deprecations, removals and changes affecting compatibility here]
18 Changes to build and runtime requirements:
20 * GCC 6.2 or later is required to build the GNU C Library.
22 Older GCC versions and non-GNU compilers are still supported when
23 compiling programs that use the GNU C Library.
25 Security related changes:
27 [Add security related changes here]
29 The following bugs are resolved with this release:
31 [The release manager will add the list generated by
32 scripts/list-fixed-bugs.py just before the release.]
39 * The getcpu wrapper function has been added, which returns the currently
40 used CPU and NUMA node. This function is Linux-specific.
42 * A new convenience target has been added for distribution maintainers
43 to build and install all locales as directories with files. The new
44 target is run by issuing the following command in your build tree:
45 'make localedata/install-locale-files', with an optional DESTDIR
46 to set the install root if you wish to install into a non-default
49 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
51 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
52 for _GNU_SOURCE, to match BSD environments.
54 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
55 indicates that it will abort the transaction prior to entering the kernel
56 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
57 suspended, and this caused some undefined side-effects issues by aborting
58 transactions manually. Glibc avoided it by abort transactions manually on
59 each syscall, but it lead to performance issues on newer kernels where the
60 HTM state is saved and restore lazily (the state being saved even when the
61 process actually does not use HTM).
63 * The functions posix_spawn_file_actions_addchdir_np and
64 posix_spawn_file_actions_addfchdir_np have been added, enabling
65 posix_spawn and posix_spawnp to run the new process in a different
66 directory. These functions are GNU extensions. The function
67 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
70 * The popen and system do not run atfork handlers anymore (BZ#17490).
71 Although it is a possible POSIX violation, the POSIX rationale in
72 pthread_atfork documentation regarding atfork handlers is to handle
73 inconsistent mutex state after a fork call in a multi-threaded process.
74 In both popen and system there is no direct access to user-defined mutexes.
76 * Support for the C-SKY ABIV2 running on Linux has been added. This port
77 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
79 - C-SKY ABIV2 soft-float little-endian
80 - C-SKY ABIV2 hard-float little-endian
82 * strftime's default formatting of a locale's alternative year (%Ey)
83 has been changed to zero-pad the year to a minimum of two digits,
84 like "%y". This improves the display of Japanese era years during
85 the first nine years of a new era, and is expected to be harmless
86 for all other locales (only Japanese locales regularly have
87 alternative year numbers less than 10). Zero-padding can be
88 overridden with the '_' or '-' flags (which are GNU extensions).
90 * As a GNU extension, the '_' and '-' flags can now be applied to
91 "%EY" to control how the year number is formatted; they have the
92 same effect that they would on "%Ey".
94 Deprecated and removed features, and other changes affecting compatibility:
96 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
97 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
99 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
100 in <sys/procfs.h>, has been corrected to match the type actually used by
101 the Linux kernel. This affects the size and layout of that structure on
102 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
104 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
105 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
106 defined in <sys/procfs.h>, has been corrected to match the type actually
107 used by the Linux kernel. This affects the size and layout of those
110 * An archaic GNU extension to scanf, under which '%as', '%aS', and '%a[...]'
111 meant to scan a string and allocate space for it with malloc, is now
112 restricted to programs compiled in C89 or C++98 mode with _GNU_SOURCE
113 defined. This extension conflicts with C99's use of '%a' to scan a
114 hexadecimal floating-point number, which is now available to programs
115 compiled as C99 or C++11 or higher, regardless of _GNU_SOURCE.
117 POSIX.1-2008 includes the feature of allocating a buffer for string input
118 with malloc, using the modifier letter 'm' instead. Programs using
119 '%as', '%aS', or '%a[...]' with the old GNU meaning should change to
120 '%ms', '%mS', or '%m[...]' respectively. Programs that wish to use the
121 C99 '%a' no longer need to avoid _GNU_SOURCE.
123 GCC's -Wformat warnings can detect most uses of this extension, as long
124 as all functions that call vscanf, vfscanf, or vsscanf are annotated with
125 __attribute__ ((format (scanf, ...))).
127 Changes to build and runtime requirements:
129 * Python 3.4 or later is required to build the GNU C Library.
131 * On most architectures, GCC 5 or later is required to build the GNU C
132 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
134 Older GCC versions and non-GNU compilers are still supported when
135 compiling programs that use the GNU C Library.
137 Security related changes:
139 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
140 denial of service due to resource exhaustion when processing getaddrinfo
141 calls with crafted host names. Reported by Guido Vranken.
143 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
144 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
145 happened, accessing the 32-bit size_t value as the full 64-bit register
146 in the assembly string/memory functions would cause a buffer overflow.
149 CVE-2016-10739: The getaddrinfo function could successfully parse IPv4
150 addresses with arbitrary trailing characters, potentially leading to data
151 or command injection issues in applications.
153 The following bugs are resolved with this release:
155 [10425] localedata: it_IT/it_CH: LC_TIME format is wrong
156 [10496] localedata: 12h time representation in multiple locales faulty
157 [10797] localedata: it_IT locale numeric does not have a separator for
159 [11319] libc: dprintf doesn't handle errors properly
160 [16346] time: mktime: potentially unsafe use of localtime_offset
161 [17248] build: glibc should not sort CFLAGS (support gcc plugins and
163 [17405] libc: Implement posix_spawn_file_actions_addchdir_np,
164 posix_spawn_file_actions_addfchdir_np
165 [17426] localedata: Indian locales: set the correct date format
166 [17490] stdio: popen should not invoke atfork handlers
167 [17783] libc: TIOCSER_TEMT conditions inconsistent
168 [18040] regex: use-after-free in regexec/get_subexp
169 [18093] libc: Corrupted aux-cache causes ldconfig to segfault
170 [20018] network: getaddrinfo should reject IP addresses with trailing
171 characters (CVE-2016-10739)
172 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
173 [20271] libc: Missing "\n" in __libc_fatal calls
174 [20480] dynamic-link: Patch: ifunc not executable, crashes sudo qemu
175 [20544] libc: RFE: atexit, __cxa_atexit, on_exit should assert function
176 pointer argument is non-NULL
177 [21037] stdio: open_memstream and freopen
178 [21286] libc: bits/siginfo.h is missing enum definition for TRAP_HWBKPT
179 [21716] time: Crash in glibc's mktime in low-memory situations
180 [22834] stdio: Subprocess forked by popen may crash in Linux when
181 multithreads call popen
182 [22927] network: crash in vn_gai_enqueue_request if requests_tail was NULL
183 and pthread_create fails.
184 [23032] hurd: sysdeps/htl/pt-barrier-init.c:39: bad call to memcmp ?
185 [23125] libc: riscv64: endless loop when throwing an exception from a
187 [23275] nptl: Race in pthread_mutex_lock while promoting to
188 PTHREAD_MUTEX_ELISION_NP.
189 [23400] libc: stdlib/test-bz22786.c creates temporary files in glibc
191 [23479] math: [mips] bits/fenv.h should not define some macros for soft-
193 [23490] libc: sysdeps/unix/sysv/linux/x86/tst-cet-property-2.c:49: off by
195 [23497] libc: readdir64@GLIBC_2.1 cannot parse the kernel directory stream
196 [23509] dynamic-link: CET enabled glibc is incompatible with the older
198 [23520] nscd: nscd: Use-after-free in addgetnetgrentX and its callers
199 [23521] nss: get_next_alias nss_files file stream leak
200 [23538] nptl: Hang in pthread_cond_broadcast
201 [23562] libc: Wrong type for si_band in Linux-specific siginfo_t
202 [23578] regex: Invalid memory access if regex pattern contains NUL byte
203 [23579] libc: Errors misreported in preadv2
204 [23597] build: support/test-container.c doesn't work with different
206 [23603] time: mktime signed integer overflow on large timestamps
207 [23606] libc: Missing ENDBR32 in sysdeps/i386/start.S
208 [23614] libc: powerpc: missing CFI register information in __mpn_*
210 [23637] string: Generic strstr/strcasestr fails with huge needles
211 [23640] libc: no way to easily clear FD_CLOEXEC in
212 posix_spawn_file_actions_adddup2()
213 [23649] libc: [microblaze/mips/nios2/riscv] sys/procfs.h pr_uid, pr_gid
215 [23656] libc: [mips n32] sys/procfs.h pr_sigpend, pr_sighold, pr_flag have
217 [23679] libc: gethostid: Missing NULL check for gethostbyname_r result
218 [23689] libc: Bug in documentation for rusage.ru_ixrss in
219 bits/types/struct_rusage.h
220 [23690] dynamic-link: Segfault in _dl_profile_fixup with a high number of
222 [23707] dynamic-link: Missing unwind info in sysdeps/powerpc/powerpc32/dl-
224 [23709] string: glibc 2.25 lacks sse2 optimized strstr()
225 [23716] dynamic-link: _dl_runtime_resolve_shstk isn't selected properly
226 [23717] libc: glibc: stdlib/tst-setcontext9 test suite failure on
228 [23724] localedata: Albanian date formats are incorrect
229 [23735] math: libnldbl_nonshared.a references internal libm symbols
230 [23740] localedata: kl_GL: Month names and date formats need update
231 [23744] regex: regex refactorings to remove BE, avoid duplication
232 [23745] time: mktime fix for Gnulib + coreutils
233 [23758] time: Improve the width of alternate representation for year in
235 [23783] libc: [mips] Missing CMSPAR bits/termios.h
236 [23789] time: mktime does not set errno on failure
237 [23791] localedata: Wrong monetary format for ca_ES locale
238 [23793] locale: c32rtomb and mbrtoc32 should not alias wcrtomb and mbrtowc
239 [23794] locale: c16rtomb does not handle surrogate pairs
240 [23821] libc: si_band in siginfo_t has wrong type long int on sparc64
241 [23822] math: ia64 static libm.a is missing exp2f, log2f and powf symbols
242 [23836] time: time/tst-mktime2 test failure on Arm (32-bit)
243 [23848] libc: [sparc] Some socket syscalls wrongly assumed to be present
244 [23861] nptl: rdlock stalls indefinitely on an unlocked pthread rwlock
245 [23862] libc: [sh] missing kernel-features.h undefines
246 [23864] libc: [riscv] missing kernel-features.h undefines
247 [23867] libc: [arm/microblaze] __ASSUME_MLOCK2 incorrect
248 [23907] malloc: Incorrect double-free malloc tcache check disregards
250 [23913] libc: off-by-one in function maybe_script_execute in
251 sysdeps/posix/spawni.c
252 [23915] libc: [arm] __ASSUME_COPY_FILE_RANGE incorrect
253 [23923] locale: Add --no-hard-links option to localedef
254 [23927] network: Linux if_nametoindex() does not close descriptor
256 [23961] math: powf can overflow to inf without setting errno in non-
257 nearest rounding mode
258 [23967] libc: [2.28 Regression]: New sigaction implementation breaks m68k
259 [23972] libc: __old_getdents64 uses wrong d_off value on overflow
260 [23993] libc: glibc 2.29 doesn't build with gcc 4.9
261 [23995] localedata: Remove execution flags from localedata/locales/bi_VU
262 [24011] localedata: Fixed small type in comment for locale bs_BA
263 [24018] libc: gettext() may return NULL
264 [24022] build: riscv build failure with Linux kernel 4.20-rc7
265 [24023] build: [2.29 Regression] FAIL: elf/check-localplt
266 [24024] string: strerror() might set errno to ENOMEM due to -fno-math-
268 [24027] malloc: glibc: realloc() ncopies 32-bit integer overflow
269 [24034] libc: tst-cancel21-static fails with SIGBUS on pre-ARMv7 when
271 [24046] localedata: en_US locale doesn't define date_fmt
272 [24063] manual: @var{errno} should be @code{errno}
273 [24066] soft-fp: Inconsistent _FP_W_TYPE_SIZE check
274 [24088] libc: VSCR field is not being correctly read in ucontext_t on
276 [24096] time: Specifying '_' or '-' flag for "%EY" does not produce the
278 [24097] string: Can't use 64-bit register for size_t in assembly codes for
280 [24110] hurd: SS_DISABLE never set in stack_t value returned by
282 [24112] network: Do not send DNS queries for non-host names (where all
283 answers will be rejected)
284 [24130] libc: alpha __remqu corrupts $f3 register
291 * The localization data for ISO 14651 is updated to match the 2016
292 Edition 4 release of the standard, this matches data provided by
293 Unicode 9.0.0. This update introduces significant improvements to the
294 collation of Unicode characters. This release deviates slightly from
295 the standard in that the collation element ordering for lowercase and
296 uppercase LATIN script characters is adjusted to ensure that regular
297 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
298 is not matched by [a-z]. With the update many locales have been
299 updated to take advantage of the new collation information. The new
300 collation information has increased the size of the compiled locale
301 archive or binary locales.
303 * The GNU C Library can now be compiled with support for Intel CET, AKA
304 Intel Control-flow Enforcement Technology. When the library is built
305 with --enable-cet, the resulting glibc is protected with indirect
306 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
307 compatible with all existing executables and shared libraries. This
308 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
309 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
310 capable of multi-byte NOPs, like x86-64 processors as well as Intel
311 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
312 x86_64 and x32 on non-CET processors. --enable-cet has been tested
313 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
316 * The GNU C Library now has correct support for ABSOLUTE symbols
317 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
318 relocated incorrectly or in some cases discarded. The GNU linker can
319 make use of the newer semantics, but it must communicate it to the
320 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
321 field) to indicate such support is required.
323 * Unicode 11.0.0 Support: Character encoding, character type info, and
324 transliteration tables are all updated to Unicode 11.0.0, using
325 generator scripts contributed by Mike FABIAN (Red Hat).
327 * <math.h> functions that round their results to a narrower type are added
328 from TS 18661-1:2014 and TS 18661-3:2015:
330 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
333 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
336 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
339 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
342 * Two grammatical forms of month names are now supported for the following
343 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
344 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
345 support two grammatical forms in abbreviated month names: Catalan, Greek,
348 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
349 include the support for two grammatical forms of month names.
351 * Building and running on GNU/Hurd systems now works without out-of-tree
354 * The renameat2 function has been added, a variant of the renameat function
355 which has a flags argument. If the flags are zero, the renameat2 function
356 acts like renameat. If the flag is not zero and there is no kernel
357 support for renameat2, the function will fail with an errno value of
358 EINVAL. This is different from the existing gnulib function renameatu,
359 which performs a plain rename operation in case of a RENAME_NOREPLACE
360 flags and a non-existing destination (and therefore has a race condition
361 that can clobber the destination inadvertently).
363 * The statx function has been added, a variant of the fstatat64
364 function with an additional flags argument. If there is no direct
365 kernel support for statx, glibc provides basic stat support based on
366 the fstatat64 function.
368 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
369 library if installed. libidn2 version 2.0.5 or later is recommended. If
370 libidn2 is not available, internationalized domain names are not encoded
371 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
372 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
373 with an encoding error.) Flags which used to change the IDN encoding and
374 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
375 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
376 deprecated. They no longer have any effect.
378 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
379 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
380 range of ELF gABI expressions including such constructs as
381 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
382 have been further restricted, and where in the past a dynamic string
383 token sequence may have been interpreted as a literal string it will
384 now cause a load failure. These load failures were always considered
385 unspecified behaviour from the perspective of the dynamic loader, and
386 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
387 results in a load failure now.
389 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
390 implementation includes all the standard functions provided by
393 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
394 thrd_detach, thrd_exit, and thrd_join for thread management.
396 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
397 mtx_destroy for mutual exclusion.
399 - call_once for function call synchronization.
401 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
402 cnd_wait for conditional variables.
404 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
406 Application developers must link against libpthread to use ISO C threads.
408 Deprecated and removed features, and other changes affecting compatibility:
410 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
411 installed. Software that was using either header should be updated to
412 use standard <stdio.h> interfaces instead.
414 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
415 This was never required by the C standard, and the macros just expanded
416 to call alternative names for the same functions. If you hoped getc and
417 putc would provide performance improvements over fgetc and fputc, instead
418 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
419 necessary, flockfile and funlockfile.
421 * All stdio functions now treat end-of-file as a sticky condition. If you
422 read from a file until EOF, and then the file is enlarged by another
423 process, you must call clearerr or another function with the same effect
424 (e.g. fseek, rewind) before you can read the additional data. This
425 corrects a longstanding C99 conformance bug. It is most likely to affect
426 programs that use stdio to read interactive input from a terminal.
429 * The macros 'major', 'minor', and 'makedev' are now only available from
430 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
431 headers that happen to include <sys/types.h>. These macros are rarely
432 used, not part of POSIX nor XSI, and their names frequently collide with
433 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
436 <sys/sysmacros.h> is a GNU extension. Portable programs that require
437 these macros should first include <sys/types.h>, and then include
438 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
440 * The tilegx*-*-linux-gnu configurations are no longer supported.
442 * The obsolete function ustat is no longer available to newly linked
443 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
444 function has been deprecated in favor of fstatfs and statfs.
446 * The obsolete function nfsservctl is no longer available to newly linked
447 binaries. This function was specific to systems using the Linux kernel
448 and could not usefully be used with the GNU C Library on systems with
449 version 3.1 or later of the Linux kernel.
451 * The obsolete function name llseek is no longer available to newly linked
452 binaries. This function was specific to systems using the Linux kernel
453 and was not declared in a header. Programs should use the lseek64 name
454 for this function instead.
456 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
457 getaddrinfo and getnameinfo functions have been deprecated. The behavior
458 previously selected by them is now always enabled.
460 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
461 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
462 restriction (rejecting '_' in host names, among other things) has been
463 removed, for increased compatibility with non-IDN name resolution.
465 * The fcntl function now have a Long File Support variant named fcntl64. It
466 is added to fix some Linux Open File Description (OFD) locks usage on non
467 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
468 fcntl and LFS support is handled transparently. Also for Linux, the OFD
469 locks act as a cancellation entrypoint.
471 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
472 ecb_crypt, and des_setparity are no longer available to newly linked
473 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
474 longer installed. These functions encrypted and decrypted data with the
475 DES block cipher, which is no longer considered secure. Software that
476 still uses these functions should switch to a modern cryptography library,
479 * Reflecting the removal of the encrypt and setkey functions above, the
480 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
481 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
484 * The obsolete function fcrypt is no longer available to newly linked
485 binaries. It was just another name for the standard function crypt,
486 and it has not appeared in any header file in many years.
488 * We have tentative plans to hand off maintenance of the passphrase-hashing
489 library, libcrypt, to a separate development project that will, we hope,
490 keep up better with new passphrase-hashing algorithms. We will continue
491 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
492 'crypt_r' should not need to change at all; however, distributions will
493 need to install <crypt.h> and libcrypt from a separate project.
495 In this release, if the configure option --disable-crypt is used, glibc
496 will not install <crypt.h> or libcrypt, making room for the separate
497 project's versions of these files. The plan is to make this the default
498 behavior in a future release.
500 Changes to build and runtime requirements:
502 GNU make 4.0 or later is now required to build glibc.
504 Security related changes:
506 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
507 been fixed by removing the glibc-internal IDNA implementation and using
508 the system-provided libidn2 library instead. Originally reported by Hanno
509 Böck and Christian Weisgerber.
511 CVE-2017-18269: An SSE2-based memmove implementation for the i386
512 architecture could corrupt memory. Reported by Max Horn.
514 CVE-2018-11236: Very long pathname arguments to realpath function could
515 result in an integer overflow and buffer overflow. Reported by Alexey
518 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
519 architecture could write beyond the target buffer, resulting in a buffer
520 overflow. Reported by Andreas Schwab.
522 The following bugs are resolved with this release:
524 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
525 [6889] manual: 'PWD' mentioned but not specified
526 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
527 when __WORDSIZE != 64
528 [13762] regex: re_search etc. should return -2 on memory exhaustion
529 [13888] build: /tmp usage during testing
530 [13932] math: dbl-64 pow unexpectedly slow for some inputs
531 [14092] nptl: Support C11 threads
532 [14095] localedata: Review / update collation data from Unicode / ISO
534 [14508] libc: -Wformat warnings
535 [14553] libc: Namespace pollution loff_t in sys/types.h
536 [14890] libc: Make NT_PRFPREG canonical.
537 [15105] libc: Extra PLT references with -Os
538 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
540 [16335] manual: Feature test macro documentation incomplete and out of
542 [16552] libc: Unify umount implementations in terms of umount2
543 [17082] libc: htons et al.: statement-expressions prevent use on global
544 scope with -O1 and higher
545 [17343] libc: Signed integer overflow in /stdlib/random_r.c
546 [17438] localedata: pt_BR: wrong d_fmt delimiter
547 [17662] libc: please implement binding for the new renameat2 syscall
548 [17721] libc: __restrict defined as /* Ignore */ even in c11
549 [17979] libc: inconsistency between uchar.h and stdint.h
550 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
551 [18023] libc: extend_alloca is broken (questionable pointer comparison,
552 horrible machine code)
553 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
555 [18471] libc: llseek should be a compat symbol
556 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
558 [18991] nss: nss_files skips large entry in database
559 [19239] libc: Including stdlib.h ends up with macros major and minor being
561 [19463] libc: linknamespace failures when compiled with -Os
562 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
563 [19527] locale: Normalized charset name not recognized by setlocale
564 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
565 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
567 [19728] network: out of bounds stack read in libidn function
568 idna_to_ascii_4i (CVE-2016-6261)
569 [19729] network: out of bounds heap read on invalid utf-8 inputs in
570 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
571 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
573 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
574 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
575 [20419] dynamic-link: files with large allocated notes crash in
577 [20530] libc: bswap_16 should use __builtin_bswap16() when available
578 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
579 [20980] manual: CFLAGS environment variable replaces vital options
580 [21163] regex: Assertion failure in pop_fail_stack when executing a
581 malformed regexp (CVE-2015-8985)
582 [21234] manual: use of CFLAGS makes glibc detect no optimization
583 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
584 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
585 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
586 [21508] locale: intl/tst-gettext failure with latest msgfmt
587 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
588 [21812] network: getifaddrs() returns entries with ifa_name == NULL
589 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
591 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
593 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
594 [22247] network: Integer overflow in the decode_digit function in
595 puny_decode.c in libidn (CVE-2017-14062)
596 [22342] nscd: NSCD not properly caching netgroup
597 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
598 [22550] localedata: es_ES locale (and other es_* locales): collation
599 should treat ñ as a primary different character, sync the collation
600 for Spanish with CLDR
601 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
602 by gcc configured with --enable-default-pie
603 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
604 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
605 crossing 2GB threshold (CVE-2017-18269)
606 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
608 [22735] time: Misleading typo in time.h source comment regarding
610 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
611 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
612 `__assert_perror_fail ()' to `__assert_fail_base ()'
613 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
615 [22786] libc: Stack buffer overflow in realpath() if input size is close
616 to SSIZE_MAX (CVE-2018-11236)
617 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
618 through an absolute DT_NEEDED path
619 [22792] build: tcb-offsets.h dependency dropped
620 [22797] libc: pkey_get() uses non-reserved name of argument
621 [22807] libc: PTRACE_* constants missing for powerpc
622 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
623 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
624 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
625 [22848] localedata: ca_ES: update date definitions from CLDR
626 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
627 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
628 [22896] localedata: Update locale data for an_ES
629 [22902] math: float128 test failures with GCC 8
630 [22918] libc: multiple common of `__nss_shadow_database'
631 [22919] libc: sparc32: backtrace yields infinite backtrace with
633 [22926] libc: FTBFS on powerpcspe
634 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
636 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
637 [22947] libc: FAIL: misc/tst-preadvwritev2
638 [22963] localedata: cs_CZ: Add alternative month names
639 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
640 [22996] localedata: change LC_PAPER to en_US in es_BO locale
641 [22998] dynamic-link: execstack tests are disabled when SELinux is
643 [23005] network: Crash in __res_context_send after memory allocation
645 [23007] math: strtod cannot handle -nan
646 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
647 [23036] regex: regex equivalence class regression
648 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
649 [23069] libc: sigaction broken on riscv64-linux-gnu
650 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
651 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
652 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
653 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
654 and libc build with -Os)
655 [23140] localedata: More languages need two forms of month names
656 [23145] libc: _init/_fini aren't marked as hidden
657 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
658 [23171] math: C++ iseqsig for long double converts arguments to double
659 [23178] nscd: sudo will fail when it is run in concurrent with commands
660 that changes /etc/passwd
661 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
663 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
664 [23208] localedata: New locale - Lower Sorbian (dsb)
665 [23233] regex: Memory leak in build_charclass_op function in file
667 [23236] stdio: Harden function pointers in _IO_str_fields
668 [23250] nptl: Offset of __private_ss differs from GCC
669 [23253] math: tgamma test suite failures on i686 with -march=x86-64
670 -mtune=generic -mfpmath=sse
671 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
673 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
674 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
676 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
677 [23277] math: nan function should not have const attribute
678 [23279] math: scanf and strtod wrong for some hex floating-point
679 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
681 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
682 [23303] build: undefined reference to symbol
683 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
684 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
686 [23313] stdio: libio vtables validation and standard file object
688 [23329] libc: The __libc_freeres infrastructure is not properly run across
690 [23349] libc: Various glibc headers no longer compatible with
692 [23351] malloc: Remove unused code related to heap dumps and malloc
694 [23363] stdio: stdio-common/tst-printf.c has non-free license
695 [23396] regex: Regex equivalence regression in single-byte locales
696 [23422] localedata: oc_FR: More updates of locale data
697 [23442] build: New warning with GCC 8
698 [23448] libc: Out of bounds access in IBM-1390 converter
699 [23456] libc: Wrong index_cpu_LZCNT
700 [23458] build: tst-get-cpu-features-static isn't added to tests
701 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
703 [23467] dynamic-link: x86/CET: A property note parser bug
710 * The GNU C Library can now be compiled with support for building static
711 PIE executables (See --enable-static-pie in INSTALL). These static PIE
712 executables are like static executables but can be loaded at any address
713 and provide additional security hardening benefits at the cost of some
714 memory and performance. When the library is built with --enable-static-pie
715 the resulting libc.a is usable with GCC 8 and above to create static PIE
716 executables using the GCC option '-static-pie'. This feature is currently
717 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
718 aarch64 with binutils 2.30 or later.
720 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
721 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
724 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
726 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
728 * In order to support faster and safer process termination the malloc API
729 family of functions will no longer print a failure address and stack
730 backtrace after detecting heap corruption. The goal is to minimize the
731 amount of work done after corruption is detected and to avoid potential
732 security issues in continued process execution. Reducing shutdown time
733 leads to lower overall process restart latency, so there is benefit both
734 from a security and performance perspective.
736 * The abort function terminates the process immediately, without flushing
737 stdio streams. Previous glibc versions used to flush streams, resulting
738 in deadlocks and further data corruption. This change also affects
739 process aborts as the result of assertion failures.
741 * On platforms where long double has the IEEE binary128 format (aarch64,
742 alpha, mips64, riscv, s390 and sparc), the math library now implements
743 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
744 These are the same interfaces added in version 2.26 for some platforms where
745 this format is supported but is not the format of long double.
747 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
748 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
749 implements interfaces for that type, as defined by ISO/IEC TS
750 18661-3:2015. These are corresponding interfaces to those supported for
753 * The math library now implements interfaces for the _Float32, _Float64 and
754 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
755 corresponding interfaces to those supported for _Float128.
757 * glibc now implements the memfd_create and mlock2 functions on Linux.
759 * Support for memory protection keys was added. The <sys/mman.h> header now
760 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
763 * The copy_file_range function was added.
765 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
767 * The ldconfig utility now processes `include' directives using the C/POSIX
768 collation ordering. Previous glibc versions used locale-specific
769 ordering, the change might break systems that relied on that.
771 * Support for two grammatical forms of month names has been added.
772 In a call to strftime, the "%B" and "%b" format specifiers will now
773 produce the grammatical form required when the month is used as part
774 of a complete date. New "%OB" and "%Ob" specifiers produce the form
775 required when the month is named by itself. For instance, in Greek
776 and in many Slavic and Baltic languages, "%B" will produce the month
777 in genitive case, and "%OB" will produce the month in nominative case.
779 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
780 are all valid and will all accept any known form of month
781 name---standalone or complete, abbreviated or full. In a call to
782 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
783 the strings used by "%B" and "%b", respectively. New query
784 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
785 used by "%OB" and "%Ob", respectively.
787 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
788 define the strings for %OB and %Ob, respectively; these have the
789 same syntax as "mon" and "abmon". These arrays are optional; if they
790 are not provided then they have the same content as "mon" and "abmon",
793 These features are provided for locales which define "alt_mon" and/or
794 "ab_alt_mon" in their locale source data. This release includes such
795 alternative month name data for the following languages: Belarusian,
796 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
798 This feature is currently a GNU extension, but it is expected to
799 be added to the next revision of POSIX, and it is also already
800 available on some BSD-derived operating systems.
802 This feature will cause existing statically compiled applications
803 to fail to load locales and fall back to the builtin C/POSIX locales.
804 See notes below for other changes affecting compatibility.
806 * Support for the RISC-V ISA running on Linux has been added. This port
807 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
808 for the following ISA and ABI pairs:
814 Deprecated and removed features, and other changes affecting compatibility:
816 * Statically compiled applications attempting to load locales compiled for the
817 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
818 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
819 support for two grammatical forms of the month names, also extends the locale
820 data binary format. Static applications needing locale support must be
821 recompiled to match the runtime and data they are deployed with. In some
822 distributions there is an upgrade window where dynamically linked applications
823 may use a new library but the old locale data and also fall back to the
824 builtin C/POSIX locales; restarting the application process is sufficient to
827 * Support for statically linked applications which call dlopen is deprecated
828 and will be removed in a future version of glibc. Applications which call
829 dlopen need to be linked dynamically instead.
831 * Support for old programs which use internal stdio data structures and
832 functions is deprecated. This includes programs which use the C++ streams
833 provided by libstdc++ in GCC 2.95. Programs which use the internal
834 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
835 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
836 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
837 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
838 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
839 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
840 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
841 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
842 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
843 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
844 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
845 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
846 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
847 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
848 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
849 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
850 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
851 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
852 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
853 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
854 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
855 future version of glibc. Unlike other symbol removals, these old
856 applications will not be supported using compatibility symbols.
858 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
859 defined by <sys/ptrace.h>.
861 * libm no longer supports SVID error handling (calling a user-provided
862 matherr function on error) or the _LIB_VERSION variable to control error
863 handling. (SVID error handling and the _LIB_VERSION variable still work
864 for binaries linked against older versions of the GNU C Library.) The
865 libieee.a library is no longer provided. math.h no longer defines struct
866 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
867 TLOSS, PLOSS and HUGE.
869 * The libm functions pow10, pow10f and pow10l are no longer supported for
870 new programs. Programs should use the standard names exp10, exp10f and
871 exp10l for these functions instead.
873 * The mcontext_t type is no longer the same as struct sigcontext. On
874 platforms where it was previously the same, this changes the C++ name
875 mangling for interfaces involving this type.
877 * The add-ons mechanism for building additional packages at the same time as
878 glibc has been removed. The --enable-add-ons configure option is now
881 * The --without-fp configure option is now ignored. Whether hardware
882 floating-point instructions are used is now configured based on whether
883 the compiler used at configure time (without any options implied by a
884 --with-cpu= configure option) uses such instructions.
886 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
887 the specified string can be parsed as a domain name.
889 * In the malloc_info output, the <heap> element may contain another <aspace>
890 element, "subheaps", which contains the number of sub-heaps.
892 * The libresolv function p_secstodate is no longer supported for new
895 * The tilepro-*-linux-gnu configuration is no longer supported.
897 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
898 and will be removed in a future release. Software that is still using
899 either header should be updated to use standard <stdio.h> interfaces
902 libio.h was originally the header for a set of supported GNU extensions,
903 but they have not been maintained as such in many years, they are now
904 standing in the way of improvements to stdio, and we don't think there are
905 any remaining external users. _G_config.h was never intended for public
906 use, but predates the bits convention.
908 Changes to build and runtime requirements:
910 * bison version 2.7 or later is required to generate code in the 'intl'
913 Security related changes:
915 CVE-2009-5064: The ldd script would sometimes run the program under
916 examination directly, without preventing code execution through the
917 dynamic linker. (The glibc project disputes that this is a security
918 vulnerability; only trusted binaries must be examined using the ldd
921 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
922 suffered from a one-byte overflow during ~ operator processing (either
923 on the stack or the heap, depending on the length of the user name).
924 Reported by Tim Rühsen.
926 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
927 would sometimes fail to free memory allocated during ~ operator
928 processing, leading to a memory leak and, potentially, to a denial
931 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
932 without GLOB_NOESCAPE, could write past the end of a buffer while
933 unescaping user names. Reported by Tim Rühsen.
935 CVE-2017-17426: The malloc function, when called with an object size near
936 the value SIZE_MAX, would return a pointer to a buffer which is too small,
937 instead of NULL. This was a regression introduced with the new malloc
938 thread cache in glibc 2.26. Reported by Iain Buclaw.
940 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
941 to the allocation of too much memory. (This is not a security bug per se,
942 it is mentioned here only because of the CVE assignment.) Reported by
945 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
946 of the number of search path components. (This is not a security
947 vulnerability per se because no trust boundary is crossed if the fix for
948 CVE-2017-1000366 has been applied, but it is mentioned here only because
949 of the CVE assignment.) Reported by Qualys.
951 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
952 for AT_SECURE or SUID binaries could be used to load libraries from the
955 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
956 succeeds without returning an absolute path due to unexpected behaviour
957 of the Linux kernel getcwd syscall. Reported by halfdog.
959 CVE-2018-6485: The posix_memalign and memalign functions, when called with
960 an object size near the value of SIZE_MAX, would return a pointer to a
961 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
963 CVE-2018-6551: The malloc function, when called with an object size near
964 the value of SIZE_MAX, would return a pointer to a buffer which is too
965 small, instead of NULL.
967 The following bugs are resolved with this release:
969 [866] glob: glob should match dangling symlinks
970 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
971 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
973 [5997] math: Very slow execution of sinf function
974 [10580] localedata: hr_HR: updated locale
975 [10871] locale: 'mon' array should contain both nominative and genitive
977 [12349] localedata: eu_ES: incorrect thousands separator
978 [13605] localedata: shn_MM: new Shan locale
979 [13805] localedata: ru_RU: currency should use ',' as radix point
980 [13953] localedata: km_KH: locale update
981 [13994] localedata: mjw_IN: new locale
982 [14121] build: make writes .mo files in po directory
983 [14333] libc: Fix the race between atexit() and exit()
984 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
985 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
987 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
988 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
989 width Latin characters
990 [15332] localedata: es_CU: locale update
991 [15436] stdio: Don't close or flush stdio streams on abort
992 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
994 [16148] localedata: ca_ES: incorrect thousands separator
995 [16750] dynamic-link: ldd should not try to execute the binaries
997 [16777] localedata: pl_PL: incorrect thousands separator in locale
998 [16905] localedata: hanzi: new collation
999 [17563] localedata: cmn_TW: add hanzi collation
1000 [17750] localedata: wrong collation order of diacritics in most locales
1001 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
1003 [17956] build: Build fails on missing definitions from header file
1004 nss/nss.h when Mozilla NSS is used for cryptography
1005 [18203] libc: realpath() does not handle unreachable paths correctly
1006 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
1007 [18812] localedata: kab_DZ: new Kabyle Algeria locale
1008 [18822] libc: Internal functions are called via PLT
1009 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
1010 [19170] libc: __gmon_start__ defined in hppa in crtn.S
1011 [19574] libc: glibc should support building static PIE binaries
1012 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
1014 [19971] glob: glob: Do not skip entries with zero d_ino values
1015 [19982] localedata: fr.po: spelling mistake for error code EXDEV
1016 [20008] localedata: km_KH: convert to translit_neutral
1017 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
1018 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
1019 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
1020 [20482] localedata: de_CH: abbreviated weekdays should be two letters
1021 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
1023 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
1024 they're set, wrong errors returned by gaih_inet when lookup functions
1026 [20756] localedata: [PATCH] Use Unicode wise thousands separator
1027 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
1029 [20952] localedata: yuw_PG: new locale
1030 [21084] localedata: charmaps/IBM858: new codepage
1031 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
1032 [21242] libc: assert gives pedantic warning in old gcc versions
1033 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
1034 __regcall calling convention
1035 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
1036 [21326] libc: C99 functions are not declared for C++11 and later if
1037 _GNU_SOURCE is not predefined by g++
1038 [21457] libc: sys/ucontext.h namespace
1039 [21530] libc: tmpfile() should be implemented using O_TMPFILE
1040 [21660] math: GCC fails to compile a formula with tgmath.h
1041 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
1042 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
1043 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
1044 [21684] math: tgmath.h handling of complex integers
1045 [21685] math: tgmath.h handling of bit-fields
1046 [21686] math: tgmath.h handling of __int128
1047 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1049 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
1051 [21750] localedata: column width of characters incompatible with classical
1053 [21754] malloc: malloc: Perform as little work as possible after heap
1054 consistency check failures
1055 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
1056 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
1057 [21791] string: Unused XXX_chk_XXX functions in libc.a
1058 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
1060 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
1062 [21845] localedata: Added new Locale bho_NP
1063 [21853] localedata: Fix abday Which looks same as day in zh_SG
1064 [21854] localedata: Added New Locale en_SC
1065 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
1066 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
1067 _dl_runtime_resolve_avx_slow
1068 [21885] network: getaddrinfo: gethosts does not release resolver context
1069 on memory allocation failure
1070 [21899] libc: XPG4.2 sigaction namespace
1071 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
1073 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
1075 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
1076 errno value, causing getaddrinfo to fail
1077 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
1078 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
1079 instead of EAI_NODATA
1080 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
1081 constant PTRACE_SEIZE_DEVEL
1082 [21930] math: C-only gcc builtins used in <math.h> isinf
1083 [21932] network: Unpaired __resolv_context_get in generic get*_r
1085 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
1086 [21944] libc: sigval namespace
1087 [21951] localedata: Update hanzi collation by stroke
1088 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
1089 sysdeps/x86_64/fpu/e_expf.S
1090 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
1091 allocation in syscall loops)
1092 [21959] localedata: Fix Country name for xh_ZA
1093 [21960] localedata: Fix abmon for bem_ZM
1094 [21966] math: AVX2 mathvec functions use FMA without checking
1095 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
1096 [21971] localedata: Added New Locale for mfe_MU
1097 [21972] libc: assert macro requires operator== (int) for its argument type
1098 [21973] math: [sparc] libm missing sqrtl compat symbol
1099 [21974] libc: Remove __bb_init_func and __bb_exit_func
1100 [21982] string: stratcliff.c: error: assuming signed overflow does not
1102 [21986] stdio: __guess_grouping is called incorrectly
1103 [21987] math: [sparc32] wrong bits/long-double.h installed
1104 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
1106 [22022] localedata: Missing country_name for mni_IN
1107 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
1109 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
1111 [22026] locale: iconv_open: heap overflow on gconv_init failure
1112 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
1113 [22035] math: [m68k] bits/math-inline.h macro namespace
1114 [22038] localedata: Fix abbreviated weeks and months for Somali
1115 [22044] localedata: Remove redundant data for Limburgish Language
1116 [22050] malloc: Linking with -lmcheck does not hook
1117 __malloc_initialize_hook correctly
1118 [22051] libc: zero terminator in the middle of glibc's .eh_frame
1119 [22052] malloc: malloc failed to compile with GCC 7 and -O3
1120 [22070] localedata: charmaps/UTF-8: wcwidth for
1121 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
1122 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
1123 Jungseong and Jongseong) should be 0
1124 [22078] nss: nss_files performance issue in multi mode
1125 [22082] math: bits/math-finite.h exp10 condition
1126 [22086] libc: pcprofiledump incorrect cross-endian condition
1127 [22093] dynamic-link: ld.so no longer searches in .../x86_64
1128 [22095] network: Name server address allocation memory leak in resolv.conf
1130 [22096] network: __resolv_conf_attach can incorrectly free passed conf
1132 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
1133 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
1135 [22111] malloc: malloc: per thread cache is not returned when thread exits
1136 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
1137 [22134] libc: [linux] implement fexecve with execveat
1138 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
1140 [22145] libc: ttyname() gives up too early in the face of namespaces
1141 [22146] math: C++ build issue with float128 on x86_64
1142 [22153] nptl: nptl: save error code before process termination
1143 [22156] libc: [hppa,ia64,microblaze] Executable stack default
1144 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
1145 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
1146 [22165] libc: [hppa] Text relocations in libc.so
1147 [22180] libc: destructor registered via __cxa_atexit is called twice
1148 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
1149 symlinks") cause "make" segfaults
1150 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
1152 [22207] libc: FAIL: stdlib/test-atexit-race
1153 [22225] math: nearbyint arithmetic moved before feholdexcept
1154 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
1155 [22235] math: iscanonical in C++ and float128
1156 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
1157 the svid compat wrapper
1158 [22244] math: ynf and yn are wrong without the svid compat wrapper
1159 [22273] libc: Improper assert in Linux posix_spawn implementation
1160 [22284] libc: -pg -pie doesn't work
1161 [22292] locale: localedef exits with error 4 when it should be error 1
1162 [22294] locale: Allow "" for int_currency_symbol definition in locales.
1163 [22295] locale: Don't warn on non-symbolic characters in locale sources in
1165 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
1166 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
1168 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
1169 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
1170 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
1171 [22322] libc: [mips64] wrong bits/long-double.h installed
1172 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
1173 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
1175 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
1176 [22343] malloc: Integer overflow in posix_memalign
1177 [22347] libc: getrandom() returns the number of bytes that were copied to
1178 the buffer even though the comments say "Return 0 on success and -1 on
1180 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1181 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1183 [22370] dynamic-link: Incorrect note padding check
1184 [22375] libc: malloc returns pointer from tcache_get when should return
1185 NULL (CVE-2017-17426)
1186 [22377] math: iseqsig, float128 and C++
1187 [22382] localedata: Error in tpi_PG locale
1188 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1189 inside the ASCII printable range
1190 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1192 [22403] localedata: Slash needs escaping in some locales
1193 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1194 [22409] network: res_hnok does not accept some host names used on the
1196 [22412] network: res_dnok, res_hnok should perform syntax checks
1197 [22413] network: ns_name_pton ignores syntactically invalid trailing
1199 [22415] stdio: setvbuf can lead to invalid free/segfault
1200 [22432] build: Non-deterministic build
1201 [22439] malloc: malloc_info should compute summary statistics for all sub-
1203 [22442] network: if_nametoindex could report index for the wrong
1204 networking interface
1205 [22446] build: aliasing violation calling readlink in handle_request
1206 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1207 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1208 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1209 __nscd_hash/__nss_hash
1210 [22463] network: p_secstodate overflow handling
1211 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1212 [22478] libc: sigwait can fail with EINTR
1213 [22505] libc: ldconfig processes include directive in locale-specific
1215 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1216 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1217 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1218 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1219 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1220 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1221 same as for Croatian
1222 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1223 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1224 [22577] libc: missing newline after "cannot allocate TLS data structures
1226 [22588] manual: manual/conf.texi: missing underscore in front of
1228 [22593] math: nextafter and nexttoward are declared with const attribute
1229 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1230 [22603] string: ia64 memchr overflows internal pointer check
1231 [22605] libc: SH clone does not set the exit code correctly
1232 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1234 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1235 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1236 modified in case of success
1237 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1238 [22615] manual: manual: ambiguous wording about errno value in case of
1240 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1242 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1243 binaries or if /proc is not mounted (CVE-2017-16997)
1244 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1245 [22630] build: $(no-pie-ldflag) is no longer effective
1246 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1247 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1248 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1249 [22637] nptl: guard size is subtracted from thread stack size instead of
1251 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1252 [22657] localedata: hu_HU: Avoid double space in date
1253 [22660] math: fmax, fmin sNaN handling on alpha
1254 [22664] libc: New warning of GCC8
1255 [22665] math: alpha: ceil and floor raise inexact exceptions
1256 [22666] math: alpha: trunc raise inexact exceptions
1257 [22667] libc: makecontext lacks stack alignment on i386
1258 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1259 [22679] libc: getcwd(3) can succeed without returning an absolute path
1261 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1262 elision and tunables
1263 [22687] math: [powerpc-nofpu] complex long double functions spurious
1265 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1266 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1267 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1269 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1270 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1272 [22701] nis: Incomplete removal of libnsl
1273 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1274 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1275 [22715] dynamic-link: FAIL: elf/tst-audit10
1276 [22719] libc: Backtrace tests fail on hppa
1277 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1278 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1279 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1280 before the first call to crypt_r () in crypt/badsalttest.c
1287 * A per-thread cache has been added to malloc. Access to the cache requires
1288 no locks and therefore significantly accelerates the fast path to allocate
1289 and free small amounts of memory. Refilling an empty cache requires locking
1290 the underlying arena. Performance measurements show significant gains in a
1291 wide variety of user workloads. Workloads were captured using a special
1292 instrumented malloc and analyzed with a malloc simulator. Contributed by
1293 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1295 * Unicode 10.0.0 Support: Character encoding, character type info, and
1296 transliteration tables are all updated to Unicode 10.0.0, using
1297 generator scripts contributed by Mike FABIAN (Red Hat).
1298 These updates cause user visible changes, especially the changes in
1299 wcwidth for many emoji characters cause problems when emoji sequences
1300 are rendered with pango, see for example:
1301 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1303 * Collation of Hungarian has been overhauled and is now consistent with "The
1304 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1307 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1309 - The GNU C Library will now detect when /etc/resolv.conf has been
1310 modified and reload the changed configuration. The new resolver option
1311 “no-reload” (RES_NORELOAD) disables this behavior.
1313 - The GNU C Library now supports an arbitrary number of search domains
1314 (configured using the “search” directive in /etc/resolv.conf);
1315 previously, there was a hard limit of six domains. For backward
1316 compatibility, applications that directly modify the ‘_res’ global
1317 object are still limited to six search domains.
1319 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1320 Library will now randomly pick a name server from the configuration as a
1321 starting point. (Previously, the second name server was always used.)
1323 * The tunables feature is now enabled by default. This allows users to tweak
1324 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1326 * New function reallocarray, which resizes an allocated block (like realloc)
1327 to the product of two sizes, with a guaranteed clean failure upon integer
1328 overflow in the multiplication. Originally from OpenBSD, contributed by
1329 Dennis Wölfing and Rüdiger Sonderfeld.
1331 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1332 These are extended versions of preadv and pwritev, respectively, taking an
1333 additional flags argument. The set of supported flags depends on the
1334 running kernel; full support currently requires kernel 4.7 or later.
1336 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1337 create a new session ID for the spawned process. This feature is
1338 scheduled to be added to the next major revision of POSIX; for the time
1339 being, it is available under _GNU_SOURCE.
1341 * errno.h is now safe to use from C-preprocessed assembly language on all
1342 supported operating systems. In this context, it will only define the
1343 Exxxx constants, as preprocessor macros expanding to integer literals.
1345 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1346 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1347 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1348 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1350 To compile programs that use this feature, the compiler must support
1351 128-bit floating point with the type name _Float128 (as defined by TS
1352 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1353 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1354 must be defined to make the new interfaces visible.
1356 The new functions and macros correspond to those present for other
1357 floating-point types (except for a few obsolescent interfaces not
1358 supported for the new type), with F128 or f128 suffixes; for example,
1359 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1360 printf or scanf formats for the new type; the strfromf128 and strtof128
1361 interfaces should be used instead.
1363 Deprecated and removed features, and other changes affecting compatibility:
1365 * The synchronization that pthread_spin_unlock performs has been changed to
1366 now be equivalent to a C11 atomic store with release memory order to the
1367 spin lock's memory location. Previously, several (but not all)
1368 architectures used stronger synchronization (e.g., containing what is
1369 often called a full barrier). This change can improve performance, but
1370 may affect odd fringe uses of spin locks that depend on the previous
1371 behavior (e.g., using spin locks as atomic variables to try to implement
1372 Dekker's mutual exclusion algorithm).
1374 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1377 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1378 will only be built and installed when the GNU C Library is configured with
1379 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1380 as TIRPC or rpcsvc-proto, to be used.
1382 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1383 libnss_compat, are deprecated, and will not be built or installed by
1386 The NIS(+) support library, libnsl, is also deprecated. By default, a
1387 compatibility shared library will be built and installed, but not headers
1388 or development libraries. Only a few NIS-related programs require this
1389 library. (In particular, the GNU C Library has never required programs
1390 that use 'gethostbyname' to be linked with libnsl.)
1392 Replacement implementations based on TIRPC, which additionally support
1393 IPv6, are available from <https://github.com/thkukuk/>. The configure
1394 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1395 name service modules, to be built and installed.
1397 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1398 support is enabled, the configured recursive resolver must support EDNS.
1399 (Responding to EDNS-enabled queries with responses which are not
1400 EDNS-enabled is fine, but FORMERR responses are not.)
1402 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1403 servers have not supported this opcode for a long time.
1405 * The _res_opcodes variable has been removed from libresolv. It had been
1406 exported by accident.
1408 * <string.h> no longer includes inline versions of any string functions,
1409 as this kind of optimization is better done by the compiler. The macros
1410 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1412 * The nonstandard header <xlocale.h> has been removed. Most programs should
1413 use <locale.h> instead. If you have a specific need for the definition of
1414 locale_t with no other declarations, please contact
1415 libc-alpha@sourceware.org and explain.
1417 * The obsolete header <sys/ultrasound.h> has been removed.
1419 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1421 * The obsolete function cfree has been removed. Applications should use
1424 * The stack_t type no longer has the name struct sigaltstack. This changes
1425 the C++ name mangling for interfaces involving this type.
1427 * The ucontext_t type no longer has the name struct ucontext. This changes
1428 the C++ name mangling for interfaces involving this type.
1430 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1431 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1432 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1433 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1434 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1435 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1436 fpregset_t type no longer has the name struct fpu. This changes the C++
1437 name mangling for interfaces involving those types.
1439 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1440 synced with the kernel:
1442 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1443 are not supported on this architecture and have been removed.
1445 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1446 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1447 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1449 Programs that assume the GET/SETREGS ptrace requests are universally
1450 available will now fail to build, instead of malfunctioning at runtime.
1452 Changes to build and runtime requirements:
1454 * Linux kernel 3.2 or later is required at runtime, on all architectures
1455 supported by that kernel. (This is a change from version 2.25 only for
1458 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1460 * On most architectures, GCC 4.9 or later is required to build the GNU C
1461 Library. On powerpc64le, GCC 6.2 or later is required.
1463 Older GCC versions and non-GNU compilers are still supported when
1464 compiling programs that use the GNU C Library. (We do not know exactly
1465 how old, and some GNU extensions to C may be _de facto_ required. If you
1466 are interested in helping us make this statement less vague, please
1467 contact libc-alpha@sourceware.org.)
1469 Security related changes:
1471 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1472 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1474 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1475 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1477 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1478 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1481 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1482 fixed (CVE-2017-12133).
1484 The following bugs are resolved with this release:
1486 [984] network: Respond to changed resolv.conf in gethostbyname
1487 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1488 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1490 [12189] libc: __stack_chk_fail should not attempt a backtrace
1492 [14096] time: Race condition on timezone/tst-timezone.out
1493 [14172] localedata: az_IR: new locale
1494 [14995] build: glibc fails to build if gold is the default linker, even if
1496 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1497 [16637] network: inet_pton function is accepting IPv6 with bad format
1498 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1499 x86 and other generic code
1500 [16875] localedata: ko_KR: fix lang_name
1501 [17225] localedata: ar_SY: localized month names for May and June are
1503 [17297] localedata: da_DK: wrong date_fmt string
1504 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1505 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1506 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1507 [19066] localedata: ar_SA abbreviated day and month names are in English
1508 [19569] network: resolv: Support an arbitrary number of search domains
1509 [19570] network: Implement random DNS server selection in the stub
1511 [19838] locale: localedef fails on PA-RISC
1512 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1513 order of 0D36 and 0D37
1514 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1516 [20098] libc: FAIL: debug/backtrace-tst on hppa
1517 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1519 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1521 [20313] localedata: Update locale data to Unicode 9.0
1522 [20424] manual: Document how to provide a malloc replacement
1523 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1525 [20686] locale: Add el_GR@euro to SUPPORTED.
1526 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1527 failures consistently
1528 [21015] dynamic-link: Document and fix --enable-bind-now
1529 [21016] nptl: pthread_cond support is broken on hppa
1530 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1531 frame-pointer on i386
1532 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1534 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1535 [21088] libc: Build fails with --enable-static-nss
1536 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1537 generic c code is used
1538 [21109] libc: Tunables broken on big-endian
1539 [21112] math: powf has large ulp errors with base close to 1 and exponent
1541 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1543 [21120] malloc: glibc malloc is incompatible with GCC 7
1544 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1546 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1547 (-0.0) when linking with -lieee
1548 [21171] math: log10, log2 and lgamma return incorrect results
1549 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1550 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1551 [21207] localedata: ce_RU: update weekdays from CLDR
1552 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1553 [21217] localedata: Update months from CLDR-31
1554 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1555 [21243] libc: support_delete_temp_file should issue warning for failed
1557 [21244] libc: support resolv_test_start() socket fd close should be
1559 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1560 new posix_spawn implementation
1561 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1562 leads to lower CPU frequency
1563 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1564 [21261] libc: [sparc64] bits/setjmp.h namespace
1565 [21267] network: [mips] bits/socket.h IOC* namespace
1566 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1567 [21270] libc: mmap64 silently truncates large offset values
1568 [21275] libc: posix_spawn always crashes on ia64 now
1569 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1570 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1571 [21289] libc: Incorrect declaration for 32-bit platforms with
1572 _FILE_OFFSET_BITS=64 causes build error
1573 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1574 not support gethostbyname4_r
1575 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1577 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1579 [21340] libc: Support POSIX_SPAWN_SETSID
1580 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1581 [21359] network: ns_name_pack needs additional byte in destination buffer
1582 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1583 against fragmentation attacks (CVE-2017-12132)
1584 [21369] network: resolv: Remove EDNS fallback
1585 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1587 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1588 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1589 [21393] stdio: Missing dup3 error check in freopen, freopen64
1590 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1591 [21399] localedata: Bad description for U00EC in
1592 localedata/charmaps/CP1254
1593 [21411] malloc: realloc documentation error
1594 [21426] network: sys/socket.h uio.h namespace
1595 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1596 [21445] libc: signal.h bsd_signal namespace
1597 [21455] network: Network headers stdint.h namespace
1598 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1600 [21475] network: resolv: Overlong search path is truncated mid-label
1601 [21511] libc: sigstack namespace
1602 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1603 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1604 bits/syscall.h: No such file or directory
1605 [21517] libc: struct sigaltstack namespace
1606 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1607 [21533] localedata: Update locale data to Unicode 10.0
1609 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1610 junk at end of line, first unrecognized character is `@'
1611 [21538] libc: SIG_HOLD missing for XPG4
1612 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1613 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1614 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1615 [21543] libc: sigevent namespace
1616 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1617 size for general purpose registers in mcontext_t structure
1618 [21550] libc: sigwait namespace
1619 [21552] libc: XPG4 bsd_signal namespace
1620 [21554] libc: sigpause namespace
1621 [21560] libc: sys/wait.h signal.h namespace
1622 [21561] libc: waitid namespace
1623 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1624 '/usr/include/stdlib.h': Permission denied
1625 [21575] libc: sys/wait.h missing struct rusage definition
1626 [21584] libc: sigaltstack etc namespace
1627 [21597] libc: siginterrupt namespace
1628 [21607] math: hppa: FAIL: math/test-tgmath
1629 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1631 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1632 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1633 stack and heap (CVE-2017-1000366)
1634 [21625] libc: wait3 namespace
1635 [21654] nss: Incorrect pointer alignment in NSS group merge result
1637 [21657] network: Parse interface zone id for node-local multicast
1638 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1640 [21666] libc: .symver is used on common symbol
1641 [21668] network: resolv: res_init cross-thread broadcast introduces race
1643 [21687] math: tgmath.h totalorder, totalordermag return type
1644 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1646 [21696] libc: Incorrect assumption of of __cpu_mask in
1647 posix/sched_cpucount.c
1648 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1649 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1651 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1652 leading to relocation crash
1653 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1654 `resolv_conf_matches (resp, conf)' failed.
1655 [21710] localedata: Added Samoan language locale for Samoa
1656 [21711] localedata: Pashto yesstr/nostr locale are missing
1657 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1658 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1659 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1660 [21723] localedata: yesstr/nostr missing for Chinese language locale
1661 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1663 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1665 [21728] localedata: New Locale for Tongan language
1666 [21729] localedata: incorrect LC_NAME fields for hi_IN
1667 [21733] localedata: yesstr and nostr are missing for zh_HK
1668 [21734] localedata: Missing yesstr and nostr are for kw_GB
1669 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1670 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1671 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1672 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1674 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1676 [21749] localedata: Wrong abbreviated day name (“abday”) for
1678 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1679 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1680 [21759] localedata: missing yesstr and nostr for Tigrinya
1681 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1682 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1683 [21767] localedata: Missing Bislama locales
1684 [21768] localedata: Missing yesstr and nostr for aa_ET
1685 [21770] localedata: Missing Field in li_NL
1686 [21778] nptl: Robust mutex may deadlock
1687 [21779] libc: MicroBlaze segfaults when loading libpthread
1688 [21783] localedata: Fix int_select international_call_prefixes
1689 [21784] localedata: Inconsistency in country_isbn
1690 [21788] localedata: Missing Country Postal Abbreviations
1691 [21794] localedata: Added-country_isbn-for-Italy
1692 [21795] localedata: Add/Fix country_isbn for France
1693 [21796] localedata: Added country_isbn for Republic of Korea
1694 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1695 [21799] localedata: Added int_select international_call_prefixes
1696 [21801] localedata: Added int_select international_call_prefixes
1697 [21804] nptl: Double semicolon in thread-shared-types.h
1698 [21807] localedata: LC_ADDRESS fix for pap_CW
1699 [21808] localedata: Fix LC_ADDRESS for pap_AW
1700 [21821] localedata: Added country_name in mai_IN
1701 [21822] localedata: Fix LC_TIME for mai_IN
1702 [21823] localedata: missing yesstr, nostr for sa_IN
1703 [21825] localedata: Fix name_mrs for mag_IN
1704 [21828] localedata: 2.26 changelog should mention user visible changes
1706 [21835] localedata: Added Maithili language locale for Nepal
1707 [21838] localedata: Removed redundant data for the_NP
1708 [21839] localedata: Fix LC_MONETARY for ta_LK
1709 [21844] localedata: Fix Latin characters and Months Sequence.
1710 [21848] localedata: Fix mai_NP Title Name
1715 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1716 24731-2:2010, is supported to enable declarations of functions from that
1717 TR. Note that not all functions from that TR are supported by the GNU C
1720 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1721 18661-1:2014, is supported to enable declarations of functions and macros
1722 from that TS. Note that not all features from that TS are supported by
1725 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1726 18661-4:2015, is supported to enable declarations of functions and macros
1727 from that TS. Note that most features from that TS are not supported by
1730 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1731 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1732 Since the GNU C Library defaults to a much newer revision of POSIX, this
1733 will only affect programs that specifically request an old conformance
1734 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1735 see a change in the visible declarations, but a program compiled with
1736 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1739 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1740 defined by all multithreaded code, but glibc has not required this for
1743 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1744 means that in a future release, the macros “major”, “minor”, and “makedev”
1745 will only be available from <sys/sysmacros.h>.
1747 These macros are not part of POSIX nor XSI, and their names frequently
1748 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1749 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1750 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1753 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1754 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1755 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1757 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1758 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1759 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1760 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1761 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1762 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1763 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1764 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1765 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1766 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1767 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1770 * New <math.h> features are added from TS 18661-1:2014:
1772 - Signaling NaN macros: SNANF, SNAN, SNANL.
1774 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1775 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1776 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1778 - llogb functions: the llogb, llogbf and llogbl functions, and the
1779 FP_LLOGB0 and FP_LLOGBNAN macros.
1781 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1784 - Comparison macros: iseqsig.
1786 - Classification macros: iscanonical, issubnormal, iszero.
1788 - Total order functions: totalorder, totalorderf, totalorderl,
1789 totalordermag, totalordermagf, totalordermagl.
1791 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1793 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1794 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1796 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1797 are added to libc. They convert a floating-point number into string.
1799 * Most of glibc can now be built with the stack smashing protector enabled.
1800 It is recommended to build glibc with --enable-stack-protector=strong.
1801 Implemented by Nick Alcock (Oracle).
1803 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1804 intended to be used instead of memset() to erase sensitive data after use;
1805 the compiler will not optimize out calls to explicit_bzero even if they
1806 are "unnecessary" (in the sense that no _correct_ program can observe the
1807 effects of the memory clear).
1809 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1810 to float instead of double. This does not affect the ABI of any libraries
1811 that are part of the GNU C Library, but may affect the ABI of other
1812 libraries that use this type in their interfaces.
1814 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1815 float_t and double_t types are now defined to long double instead of float
1816 and double. These options are not the default, and this does not affect
1817 the ABI of any libraries that are part of the GNU C Library, but it may
1818 affect the ABI of other libraries that use this type in their interfaces,
1819 if they are compiled or used with those options.
1821 * The getentropy and getrandom functions, and the <sys/random.h> header file
1824 * The buffer size for byte-oriented stdio streams is now limited to 8192
1825 bytes by default. Previously, on Linux, the default buffer size on most
1826 file systems was 4096 bytes (and thus remains unchanged), except on
1827 network file systems, where the buffer size was unpredictable and could be
1828 as large as several megabytes.
1830 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1831 for the Linux quota interface which predates kernel version 2.4.22 has
1834 * The malloc_get_state and malloc_set_state functions have been removed.
1835 Already-existing binaries that dynamically link to these functions will
1836 get a hidden implementation in which malloc_get_state is a stub. As far
1837 as we know, these functions are used only by GNU Emacs and this change
1838 will not adversely affect already-built Emacs executables. Any undumped
1839 Emacs executables, which normally exist only during an Emacs build, should
1840 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1842 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1843 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1844 “no-ip6-dotint” had already been the default, and support for the
1845 “ip6-dotint” option was removed from the Internet in 2006.
1847 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1848 flag from <resolv.h> have been removed. The option relied on a
1849 backwards-incompatible DNS extension which was never deployed on the
1852 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1853 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1854 They were already unimplemented.
1856 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1857 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1858 removed again from the IETF name lookup interface specification in RFC
1859 2553. Applications should use getaddrinfo instead.
1861 * DNSSEC-related declarations and definitions have been removed from the
1862 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1863 decode the data part of DNSSEC record types. Previous versions of glibc
1864 only implemented minimal support for the previous version of DNSSEC, which
1865 is incompatible with the currently deployed version.
1867 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1868 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1869 <arpa/nameser.h> header file because the distinction between RR types and
1870 meta-RR types is not officially standardized, subject to revision, and
1871 thus not suitable for encoding in a macro.
1873 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1874 and rhook members of the res_state type in <resolv.h> have been removed.
1875 The glibc stub resolver did not support these hooks, but the header file
1876 did not reflect that.
1878 * For multi-arch support it is recommended to use a GCC which has
1879 been built with support for GNU indirect functions. This ensures
1880 that correct debugging information is generated for functions
1881 selected by IFUNC resolvers. This support can either be enabled by
1882 configuring GCC with '--enable-gnu-indirect-function', or by
1883 enabling it by default by setting 'default_gnu_indirect_function'
1884 variable for a particular architecture in the GCC source file
1887 * GDB pretty printers have been added for mutex and condition variable
1888 structures in POSIX Threads. When installed and loaded in gdb these pretty
1889 printers show various pthread variables in human-readable form when read
1890 using the 'print' or 'display' commands in gdb.
1892 * Tunables feature added to allow tweaking of the runtime for an application
1893 program. This feature can be enabled with the '--enable-tunables' configure
1894 flag. The GNU C Library manual has details on usage and README.tunables has
1895 instructions on adding new tunables to the library.
1897 * A new version of condition variables functions have been implemented in
1898 the NPTL implementation of POSIX Threads to provide stronger ordering
1901 * A new version of pthread_rwlock functions have been implemented to use a more
1902 scalable algorithm primarily through not using a critical section anymore to
1905 Security related changes:
1907 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1908 have been created with makecontext could fail to terminate due to a
1909 missing .cantunwind annotation. This has been observed to lead to a hang
1910 (denial of service) in some Go applications compiled with gccgo. Reported
1911 by Andreas Schwab. (CVE-2016-6323)
1913 * The DNS stub resolver functions would crash due to a NULL pointer
1914 dereference when processing a query with a valid DNS question type which
1915 was used internally in the implementation. The stub resolver now uses a
1916 question type which is outside the range of valid question type values.
1919 The following bugs are resolved with this release:
1921 [4099] stdio: Overly agressive caching by stream i/o functions.
1922 [7065] build: Support building glibc with -fstack-protector or -fstack-
1924 [9842] localedata: en_CA: incorrect date format
1925 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1926 before it started waiting
1927 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1928 [15765] nptl: sem_open is wrongly a cancellation point
1929 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1930 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1932 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1933 library linked with pthread
1934 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1935 "FIXME: Ingo" issue)
1936 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1937 [17252] libc: getrandom and getentropy syscall
1938 [17730] malloc: thread-local storage is sometimes improperly free()'d
1939 after being __libc_memalign()'d
1940 [18241] stdio: failed fseek on memstream does not set errno and can fail
1942 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1944 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1945 PTHREAD_PRIO_INHERIT mutex on ARM
1946 [18784] network: res_query and related function crash for special record
1947 type queries (CVE-2015-5180)
1948 [19380] math: strtod does not raise "inexact"
1949 [19387] string: Integer overflow in memchr
1950 [19390] string: Integer overflow in strncat
1951 [19398] build: linknamespace tests fail with massively parallel build
1952 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1954 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1955 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1956 [19514] libc: [PATCH] Fix spelling errors in spelling
1957 "implement"/"implementation" in several places
1958 [19582] network: Deprecate RES_USE_INET6
1959 [19673] manual: clog10 docs appear to be erroneous
1960 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1961 causes a segmentation fault
1962 [19826] libc: invalid pointer returned from __tls_get_addr with static
1964 [20016] network: resolv: Remove hooks support from the API
1965 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1967 [20033] math: [x86_64] vectorized math function don't call the __finite
1969 [20116] nptl: use after free in pthread_create
1970 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1971 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1973 [20311] nptl: please install proc_service.h
1974 [20366] libc: Compilation errors in installed headers in strict-compliance
1976 [20370] malloc: malloc: Arena free list management is still racy
1977 (incorrect fix in bug 19243)
1978 [20386] libc: assert (X = 0) does not result in GCC warning
1979 [20432] malloc: malloc: Minimize interface required for interposition
1980 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1981 in _Unwind_Backtrace (CVE-2016-6323)
1982 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1983 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1985 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1986 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1987 [20477] network: resolv: incorrect double-checked locking related to
1989 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1991 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1993 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1994 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1995 cause transition penalty
1996 [20517] math: sparcv9 missing fdiml compat symbol
1997 [20524] manual: strverscmp is inconsistent
1998 [20525] libc: <sys/quota.h> should be based on kernel headers
1999 [20539] math: GCC 7 -static -lm fails to link at -O3
2000 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
2002 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
2004 [20591] network: Remove obsolete DNSSEC support
2005 [20592] network: DNS resource record type classification macros in
2006 <arpa/nameser.h> are incorrect
2007 [20593] network: Update DNS RR type definitions
2008 [20611] network: getaddrinfo accepts invalid numeric scope IDs
2009 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
2011 [20629] network: libresolv: Remove support for bitstring labels
2013 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
2014 assertion: "maxidx >= 2"
2015 [20660] math: [arm] Use VSQRT
2016 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
2017 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
2018 [20689] libc: Test for FMA should also check for AVX.
2019 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
2020 [20715] math: iszero macro breaks existing code
2021 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
2022 [20728] libc: powerpc: Missing TOC stub in clone
2023 [20729] build: glibc-2.24 fails to build for i486 with -Os
2024 [20750] build: Build fails with default PIE enabled gcc-6.x
2025 [20768] math: [alpha] sqrt fegetenv namespace
2026 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
2028 [20787] math: float_t is defined as float incorrectly on x86_64 even if
2029 __FLT_EVAL_METHOD__ is 2
2030 [20790] network: rpcgen buffer overrun in get_prog_declaration
2031 [20822] nptl: race condition in __lll_unlock_elision on powerpc
2032 [20829] libc: crypt snprintf namespace
2033 [20847] libc: tst-vfork3 failure
2034 [20855] math: Default bits/mathdef.h has inappropriate float_t
2035 [20859] math: [sh4] FP_ILOGB0 invalid
2036 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
2037 U+20AC), not same as GBK
2038 [20915] dynamic-link: global-dynamic TLS broken on aarch64
2039 [20916] math: pow handling of sNaN arguments
2040 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
2041 [20919] math: Bad pow (qNaN, 0) result with -lieee
2042 [20940] math: hypot sNaN handling
2043 [20947] math: fmax, fmin sNaN handling
2044 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
2045 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
2047 [20971] string: powerpc64/power7 memchr overflows internal pointer check
2048 [20973] nptl: robust mutexes: Lost wake-ups
2049 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
2050 [20978] nis: strlen on null pointer in nss_nisplus
2051 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
2053 [21014] string: i686 memchr overflows internal pointer check
2054 [21019] libc: [mips] n32 lseek incorrect on overflow
2055 [21022] libc: [microblaze] __backtrace get_frame_size namespace
2056 [21026] libc: [MIPS] readahead syscall is broken on n64
2057 [21028] math: Fallback fesetexceptflag should always succeed
2058 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
2059 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
2061 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
2062 [21061] librt: [microblaze] librt lost clock_* exports
2063 [21073] libc: tunables: insecure environment variables passed to
2064 subprocesses with AT_SECURE
2065 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
2069 * The minimum Linux kernel version that this version of the GNU C Library
2070 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
2071 kernel version 2.6.32 or later suffices (on architectures that already
2072 required kernel versions more recent than 3.2, those requirements remain
2073 unchanged). Linux 3.2 or later kernel headers are required on all
2076 * The pap_AN locale has been deleted. This has been deprecated for a long
2077 time. It has been replaced by pap_AW & pap_CW, both of which have long
2078 been included in previous releases.
2080 * The readdir_r and readdir64_r functions have been deprecated. It is
2081 recommended to use readdir and readdir64 instead.
2083 * The type “union wait” has been removed. It was deprecated in the early
2084 1990s and never part of POSIX. Application code should use the int type
2085 instead of “union wait”.
2087 * A new NSS action is added to facilitate large distributed system
2088 administration. The action, MERGE, allows remote user stores like LDAP
2089 to be merged into local user stores like /etc/groups in order to provide
2090 easy to use, updated, and managed sets of merged credentials. The new
2091 action can be used by configuring it in /etc/nsswitch.conf:
2092 group: files [SUCCESS=merge] nis
2093 Implemented by Stephen Gallagher (Red Hat).
2095 * The deprecated __malloc_initialize_hook variable has been removed from the
2098 * The long unused localedef --old-style option has been removed. It hasn't
2099 done anything in over 16 years. Scripts using this option can safely
2102 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
2103 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
2104 return the next representable value in the direction of positive infinity
2105 and the nextdown functions return the next representable value in the
2106 direction of negative infinity. These are currently enabled as GNU
2109 Security related changes:
2111 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
2112 could result in a stack overflow when getnetbyname was called with an
2113 overly long name. (CVE-2016-3075)
2115 * Previously, getaddrinfo copied large amounts of address data to the stack,
2116 even after the fix for CVE-2013-4458 has been applied, potentially
2117 resulting in a stack overflow. getaddrinfo now uses a heap allocation
2118 instead. Reported by Michael Petlan. (CVE-2016-3706)
2120 * The glob function suffered from a stack-based buffer overflow when it was
2121 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
2122 Reported by Alexander Cherepanov. (CVE-2016-1234)
2124 * The Sun RPC UDP client could exhaust all available stack space when
2125 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
2126 alloca plugin for GCC. (CVE-2016-4429)
2128 * The IPv6 name server management code in libresolv could result in a memory
2129 leak for each thread which is created, performs a failing naming lookup,
2130 and exits. Over time, this could result in a denial of service due to
2131 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
2133 The following bugs are resolved with this release:
2135 [1170] localedata: ne_NP: update Nepali locale definition file
2136 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
2138 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
2139 [6796] math: fdim() does not set errno on overflow
2140 [10354] libc: posix_spawn should use vfork() in more cases than presently
2141 [11213] localedata: localedata: add copyright disclaimer to locale files
2142 [12143] localedata: chr_US: new Cherokee locale
2143 [12450] localedata: sgs_LT: new locale
2144 [12676] localedata: ln_CD: new locale
2145 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
2147 [13304] math: fma, fmaf, fmal produce wrong results
2148 [14259] build: --localedir arg to configure is ignored
2149 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
2150 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
2151 [14934] localedata: es_CL: wrong first weekday chilean locale
2152 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
2154 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
2156 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
2157 [15368] nptl: raise() is not async-signal-safe
2158 [15479] math: ceil, floor, round and trunc raise inexact exception
2159 [15578] localedata: kk_KZ: various updates
2160 [16003] localedata: pap_AN: punt old locale
2161 [16137] localedata: iw_IL: punt old locale
2162 [16190] localedata: eo: new esperanto locale
2163 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
2164 [16742] malloc: race condition: pthread_atfork() called before first
2165 malloc() results in unexpected locking behaviour/deadlocks
2166 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
2168 [16983] localedata: postal_fmt does not allow %l and %n modifiers
2169 [17565] localedata: pt_PT: wrong (work-)week start
2170 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
2171 [17950] build: Build fails with -msse
2172 [18205] localedata: be_BY*: wrong first_weekday and first_workday
2173 [18433] libc: posix_spawn does not return correctly upon failure to
2175 [18453] localedata: charmaps/IBM875: incorrect codes
2176 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2178 [18896] localedata: he_IL: improvements for currency
2179 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2180 Romanian locale data
2181 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2183 [19056] libc: Deprecate readdir_r
2184 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2186 [19198] localedata: nl_NL: small improvements for Dutch locales
2187 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2188 nameservers (CVE-2016-5417)
2189 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2191 [19400] locale: Language missing in "iso-639.def", trivial fix in
2193 [19431] malloc: Deadlock between fflush, getdelim, and fork
2194 [19505] libc: Incorrect file descriptor validity checks in
2195 posix_spawn_file_actions_add{open,close,dup2}
2196 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2197 when using RTLD_NEXT
2198 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2199 `intl/{gettextP,loadinfo}.h'
2200 [19534] libc: execle, execlp may use malloc
2201 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2202 [19573] network: res_nclose and __res_maybe_init disagree about name
2203 server initialization, breaking Hesiod
2204 [19575] localedata: Status of GB18030 tables
2205 [19581] localedata: sr_* date_fmt string contains additional newline
2206 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2208 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2209 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2210 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2211 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2213 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2215 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2217 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2218 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2219 [19633] locale: strfmon_l applies global locale to number formatting
2220 [19642] network: Memory leak in getnameinfo
2221 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2222 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2224 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2225 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2227 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2228 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2230 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2232 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2234 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2235 pointers and lengths in error-case.
2236 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2237 report errors on UTF-16 surrogates values.
2238 [19755] nscd: nscd assertion failure in gc
2239 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2240 [19759] libc: mempcpy shouldn't be inlined
2241 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2242 [19765] libc: s390 needs an optimized mempcpy
2243 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2244 NAME_MAX limit assumption (CVE-2016-1234)
2245 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2246 [19787] network: Missing and incorrect truncation checks in getnameinfo
2247 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2249 [19791] network: Assertion failure in res_query.c with un-connectable name
2251 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2252 [19822] math: libm.so install clobbers old version
2253 [19825] network: resolv: send_vc can return uninitialized data in second
2254 response to getaddrinfo
2255 [19830] network: nss_dns: should check RDATA length against buffer length
2256 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2257 confronted with A/AAAA records of invalid size
2258 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2259 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2260 [19853] stdio: Printing IBM long double in decimal with high precision is
2262 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2263 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2264 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2266 [19865] network: Assertion failure or memory leak in
2267 _nss_dns_getcanonname_r
2268 [19868] network: nss_dns: netent code does not skip over non-PTR records
2269 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2271 [19881] string: Improve x86-64 memset
2272 [19907] string: Incorrect memcpy tests
2273 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2275 [19925] libc: termios.h XCASE namespace
2276 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2277 [19929] libc: limits.h NL_NMAX namespace
2278 [19931] stdio: Memory leak in vfprintf
2279 [19957] libc: clone(CLONE_VM) access invalid parent memory
2280 [19963] localedata: en_IL: New locale
2281 [19989] stdio: stdio.h cuserid namespace
2282 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2284 [19996] locale: langinfo.h nl_langinfo_l namespace
2285 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2286 [20010] network: getaddrinfo: Stack overflow in hostent translation
2288 [20012] stdio: libio: fmemopen append mode failure
2289 [20014] stdio: stdio.h namespace for pre-threads POSIX
2290 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2291 [20023] libc: fcntl.h timespec namespace
2292 [20024] math: [x86_64] vectorized sincos trashes the stack
2293 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2294 [20041] time: sys/time.h timespec namespace
2295 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2296 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2297 [20051] libc: ttyslot in wrong header under wrong conditions
2298 [20054] libc: gethostname not declared for XPG4
2299 [20055] libc: termios.h missing tcgetsid for XPG4
2300 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2302 [20073] libc: sys/stat.h fchmod namespace
2303 [20074] libc: stdlib.h rand_r namespace
2304 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2305 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2307 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2308 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2310 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2311 [20119] libc: Wrong mask for processors level type from CPUID
2312 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2313 AS not supporting AVX512
2314 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2316 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2317 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2319 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2320 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2321 [20178] libc: posix_spawn{p} should not call exit
2322 [20191] stdio: libio: vtables hardening
2323 [20195] string: FMA4 detection requires CPUID execution with register
2325 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2326 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2328 [20212] math: acos (sNaN) returns sNaN
2329 [20213] math: asin (sNaN) returns sNaN
2330 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2331 [20218] math: [i386] asinhl (sNaN) returns sNaN
2332 [20219] math: [i386] atanhl (sNaN) returns sNaN
2333 [20222] stdio: fopencookie: Mangle function pointers
2334 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2335 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2336 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2338 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2339 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2340 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2341 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2342 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2343 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2344 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2345 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2346 [20240] math: modf (sNaN) returns sNaN
2347 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2348 [20250] math: frexp (sNaN) returns sNaN
2349 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2350 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2352 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2353 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2354 subscript is below array bounds [-Werror=array-bounds]
2355 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2356 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2358 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2359 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2361 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2362 "invalid" exceptions
2363 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2364 [20316] localedata: id_ID: Februari instead of Pebruari
2365 [20327] string: POWER8 strcasecmp returns incorrect result
2366 [20347] math: Failure: Test: j0_downward (0xap+0)
2367 [20348] libc: FAIL: misc/tst-preadvwritev64
2368 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2369 [20350] libc: There is no test for p{read,write}64
2370 [20357] math: Incorrect cos result for 1.5174239687223976
2371 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2375 * Unicode 8.0.0 Support: Character encoding, character type info, and
2376 transliteration tables are all updated to Unicode 8.0.0, using new
2377 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2378 These updates cause user visible changes, such as the fixes for bugs
2379 89, 16061, and 18568.
2381 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2382 kernel-internal CPU set size. This means that requests that change the
2383 CPU affinity which failed before (for example, an all-ones CPU mask) will
2384 now succeed. Applications that need to determine the effective CPU
2385 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2386 setting it because the kernel can adjust it (and the previous size check
2387 would not detect this in the majority of cases).
2389 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2390 the following new symbols are used: fts64_children, fts64_close,
2391 fts64_open, fts64_read and fts64_set.
2393 * getaddrinfo now detects certain invalid responses on an internal netlink
2394 socket. If such responses are received, an affected process will
2395 terminate with an error message of "Unexpected error <number> on netlink
2396 descriptor <number>" or "Unexpected netlink response of size <number> on
2397 descriptor <number>". The most likely cause for these errors is a
2398 multi-threaded application which erroneously closes and reuses the netlink
2399 file descriptor while it is used by getaddrinfo.
2401 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2402 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2403 unnecessary serialization of memory allocation requests across threads.
2404 The defect is now corrected. Users should see a substantial increase in
2405 the concurent throughput of allocation requests for applications which
2406 trigger this bug. Affected applications typically create and
2407 destroy threads frequently. (Bug 19048 was reported and analyzed by
2410 * There is now a --disable-timezone-tools configure option for disabling the
2411 building and installing of the timezone related utilities (zic, zdump, and
2412 tzselect). This is useful for people who build the timezone data and code
2413 independent of the GNU C Library.
2415 * The obsolete header <regexp.h> has been removed. Programs that require
2416 this header must be updated to use <regex.h> instead.
2418 * The obsolete functions bdflush, create_module, get_kernel_syms,
2419 query_module and uselib are no longer available to newly linked binaries;
2420 the header <sys/kdaemon.h> has been removed. These functions and header
2421 were specific to systems using the Linux kernel and could not usefully be
2422 used with the GNU C Library on systems with version 2.6 or later of the
2425 * Optimized string, wcsmbs and memory functions for IBM z13.
2426 Implemented by Stefan Liebler.
2428 * Newly linked programs that define a variable called signgam will no longer
2429 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2430 require signgam to be set by those functions must ensure that they use the
2431 variable provided by the GNU C Library and declared in <math.h>, without
2432 defining their own copy.
2434 * The minimum GCC version that can be used to build this version of the GNU
2435 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2436 still be used to compile programs using the GNU C Library.
2438 Security related changes:
2440 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2441 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2443 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2444 the pointer guard feature. It is always enabled. Previously,
2445 LD_POINTER_GUARD could be used to disable security hardening in binaries
2446 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2449 * An integer overflow in hcreate and hcreate_r could lead to an
2450 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2452 * The catopen function no longer has unbounded stack usage. Reported by
2453 Max. (CVE-2015-8779)
2455 * The nan, nanf and nanl functions no longer have unbounded stack usage
2456 depending on the length of the string passed as an argument to the
2457 functions. Reported by Joseph Myers. (CVE-2014-9761)
2459 * A stack-based buffer overflow was found in libresolv when invoked from
2460 libnss_dns, allowing specially crafted DNS responses to seize control
2461 of execution flow in the DNS client. The buffer overflow occurs in
2462 the functions send_dg (send datagram) and send_vc (send TCP) for the
2463 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2464 family. The use of AF_UNSPEC triggers the low-level resolver code to
2465 send out two parallel queries for A and AAAA. A mismanagement of the
2466 buffers used for those queries could result in the response of a query
2467 writing beyond the alloca allocated buffer created by
2468 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2469 the overflow. Thanks to the Google Security Team and Red Hat for
2470 reporting the security impact of this issue, and Robert Holiday of
2471 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2473 The following bugs are resolved with this release:
2475 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2476 [887] math: Math library function "logb" and "nextafter" inconsistent
2477 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2478 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2479 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2480 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2482 [4404] localedata: German translation of "Alarm clock" is misleading
2483 [6799] math: nextafter() and nexttoward() doen't set errno on
2484 overflow/underflow errors
2485 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2487 [10432] nis: _nss_nis_setnetgrent assertion failure
2488 [11460] libc: fts has no LFS support
2489 [12926] network: getaddrinfo()/make_request() may spin forever
2490 [13065] nptl: Race condition in pthread barriers
2491 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2492 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2494 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2495 [14912] libc: Rename non-installed bits/*.h headers
2496 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2497 [15367] math: Let gcc use __builtin_isinf
2498 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2499 [15421] math: lgamma wrongly sets signgam for ISO C
2500 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2501 FE_INVALID with argument out of range
2502 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2504 [15786] dynamic-link: ifunc resolver functions can smash function
2506 [15918] math: Unnecessary check for equality in hypotf()
2507 [16061] localedata: Review / update transliteration data
2508 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2510 [16141] time: strptime %z offset restriction
2511 [16171] math: drem should be alias of remainder
2512 [16296] math: fegetround is pure?
2513 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2514 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2516 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2517 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2519 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2520 "Exception "Inexact" set" on ppc32
2521 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2522 [16517] math: Missing underflow exception from tanf/tan/tanl
2523 [16519] math: Missing underflow exception from sinhf
2524 [16520] math: Missing underflow exception from tanhf
2525 [16521] math: Missing underflow exception from exp2
2526 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2528 [16734] stdio: fopen calls mmap to allocate its buffer
2529 [16961] math: nan function incorrect handling of bad sequences
2530 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2531 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2532 [16985] locale: localedef: confusing error message when opening output
2534 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2535 [17197] locale: Redundant shift character in iconv conversion output at
2537 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2538 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2539 [17250] dynamic-link: static linking breaks nss loading
2540 (getaddrinfo/getpwnam/etc...)
2541 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2543 [17441] math: isnan() should use __builtin_isnan() in GCC
2544 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2545 (related to lock elision)
2546 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2547 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2548 [17887] time: strptime should be able to parse "+01:00" style timezones
2549 [17905] libc: catopen() Multiple unbounded stack allocations
2551 [18084] libc: backtrace (..., 0) dumps core on x86
2552 [18086] libc: nice() sets errno to 0 on success
2553 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2554 is too large (CVE-2015-8778)
2555 [18251] dynamic-link: SONAME missing when audit modules provides path
2556 [18265] libc: add attributes for wchar string and memory functions
2557 [18370] math: csqrt missing underflows
2558 [18421] libc: [hppa] read-only segment has dynamic relocations
2559 [18472] libc: Obsolete syscall wrappers should be compat symbols
2560 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2561 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2562 [18525] localedata: Remove locale timezone information
2563 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2564 [18568] localedata: Update locale data to Unicode 8.0
2565 [18589] locale: sort-test.sh fails at random
2566 [18595] math: ctan, ctanh missing underflows
2567 [18604] libc: assert macro-expands its argument
2568 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2569 contains a vector instruction exception.
2570 [18611] math: j1, jn missing errno setting on underflow
2571 [18618] localedata: sync Chechen locale definitions with other *_RU
2573 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2574 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2575 [18665] network: In send_dg, the recvfrom function is NOT always using the
2576 buffer size of a newly created buffer (CVE-2015-7547)
2577 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2579 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2581 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2582 [18699] math: tilegx cproj() for various complex infinities does not yield
2584 [18724] libc: Harden put*ent functions against data injection
2585 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2586 [18755] build: build errors with -DNDEBUG
2587 [18757] stdio: fmemopen fails to set errno on failure
2588 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2590 [18781] libc: openat64 lacks O_LARGEFILE
2591 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2592 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2593 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2594 [18790] math: [ldbl-128ibm] tanhl inaccurate
2595 [18795] libc: stpncpy fortification misses buffer lengths that are
2596 statically too large
2597 [18796] build: build fails for --disable-mathvec
2598 [18803] math: hypot missing underflows
2599 [18820] stdio: fmemopen may leak memory on failure
2600 [18823] math: csqrt spurious underflows
2601 [18824] math: fma spurious underflows
2602 [18825] math: pow missing underflows
2603 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2604 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2605 [18870] build: sem_open.c fails to compile with missing symbol
2607 [18872] stdio: Fix memory leak in printf_positional
2608 [18873] libc: posix_fallocate overflow check ineffective
2609 [18875] math: Excess precision leads incorrect libm
2610 [18877] libc: arm: mmap offset regression
2611 [18887] libc: memory corruption when using getmntent on blank lines
2612 [18918] localedata: hu_HU: change time to HH:MM:SS format
2613 [18921] libc: Regression: extraneous stat() and fstat() performed by
2615 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2616 binaries (CVE-2015-8777)
2617 [18951] math: tgamma missing underflows
2618 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2620 [18953] localedata: lt_LT: change currency symbol to the euro
2621 [18956] math: powf inaccuracy
2622 [18961] math: [i386] exp missing underflows
2623 [18966] math: [i386] exp10 missing underflows
2624 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2625 [18969] build: multiple string test failures due to missing locale
2627 [18970] libc: Reference of pthread_setcancelstate in libc.a
2628 [18977] math: float / long double Bessel functions not in XSI POSIX
2629 [18980] math: i386 libm functions return with excess range and precision
2630 [18981] math: i386 scalb*, ldexp return with excess range and precision
2631 [18982] stdio: va_list and vprintf
2632 [18985] time: Passing out of range data to strftime() causes a segfault
2634 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2635 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2636 [19012] locale: iconv_open leaks memory on error path
2637 [19016] math: clog, clog10 inaccuracy
2638 [19018] nptl: Mangle function pointers in tls_dtor_list
2639 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2640 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2641 [19048] malloc: malloc: arena free list can become cyclic, increasing
2643 [19049] math: [powerpc] erfc incorrect zero sign
2644 [19050] math: [powerpc] log* incorrect zero sign
2645 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2646 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2647 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2648 [19074] network: Data race in _res_hconf_reorder_addrs
2649 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2650 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2651 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2653 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2655 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2656 [19086] manual: posix_fallocate64 documented argument order is wrong.
2657 [19088] math: lround, llround missing exceptions close to overflow
2659 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2660 [19095] math: dbl-64 lrint incorrect for 64-bit long
2661 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2662 [19124] dynamic-link: ld.so failed to build with older assmebler
2663 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2664 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2665 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2666 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2667 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2668 pthread_setaffinity_np
2669 [19156] math: [ldbl-128] j0l spurious underflows
2670 [19164] nptl: tst-getcpu fails with many possible CPUs
2671 [19168] math: math/test-ildoubl and math/test-ldouble failure
2672 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2673 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2675 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2676 (FE_NOMASK_ENV) do not clear SSE exceptions
2677 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2678 _int_new_arena/reused_arena
2679 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2680 [19201] math: dbl-64 remainder incorrect sign of zero result
2681 [19205] math: bits/math-finite.h conditions do not match math.h and
2683 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2684 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2685 for C99-based standards
2686 [19212] libc: features.h not -Wundef clean
2687 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2689 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2690 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2691 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2693 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2694 "inexact" exceptions
2695 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2697 [19242] libc: strtol incorrect in Turkish locales
2698 [19243] malloc: reused_arena can pick an arena on the free list, leading
2699 to an assertion failure and reference count corruption
2700 [19253] time: tzset() ineffective when temporary TZ did not include DST
2702 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2703 [19270] math: [hppa] Shared libm missing __isnanl
2704 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2705 MAP_HUGETLB and MAP_STACK defines
2706 [19313] nptl: Wrong __cpu_mask for x32
2707 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2708 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2709 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2710 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2711 [19363] time: x32: times() return value wrongly truncates/sign extends
2713 [19367] dynamic-link: Improve branch prediction on Silvermont
2714 [19369] network: Default domain name not reset by res_ninit when "search"
2715 / "domain" entry is removed from resolv.conf
2716 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2718 [19385] localedata: bg_BG: time separator should be colon, not comma
2719 [19408] libc: linux personality syscall wrapper may erroneously return an
2720 error on 32-bit architectures
2721 [19415] libc: dladdr returns wrong names on hppa
2722 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2723 IBM905, IBM907, and IBM909
2724 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2725 [19443] build: build failures with -DDEBUG
2726 [19451] build: Make check fails on test-double-vlen2
2727 [19462] libc: Glibc failed to build with -Os
2728 [19465] math: Wrong code with -Os
2729 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2731 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2733 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2734 [19486] math: S390: Math tests fail with "Exception Inexact set".
2735 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2736 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2738 [19590] math: Fail to build shared objects that use libmvec.so functions.
2742 * The following bugs are resolved with this release:
2744 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2745 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2746 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2747 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2748 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2749 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2750 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2751 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2752 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2753 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2754 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2755 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2756 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2757 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2758 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2759 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2760 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2761 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2762 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2763 18657, 18676, 18694, 18696, 18887.
2765 * Cache information can be queried via sysconf() function on s390 e.g. with
2766 _SC_LEVEL1_ICACHE_SIZE as argument.
2768 * A buffer overflow in gethostbyname_r and related functions performing DNS
2769 requests has been fixed. If the NSS functions were called with a
2770 misaligned buffer, the buffer length change due to pointer alignment was
2771 not taken into account. This could result in application crashes or,
2772 potentially arbitrary code execution, using crafted, but syntactically
2773 valid DNS responses. (CVE-2015-1781)
2775 * The time zone file parser has been made more robust against crafted time
2776 zone files, avoiding heap buffer overflows related to the processing of
2777 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2778 large time zone data files. Overly long time zone specifiers in the TZ
2779 variable no longer result in stack overflows and crashes.
2781 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2782 for LD and GD on x86 and x86-64, has been implemented. You will need
2783 binutils-2.24 or later to enable this optimization.
2785 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2786 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2787 Hat). These updates cause user visible changes, such as the fix for bug
2790 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2791 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2792 condition in some applications.
2794 * Added vector math library named libmvec with the following vectorized x86_64
2795 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2797 The library can be disabled with --disable-mathvec. Use of the functions is
2798 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2799 Shared library libmvec.so is linked in as needed when using -lm (no need to
2800 specify -lmvec explicitly for not static builds).
2801 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2803 * A new fmemopen implementation has been added with the goal of POSIX
2804 compliance. The new implementation fixes the following long-standing
2805 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2806 old implementation is still present for use be by existing binaries.
2808 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2809 releases. It has been fixed to match 2.19 and older, but binaries built
2810 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2812 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2813 Contributed by Roland McGrath (Google).
2815 * The header <regexp.h> is deprecated, and will be removed in a future
2816 release. Use of this header will trigger a deprecation warning.
2817 Application developers should update their code to use <regex.h> instead.
2819 This header was formerly part of SUS, but was deprecated in 1994 and
2820 removed from the standard in 2001. Also, the glibc implementation
2821 leaks memory. See BZ#18681 for more details.
2825 * The following bugs are resolved with this release:
2827 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2828 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2829 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2830 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2831 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2832 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2833 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2834 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2835 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2836 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2839 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2840 too little memory for the to-be-scanned arguments and overflow the
2841 allocated buffer. The implementation now correctly computes the required
2842 buffer size when using malloc, and switches to malloc from alloca as
2845 * A new semaphore algorithm has been implemented in generic C code for all
2846 machines. Previous custom assembly implementations of semaphore were
2847 difficult to reason about or ensure that they were safe. The new version
2848 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2849 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2850 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2852 * Port to Altera Nios II has been contributed by Mentor Graphics.
2854 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2855 implementations for powerpc64/powerpc64le.
2856 Implemented by Adhemerval Zanella (IBM).
2858 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2859 and powerpc64le. This may improve lock scaling of existing programs on
2860 HTM capable systems. The lock elision code is only enabled with
2861 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2862 powerpc will issue a transaction abort on every syscall to avoid side
2863 effects being visible outside transactions.
2865 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2866 AArch64. Contributed by ARM Ltd.
2868 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2870 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2871 under certain input conditions resulting in the execution of a shell for
2872 command substitution when the application did not request it. The
2873 implementation now checks WRDE_NOCMD immediately before executing the
2874 shell and returns the error WRDE_CMDSUB as expected.
2876 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2877 processing format strings with a large number of format specifiers.
2879 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2880 infinite loop if the DNS response contained a PTR record of an unexpected
2883 * The minimum GCC version that can be used to build this version of the GNU
2884 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2885 still be used to compile programs using the GNU C Library.
2887 * The GNU C Library is now built with -Werror by default. This can be
2888 disabled by configuring with --disable-werror.
2890 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2892 * The obsolete sigvec function has been removed. This was the original
2893 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2894 programs have been using instead for about 25 years. Of course, ABI
2895 compatibility for old binaries using sigvec remains intact.
2897 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2898 with newer versions of bison.
2900 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2901 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2902 registers overlay two consecutive single-precision registers. MIPS32R2
2903 introduced a new FPU mode (FR=1) where double-precision registers extend the
2904 corresponding single-precision registers which is incompatible with the
2905 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2906 require the use of FR=1 making a transition necessary. New o32 ABI
2907 extensions enable users to migrate over time from the original o32 ABI
2908 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2909 tracks the ABI of any loaded object and verifies that new objects are
2910 compatible. Mode transitions will also be requested as required and
2911 unsupportable objects will be rejected. The ABI checks include both soft and
2912 hard float ABIs for o32, n32 and n64.
2914 GCC 5 with GNU binutils 2.25 onwards:
2915 It is strongly recommended that all o32 system libraries are built using the
2916 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2917 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2918 compiler using --with-fp-32=xx to set this by default.
2922 * The following bugs are resolved with this release:
2924 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2925 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2926 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2927 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2928 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2929 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2930 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2931 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2932 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2933 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2934 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2935 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2936 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2937 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2938 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2940 * Reverted change of ABI data structures for s390 and s390x:
2941 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2942 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2943 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2944 are preserved pointing straight to the same implementation as the old ones.
2945 Given that, new callers will simply provide a too-big buffer to these
2946 functions. Any applications/libraries out there that embed jmp_buf or
2947 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2948 against 2.19 headers will have to rebuilt again. This is necessary in any
2949 case to revert the breakage in their ABI caused by the glibc change.
2951 * Support for file description locks is added to systems running the
2952 Linux kernel. The standard file locking interfaces are extended to
2953 operate on file descriptions, not file descriptors, via the use of
2954 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2955 are associated with an open file instead of a process.
2957 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2959 * The minimum Linux kernel version that this version of the GNU C Library
2960 can be used with is 2.6.32.
2962 * Running the testsuite no longer terminates as soon as a test fails.
2963 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2964 with PASS or FAIL lines for individual tests. A summary of the results is
2965 printed, including a list of failing lists, and "make check" exits with
2966 error status if there were any unexpected failures. "make check
2967 stop-on-test-failure=y" may be used to keep the old behavior.
2969 * The am33 port, which had not worked for several years, has been removed
2972 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2973 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2974 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2975 conflicted with POSIX (support for which was removed in 2.19), the
2976 interfaces those macros enabled remain available when compiling with
2977 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2978 test macros defined.
2980 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2982 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2983 This may improve lock scaling of existing programs on TX capable systems.
2984 The lock elision code is only built with --enable-lock-elision=yes and
2985 then requires a GCC version supporting the TX builtins. With lock elision
2986 default mutexes are elided via __builtin_tbegin, if the cpu supports
2987 transactions. By default lock elision is not enabled and the elision code
2990 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2991 copy the path argument. This allowed programs to cause posix_spawn to
2992 deference a dangling pointer, or use an unexpected pathname argument if
2993 the string was modified after the posix_spawn_file_actions_addopen
2996 * All supported architectures now use the main glibc sysdeps directory
2997 instead of some being in a separate "ports" directory (which was
2998 distributed separately before glibc 2.17).
3000 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
3001 On configurations that support it (all Linux configurations), it's now
3002 used regardless of the --enable-add-ons switch to configure. It is no
3003 longer possible to build such configurations without pthreads support.
3005 * Locale names, including those obtained from environment variables (LANG
3006 and the LC_* variables), are more tightly checked for proper syntax.
3007 setlocale will now fail (with EINVAL) for locale names that are overly
3008 long, contain slashes without starting with a slash, or contain ".." path
3009 components. (CVE-2014-0475) Previously, some valid locale names were
3010 silently replaced with the "C" locale when running in AT_SECURE mode
3011 (e.g., in a SUID program). This is no longer necessary because of the
3014 * On x86-64, the dynamic linker's lazy-binding support is now compatible
3015 with application code using Intel MPX instructions. (With all previous
3016 versions, the MPX register state could be clobbered when making calls
3017 into or out of a shared library.) Note that while the new dynamic
3018 linker is compatible with all known x86 hardware whether or not it
3019 supports Intel MPX, some x86 instruction-set emulators might fail to
3020 handle the new instruction encodings. This is known to affect Valgrind
3021 versions up through 3.9 (but will be fixed in the forthcoming 3.10
3022 release), and might affect other tools that do instruction emulation.
3024 * Support for loadable gconv transliteration modules has been removed.
3025 The support for transliteration modules has been non-functional for
3026 over a decade, and the removal is prompted by security defects. The
3027 normal gconv conversion modules are still supported. Transliteration
3028 with //TRANSLIT is still possible, and the //IGNORE specifier
3029 continues to be supported. (CVE-2014-5119)
3031 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
3032 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
3033 resulting a denial-of-service security vulnerability in applications which
3034 use functions related to iconv. (CVE-2014-6040)
3038 * The following bugs are resolved with this release:
3040 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
3041 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
3042 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
3043 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
3044 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
3045 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
3046 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
3047 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
3048 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
3049 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
3050 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
3051 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
3052 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
3053 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
3054 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
3055 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
3056 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
3057 16431, 16453, 16474, 16506, 16510, 16529
3059 * Slovenian translations for glibc messages have been contributed by the
3060 Translation Project's Slovenian team of translators.
3062 * The public headers no longer use __unused nor __block. This change is to
3063 support compiling programs that are derived from BSD sources and use
3064 __unused internally, and to support compiling with Clang's -fblock
3065 extension which uses __block.
3067 * CVE-2012-4412 The strcoll implementation caches indices and rules for
3068 large collation sequences to optimize multiple passes. This cache
3069 computation may overflow for large collation sequences and may cause a
3070 stack or buffer overflow. This is now fixed to use a slower algorithm
3071 which does not use a cache if there is an integer overflow.
3073 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
3074 rules for large collation sequences to optimize multiple passes and falls
3075 back to alloca if malloc fails, resulting in a possible stack overflow.
3076 The implementation now falls back to an uncached collation sequence lookup
3079 * CVE-2013-4788 The pointer guard used for pointer mangling was not
3080 initialized for static applications resulting in the security feature
3081 being disabled. The pointer guard is now correctly initialized to a
3082 random value for static applications. Existing static applications need
3083 to be recompiled to take advantage of the fix (bug 15754).
3085 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
3086 to the d_name member of struct dirent, or omit the terminating NUL
3087 character. (Bugzilla #14699).
3089 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
3090 aligned_alloc functions could allocate too few bytes or corrupt the
3091 heap when passed very large allocation size values (Bugzilla #15855,
3094 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
3095 for AF_INET6 has been fixed (Bugzilla #16072).
3097 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
3098 pap_CW, quz_PE, the_NP.
3100 * Substantially revised locales: gd_GB, ht_HT
3102 * The LC_ADDRESS field was updated to support country_car for almost all
3105 * ISO 1427 definitions were updated.
3107 * ISO 3166 definitions were updated.
3109 * The localedef utility now supports --big-endian and --little-endian
3110 command-line options to generate locales for a different system from that
3111 for which the C library was built.
3113 * Binary locale files now only depend on the endianness of the system for
3114 which they are generated and not on other properties of that system. As a
3115 consequence, binary files generated with new localedef may be incompatible
3116 with old versions of the GNU C Library, and binary files generated with
3117 old localedef may be incompatible with this version of the GNU C Library,
3118 in the following circumstances:
3120 + Locale files may be incompatible on m68k systems.
3122 + Locale archive files (but not separate files for individual locales) may
3123 be incompatible on systems where plain "char" is signed.
3125 * The configure option --disable-versioning has been removed. Builds with
3126 --disable-versioning had not worked for several years.
3128 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
3129 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
3131 * SystemTap probes for malloc have been introduced.
3133 * SystemTap probes for slow multiple precision fallback paths of
3134 transcendental functions have been introduced.
3136 * Support for powerpc64le has been added.
3138 * The soft-float powerpc port now supports e500 processors.
3140 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
3142 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
3143 set of header declarations that are enabled by default, even when other
3144 feature test macros or compiler options such as -std=c99 would otherwise
3145 disable some of those declarations.
3147 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
3148 conflict with POSIX. The libbsd-compat library (which was a dummy library
3149 that did nothing) has also been removed.
3151 * Preliminary documentation about Multi-Thread, Async-Signal and
3152 Async-Cancel Safety has been added.
3154 * Change of ABI data structures for s390 and s390x:
3155 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
3156 allow for future hardware extensions. All library functions that accept or
3157 return these structures were versioned in order to provide backwards
3158 compatibility with existing code. However, not all cases can be handled
3159 with interface versioning. If an application embeds either structure into
3160 another structure and that structure is passed to another compilation unit
3161 which expects the newer larger-sized object then any access to the new
3162 fields will result in undefined behaviour. Similarly any access to fields
3163 that were moved by the enlarging of the structures will cause undefined
3164 behaviour. To fix the undefined behaviour all objects that are part of the
3165 final in-memory image must agree on the size of structures, and this may
3166 require recompilation.
3170 * The following bugs are resolved with this release:
3172 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
3173 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
3174 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
3175 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
3176 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3177 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3178 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3179 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3180 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3181 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3182 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3183 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3184 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3187 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3188 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3189 Distributions can re-enable building and using pt_chown via the new configure
3190 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3191 considerable security risks and should only be used if the distribution
3192 understands and accepts the risks.
3194 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3197 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3198 fixed (Bugzilla #15330).
3200 * Add support for calling C++11 thread_local object destructors on thread
3201 and program exit. This needs compiler support for offloading C++11
3202 destructor calls to glibc.
3204 * Improved worst case performance of libm functions with double inputs and
3207 * Support for priority inherited mutexes in pthread condition variables on
3208 non-x86 architectures.
3210 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3212 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3214 * Optimized string functions for ARM. Implemented by Will Newton and
3217 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3219 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3222 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3223 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3225 * Added a benchmark framework to track performance of functions in glibc.
3227 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3228 It is based on draft TS 18661 and currently enabled as a GNU extension.
3230 * On Linux, the clock function now uses the clock_gettime system call
3231 for improved precision, rather than old times system call.
3233 * Added support for version-3 tz format files. This is needed when using
3234 the tz database release 2013e or later, and affects a few unusual cases --
3235 currently only TZ='America/Godthab' for time stamps after 2037.
3237 * Added new API functions pthread_getattr_default_np and
3238 pthread_setattr_default_np to get and set the default pthread
3239 attributes of a process.
3241 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3242 This may improve lock scaling of existing programs on TSX capable systems.
3243 When the --enable-lock-elision=yes parameter is specified at configure
3244 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3247 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3248 availability). Implemented by Ryan S. Arnold.
3250 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3254 * The following bugs are resolved with this release:
3256 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3257 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3258 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3259 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3260 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3261 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3262 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3263 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3264 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3265 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3266 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3267 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3268 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3269 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3270 14879, 14889, 14893, 14898, 14914.
3272 * Optimization of memcpy for MIPS.
3274 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3275 EMFILE has been fixed (Bugzilla #14889).
3277 * The add-on ports collection is now distributed in the "ports" subdirectory
3278 of the main GNU C Library distribution, rather than separately.
3280 * Port to ARM AArch64 contributed by Linaro.
3282 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3283 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3285 Implemented by Andreas Krebbel.
3287 * The new function secure_getenv allows secure access to the environment,
3288 returning NULL if running in a SUID/SGID process. This function replaces
3289 the internal function __secure_getenv.
3291 * SystemTap static probes have been added into the dynamic linker.
3292 Implemented by Gary Benson.
3294 * Optimizations of string functions strstr, strcasestr and memmem.
3295 Implemented by Maxim Kuvyrkov.
3297 * The minimum Linux kernel version that this version of the GNU C Library
3298 can be used with is 2.6.16.
3300 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3301 powerpc POWER7. Implemented by Will Schmidt.
3303 * New configure option --disable-nscd builds the C library such that it
3304 never attempts to contact the Name Service Caching Daemon (nscd).
3305 New configure option --disable-build-nscd avoids building nscd itself;
3306 this is the default if --disable-nscd is used.
3308 * Improved support for cross-compilation, including cross-testing and
3309 bootstrap builds without a previously built glibc.
3311 * Several testsuite tests are now able to test multiple IFUNC variants of an
3312 interface, rather than just testing the one that would be chooen by
3315 * New configure options --with-bugurl and --with-pkgversion, for
3316 distributors to use to embed their bug-reporting and package version
3317 information in --help and --version output.
3319 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3320 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3321 allows creation of chroots without the procfs mounted on /proc.
3323 * The `crypt' function now fails if passed salt bytes that violate the
3324 specification for those values. On Linux, the `crypt' function will
3325 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3326 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3327 when the mode is enabled.
3329 * The `clock_*' suite of functions (declared in <time.h>) is now available
3330 directly in the main C library. Previously it was necessary to link with
3331 -lrt to use these functions. This change has the effect that a
3332 single-threaded program that uses a function such as `clock_gettime' (and
3333 is not linked with -lrt) will no longer implicitly load the pthreads
3334 library at runtime and so will not suffer the overheads associated with
3335 multi-thread support in other code such as the C++ runtime library.
3337 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3342 * The following bugs are resolved with this release:
3344 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3345 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3346 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3347 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3348 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3349 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3350 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3351 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3352 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3353 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3354 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3355 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3356 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3357 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3358 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3359 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3360 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3361 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3362 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3363 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3364 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3365 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3368 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3369 configuring glibc with:
3370 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3371 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3372 Implemented by H.J. Lu.
3376 + define static_assert
3378 + do not declare gets
3380 + declare at_quick_exit and quick_exit also for ISO C11
3382 + aligned_alloc. NB: The code is deliberately allows the size parameter
3383 to not be a multiple of the alignment. This is a moronic requirement
3384 in the standard but it is only a requirement on the caller, not the
3387 + timespec_get added
3389 + uchar.h support added
3391 + CMPLX, CMPLXF, CMPLXL added
3393 Implemented by Ulrich Drepper.
3395 * Support for the IA-64 has been moved to ports.
3397 * Remove support for anything but ELF binary format
3399 * Checking versions of poll, ppoll added.
3400 Implemented by Ulrich Drepper.
3402 * More generic and 64-bit performance optimizations to math functions.
3403 Implemented by Ulrich Drepper.
3405 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3406 headers and functions available at compile time as they were before
3407 version 2.14. This option will be removed at some time in the future
3408 after the TI-RPC library becomes fully sufficient for the needs of
3409 existing applications.
3411 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3412 Note that glibc is not expected to work with any Linux kernel version
3415 * New header <sys/auxv.h> and function getauxval allowing easy access to
3416 the AT_* key-value pairs passed from the Linux kernel. The header also
3417 defines the HWCAP_* bits associated with the AT_HWCAP key.
3419 * New locales: mag_IN
3421 * New configure option --enable-systemtap builds SystemTap static probes
3422 into libc for setjmp and longjmp and into libpthread for various operations.
3423 So far the setjmp/longjmp probes and some of the libpthread probes are
3424 provided only for i*86 and x86_64.
3425 Implemented by Roland McGrath and Rayson Ho.
3427 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3429 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3432 * More optimized functions for SPARC. Implemented by David S. Miller.
3434 * Improved support for cross-compilation, especially bootstrap builds
3435 without a previously built glibc.
3437 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3438 Chris Metcalf from Tilera.
3440 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3441 now supported for ARM processors.
3443 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3444 as the name of the dynamic linker, to distinguish it from the
3445 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3447 * Support for CRIS has been removed from ports.
3449 * A new class of installed header has been documented for low-level
3450 platform-specific functionality. PowerPC added the first instance with a
3451 function to provide time base register access. Contributed by Tulio
3452 Magno Quites Machado Filho.
3454 * ix86 configurations now install header files that are consistent with
3455 what x86-64 configurations install. These same header files can be used
3456 for -m32, -m64, or -mx32 builds.
3457 Contributed by H.J. Lu.
3459 * Math library bug fixes. A thorough audit of all open math library bugs was
3460 conducted by Joseph Myers. Significant progress was made on many math
3461 library bugs resulting in more accurate exceptions and function results.
3462 Many thanks to all those that contributed including Andreas Jaeger for his
3463 patch review and work on the x87 trigonometric instruction issues.
3465 * Timezone data is no longer installed. Timezone-related binaries and scripts
3466 will continue to be installed. Users should obtain their timezone data from
3467 their distribution provider or from the tzdata package at
3468 <ftp://munnari.oz.au/pub/>.
3472 * The following bugs are resolved with this release:
3474 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3475 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3476 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3477 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3478 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3479 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3480 13484, 13506, 13515, 13523, 13524, 13538, 13540
3482 * New program pldd to list loaded object of a process
3483 Implemented by Ulrich Drepper.
3485 * Add nss_db support back to glibc. No more dependency on Berkeley db
3486 and support for initgroups lookups.
3487 Implemented by Ulrich Drepper.
3489 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3490 Contributed by HJ Lu.
3492 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3493 Contributed by HJ Lu.
3495 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3496 on x86-32 and x86-64.
3497 Contributed by Liubov Dmitrieva.
3499 * Optimized strchr and strrchr for SSE on x86-32.
3500 Contributed by Liubov Dmitrieva.
3502 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3503 for x86-64 and x86-32.
3504 Contributed by Liubov Dmitrieva.
3506 * New interfaces: scandirat, scandirat64
3507 Implemented by Ulrich Drepper.
3509 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3510 Implemented by Ulrich Drepper.
3512 * nscd now also caches the netgroup database.
3513 Implemented by Ulrich Drepper.
3515 * Integrate libm with gcc's -ffinite-math-only option.
3516 Implemented by Ulrich Drepper.
3518 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3519 to math functions. Implemented by Ulrich Drepper.
3521 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3522 Implemented by Ulrich Drepper.
3524 * New Linux interfaces: process_vm_readv, process_vm_writev
3526 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3527 Implemented by Ulrich Drepper.
3529 * Optimized nearbyint and strcasecmp for PPC.
3530 Implemented by Adhemerval Zanella.
3532 * New locales: bho_IN, unm_US, es_CU, ta_LK
3536 * The following bugs are resolved with this release:
3538 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3539 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3540 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3541 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3542 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3543 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3544 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3545 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3546 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3547 12795, 12811, 12813, 12814, 12841
3549 * The RPC implementation in libc is obsoleted. Old programs keep working
3550 but new programs cannot be linked with the routines in libc anymore.
3551 Programs in need of RPC functionality must be linked against TI-RPC.
3552 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3554 Visible changes of this change include (obviously) the inability to link
3555 programs using RPC functions without referencing the TI-RPC library and the
3556 removal of the RPC headers from the glibc headers.
3557 Implemented by Ulrich Drepper.
3559 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3560 syncfs, setns, sendmmsg
3562 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3563 yue_HK, lij_IT, mhr_RU
3565 * New encodings: CP770, CP771, CP772, CP773, CP774
3567 * New program sotruss to trace calls through PLTs
3568 Implemented by Ulrich Drepper.
3570 * The malloc hook implementation is marked deprecated and will be removed
3571 from the default implementation in the next version. The design never
3572 worked ever since the introduction of threads. Even programs which do
3573 not create threads themselves can use multiple threads created internally.
3577 * The following bugs are resolved with this release:
3579 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3580 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3581 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3582 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3585 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3587 * POWER7 optimizations: memset, memcmp, strncmp
3589 * New optimized string functions for x86-64: strnlen (SSE2),
3590 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3591 Implemented by Ulrich Drepper.
3595 * The following bugs are resolved with this release:
3597 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3598 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3599 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3600 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3601 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3602 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3603 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3604 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3606 * New interfaces: pthread_getname_np, pthread_setname_np
3608 * New Linux interface: recvmmsg
3610 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3612 * The dynamic linker now recognizes supported ABI versions from the
3613 EI_ABIVERSION field in the ELF header.
3614 Implemented by Ulrich Drepper.
3616 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3617 will not be used to fill in password fields in the passwd.byname replies.
3618 Instead it is used to synthesize the shadow.byname table, should it be
3619 missing. This is a useful mode in some installations involving Solaris.
3620 Implemented by Ulrich Drepper.
3622 * New locales: kok_IN, sq_MK, cv_RU
3626 * The following bugs are resolved with this release:
3628 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3629 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3630 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3631 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3632 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3633 10789, 10817, 10823, 10840, 10847
3635 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3637 Implemented by Ulrich Drepper.
3639 * Checking version of longjmp added that fails if an uninitialized stack
3640 frame would be created. Implemented by Ulrich Drepper.
3642 * STT_GNU_IFUNC is now supported in static executables.
3643 Implemented by H.J. Lu.
3645 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3647 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3648 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3649 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3650 Contributed by H.J. Lu.
3652 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3653 Implemented by Ulrich Drepper.
3655 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3657 Contributed by H.J. Lu.
3659 * Support for fma instruction in AVX on x86-64.
3660 Implemented by H.J. Lu and Ulrich Drepper.
3662 * AVX support in x86-64 auditing support in ld.so.
3663 Implemented by H.J. Lu.
3665 * STB_GNU_UNIQUE support added.
3666 Implemented by Ulrich Drepper.
3668 * Implement second fallback mode for optimized DNS lookup for even more
3669 broken environments. If two requests from the same port are not
3670 handled correctly close the socket and open a new one before sending
3671 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3672 can be used to select this mode right away, instead of rediscovering the
3673 necessity is every process again.
3674 Implemented by Ulrich Drepper.
3676 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3677 Implemented by Adam Tkac.
3679 * Optimized iconv conversions for S390x.
3680 Implemented by Andreas Krebbel.
3682 * Using condvars with PI mutexes is now more efficient due to kernel
3683 support for requeueing to PI futexes. NPTL support added for x86-64.
3684 Implemented by Ulrich Drepper.
3686 * New locales: ps_AF, my_MM
3690 * The following bugs are resolved with this release:
3692 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3693 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3694 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3695 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3696 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3698 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3699 Implemented by Ulrich Drepper.
3701 * New Linux interfaces: accept4, fallocate, fallocate64.
3702 Implemented by Ulrich Drepper.
3704 * Correct declarations of string function when used in C++ code. This
3705 could lead to compile errors for invalid C++ code.
3707 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3710 * New POSIX 2008 interface: psiginfo
3711 Implemented by Ulrich Drepper.
3713 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3714 Implemented by Ulrich Drepper.
3716 * Support for selecting between multiple function definitions at runtime
3717 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3719 * The libcrypt library can now use the hash function implementations in
3720 NSS. Implemented by Ulrich Drepper.
3722 * The malloc implementation can be compiled to be less memory efficient
3723 but higher performing in multi-threaded programs.
3724 Implemented by Ulrich Drepper.
3726 * New locales: nan_TW@latin, ks_IN
3728 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3729 Implemented by Ulrich Drepper.
3731 * Extended printf hook support. It is possible to use user-defined types
3732 and extend existing format specifiers.
3733 Implemented by Ulrich Drepper.
3735 * Handling for group shadow files has been added.
3736 Implemented by Ulrich Drepper.
3738 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3739 servers (the case, e.g., for some people using the built-in DNS
3740 server in ADSL modems/routers). There is a once-per-process timeout
3741 in case of a broken server. To avoid it, users can run nscd or put
3742 'options single-request' in /etc/resolv.conf.
3743 Implemented by Ulrich Drepper.
3747 * The following bugs are resolved with this release:
3749 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3750 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3751 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3752 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3753 6968, 6974, 6980, 6995, 7008, 7009, 7029
3755 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3756 up at the same time. Implemented by Ulrich Drepper.
3758 * TLS descriptors for LD and GD on x86 and x86-64.
3759 Implemented by Alexandre Oliva.
3761 * getaddrinfo now handles DCCP and UDPlite.
3762 Implemented by Ulrich Drepper.
3764 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3765 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3766 Implemented by Ulrich Drepper.
3768 * New implementation of memmem, strstr, and strcasestr which is O(n).
3769 Implemented by Eric Blake.
3771 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3773 * Implement "e" option for popen to open file descriptor with the
3774 close-on-exec flag set. Implemented by Ulrich Drepper.
3776 * Implement "b" mode for fmemopen. In this mode writes writes don't
3777 implicitly add a NUL byte and seeks from the end of the buffer really
3778 use the buffer end, not the string length as the basis.
3779 Implemented by Ulrich Drepper.
3781 * Many functions, exported and internal, now atomically set the close-on-exec
3782 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3784 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3786 Implemented by Pravin Satpute.
3788 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3792 * The following bugs are resolved with this release:
3794 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3795 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3796 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3797 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3798 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3799 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3800 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3802 * New locales: bo_CN, bo_IN, shs_CA.
3804 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3806 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3807 Implemented by Pravin Satpute.
3809 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3811 * nscd caches now all timeouts for DNS entries
3812 Implemented by Ulrich Drepper.
3814 * nscd is more efficient and wakes up less often.
3815 Implemented by Ulrich Drepper.
3817 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3818 vdprintf, and obstack_vprintf.
3819 Implemented by Jakub Jelinek.
3821 * Faster memset for x86-64.
3822 Implemented by Harsha Jagasia and H.J. Lu.
3824 * Faster memcpy on x86.
3825 Implemented by Ulrich Drepper.
3827 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3828 Implemented by Ulrich Drepper.
3830 * Faster sqrt and sqrtf implemention for some PPC variants.
3831 Implemented by Steven Munroe.
3835 * The following bugs are resolved with this release:
3837 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3838 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3839 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3840 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3841 5071, 5103, 5104, 5112, 5113, 5184, 5186
3843 * More checking functions: fread, fread_unlocked, open*, mq_open.
3844 Implemented by Jakub Jelinek and Ulrich Drepper.
3846 * Extend fortification to C++. Implemented by Jakub Jelinek.
3848 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3849 by not recognizing 'a' as a modifier when those specs are requested.
3850 Implemented by Jakub Jelinek.
3852 * PPC optimizations to math and string functions.
3853 Implemented by Steven Munroe.
3855 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3856 options to be passed. Implemented by Ulrich Drepper.
3858 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3860 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3861 Implemented by Ulrich Drepper.
3863 * Handle private futexes in the NPTL implementation.
3864 Implemented by Jakub Jelinek and Ulrich Drepper.
3866 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3867 Implemented by Roland McGrath and Ulrich Drepper.
3869 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3871 * SHA-256 and SHA-512 based password encryption.
3872 Implemented by Ulrich Drepper.
3874 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3875 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3878 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3879 Implemented by Ulrich Drepper.
3883 * The following bugs are resolved with this release:
3885 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3886 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3887 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3888 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3889 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3890 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3891 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3894 * New Linux interfaces: epoll_pwait, sched_getcpu.
3896 * New generic interfaces: strerror_l.
3898 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3902 * The following bugs are resolved with this release:
3904 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3905 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3906 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3907 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3908 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3909 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3910 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3911 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3912 3225, 3251, 3252, 3253, 3273
3914 * For Linux, the sorting of addresses returned by getaddrinfo now also
3915 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3916 Implemented by Ulrich Drepper.
3918 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3919 Implemented by Ulrich Drepper.
3921 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3923 * New iconv module for MIK. Contributed by Alexander Shopov.
3925 * For sites with broken group and/or passwd database, the auto-propagate
3926 option of nscd can prevent creating ID lookup entries from the results
3927 of a name lookup and vice versa. This usually is no problem but some
3928 site might have problems with the default behavior.
3929 Implemented by Ulrich Drepper.
3931 * Iterating over entire database in NIS can be slow. With the
3932 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3933 to trade time for memory. The entire database will be read at once.
3934 Implemented by Ulrich Drepper.
3936 * The interfaces introduced in RFC 3542 have been implemented by
3939 * Support for the new ELF hash table format was added by Ulrich Drepper.
3941 * Support for priority inheritance mutexes added by Jakub Jelinek and
3944 * Support for priority protected mutexes added by Jakub Jelinek.
3946 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3947 or_IN, csb_PL, fy_NL, sr_ME.
3951 * More overflow detection functions.
3953 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3954 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3956 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3957 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3958 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3959 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3960 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3961 by Masahide Washizawa.
3963 * It is now possible to install audit libraries for the dynamic linker using
3964 LD_AUDIT. Implemented by Ulrich Drepper.
3966 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3967 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3968 For a libc and libpthread that works well on Linux 2.4 kernels, we
3969 recommend using the stable 2.3 branch.
3971 * The new function open_wmemstream acts like open_memstream,
3972 but uses a wchar_t wide character buffer.
3974 * The new function ppoll is an improved version of poll, similar to pselect.
3976 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3977 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3978 renameat, unlinkat, linkat, symlinkat, readlinkat.
3980 * New Linux kernel system calls: unshare,
3981 inotify_init, inotify_add_watch, inotify_rm_watch.
3983 * The euidaccess function is now also known by the alias eaccess,
3984 for compatibility with some other systems.
3986 * Timezone data updated to 2006b version.
3990 * The following bugs are resolved with this release:
3992 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3993 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3994 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3995 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3996 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3997 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3999 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4001 * As of this release, GCC 4 can be used to compile the C Library.
4003 * Timezone data updated to 2005m version.
4007 * The following bugs are resolved with this release:
4009 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
4010 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
4011 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
4012 777, 787, 821, 822, 823, 825
4014 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
4018 * Support for RFC 3678. Real implementations exist only for Linux so far.
4019 Implemented by Ulrich Drepper.
4021 * nscd can now cache entries persistently. Expiring entries are reloaded.
4022 For speedups the cache can be shared in memory with client processes.
4023 Implemented by Ulrich Drepper.
4025 * nscd can now perform SELinux checks.
4026 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
4028 * getaddrinfo queries are now cached. Canonical name lookup is performed
4030 Implemented by Ulrich Drepper.
4032 * The nothrow function attribute is used when headers are used by gcc when
4033 compiling C code. This can avoid significant amounts of exception
4036 * The malloc functions perform more error checking and are stricter when
4037 it comes to reacting on errors. The default action is to terminate
4038 the process after showing an error message. Implemented by Ulrich Drepper.
4040 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
4041 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
4043 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
4044 namespaces using the new function dlmopen(). This feature is of course,
4045 like most other dynamic loading functionality, not available in statically
4046 linked applications. Implemented by Ulrich Drepper.
4048 * Low-overhead boundary checking variants of string and some stdio functions
4049 were added. These are to be used in conjunction with a gcc patch by
4050 Jakub Jelinek which adds calls to these functions if possible.
4051 Implemented by Jakub Jelinek and Ulrich Drepper.
4053 * Old code for several operating systems and machine architectures that
4054 have not been in working condition in a long time have been removed from
4055 the main source tree maintained by the GNU C Library's maintainers.
4056 These files are now reside in the separate `ports' source module
4057 that is usable as an add-on when building the library.
4061 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
4062 interrogate the dynamic linker, compatible with the Solaris interface.
4064 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
4065 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
4067 * getifaddrs now uses the netlink interface on Linux to get its information.
4068 Implemented by Thorsten Kukuk.
4070 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
4071 Implemented by Ulrich Drepper.
4073 * support for non-executable stacks on x86 has been added. Changes mostly
4076 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
4079 * getaddrinfo now performs destination address selection according to
4084 * Thread-safe interfaces for many functions that access locale data
4085 were added in version 2.3, but these features were omitted from NEWS.
4086 Many functions have variants with an `_l' suffix that take a `locale_t'
4087 object as a parameter rather than consulting the current locale.
4088 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
4089 create and maintain `locale_t' objects. Additionally, the new function
4090 `uselocale' sets "the current locale" (as used by functions not so
4091 parameterized) set for an individual thread. These features were added
4092 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
4094 * The functions getresuid, getresgid, setresuid, and setresgid, which
4095 have long been available on Linux, are now declared in <unistd.h>
4096 and are now also available on the Hurd.
4098 * ELF thread-local storage support (TLS) now works on x86-64.
4100 * The new dynamic string token $LIB is expanded in shared library names.
4101 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
4103 * Aldy Hernandez contributed complete software floating point support for
4104 PowerPC machines with no FPU.
4106 * fexecve is implemented on Linux.
4108 * The `btowc' function should work at least twice as fast due to
4109 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
4111 * With approriate thread add-ons cancelable functions are now implemented
4112 in libc.so as well. No need to call the function in libpthread. This
4113 change allowed to finally disable the incorrect and expensive handling
4114 of weak definition in ld.so.
4116 * Yet more PLT entries in libc.so have been removed. We finally arrived
4117 at the bare minimum. Startup times improved appropriately.
4119 * Support for the new Linux/x86 system call interface was added. The
4120 AT_SYSINFO auxiliary vector entry is recognized and handled.
4124 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
4127 * iconv (the program and the interface) now accepts empty names (excluding
4128 options like //TRANSLIT) to mean "use charset of current locale".
4130 * localedef can now transliterate characters in strings which are not in
4131 the provided charmap. The information from the input locale is used.
4133 * Prelinking support was added for ELF targets. This requires additional
4134 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
4136 * Read-only stdio streams now use mmap to speed up operation by eliminating
4137 copying and buffer underflows. To use add 'm' to the mode string of
4138 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
4140 * The malloc functions were completely rewritten by Wolfram Gloger based
4141 on Doug Lea's malloc-2.7.0.c.
4143 * Isamu Hasegawa contributed a completely new and POSIX-conformant
4144 implementation of regex.
4146 * Bruno Haible upgraded the iconv and locale implementation to support
4149 * Contents of the LC_* and LANG environment variables in the CEN style are
4150 not recognized anymore. It never was used. Change by Ulrich Drepper.
4152 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
4153 thread-local storage (TLS) ABI on some platforms.
4154 Changes by Ulrich Drepper. SH support by Kaz Kojima.
4156 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
4157 EUC-JISX0213, and TSCII.
4159 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
4160 BSD-compatible interface for getting all network interface addresses.
4161 Implementation for IPv4 by Roland McGrath.
4163 * Loading of locale data is faster due to the introduction of a locale
4164 archive. Implemented by Roland McGrath and Ulrich Drepper.
4166 * Startup times are significantly reduced by not using exported functions
4167 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
4170 * Steven Munroe contributed a port to PowerPC64/Linux.
4174 * The Hurd now uses the GNU libio implementation of stdio rather than the
4175 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
4177 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4178 with the same behavior as the Linux system call of the same name.
4182 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4183 128-bit long double format.
4185 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4186 IBM1161, and IBM1162 charsets.
4188 * Andreas Jaeger contributed a port to x86-64/Linux.
4190 * Peter Bruin contributed a port to PowerPC/Hurd.
4192 * libc functions using I/O streams now can handle wide-oriented streams
4195 * optimizations in the dynamic linker. Binaries created by recent binutils
4196 versions start up quicker due to reduced time spent on relocations.
4198 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4202 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4203 asinh, atanh, j0 for the 128-bit long double format.
4205 * Bruno Haible updated all the code handling Unicode in some form to
4206 support Unicode 3.1.
4208 * Speed of regex for single-byte locales is back to previous levels.
4209 Patch by Isamu Hasegawa.
4211 * Alpha, SPARC, and IA-64 now also using floating stacks.
4213 * Startup time of internationalized applications greatly improved through
4214 iconv cache. Use iconvconfig to generate the cache file.
4215 Contributed by Ulrich Drepper.
4217 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4218 severe performance problems. Patch by David Mosberger.
4220 * The Hurd port got a lot more functionality like AIO, various stdio
4221 extensions, etc. Mainly done by Roland McGrath.
4223 * mtrace can now lookup symbols in shared libraries.
4227 * Intel's IA-64 math library is largely integrated. It provides fast and
4228 accurate implementatations for most basic and standard math functions
4229 in float, double, and long double format.
4231 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4232 and asin for the 96-bit long double format and asin, log, tan for the
4233 128-bit long double format.
4235 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4236 The basic double functions exist today. Contributed by Abraham Ziv
4237 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4238 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4240 * An asynchronous name lookup library was added. The interface is designed
4241 after POSIX AIO. The proposal was circulated beforehand to get comments.
4242 No negative ones came in. Implemented by Ulrich Drepper.
4244 * Port to S390/64bit contributed by Martin Schwidefsky
4245 <schwidefsky@de.ibm.com>.
4247 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4248 of functions for Linux/IA-64.
4250 * The RPC code is now thread safe. Threads can now use the same service
4251 of different services at the same time. Patch by Eric Norum
4252 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4254 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4255 family of functions for Linux/S390.
4257 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4258 of functions for Linux/x86.
4260 * Port to Linux/CRIS contributed by Axis Communications.
4264 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4265 we can now check for namespace violations and missing declarations. The
4266 result is that almost all headers are now Unix-compliant (as defined in
4267 the upcoming XPG6). The negative side is that some programs might need
4268 corrections, too, if they depend on the incorrect form of the headers in
4269 previous versions which defined too many symbols and included too many
4272 * regex now handles multibyte character sets correctly.
4273 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4275 * iconv (the program) does now conform to the upcoming XPG6 and handles
4276 charmaps. Instead of the charset names the path of charmaps can be
4277 provided and the conversion happens based on this data.
4278 Contributed by Ulrich Drepper.
4280 * The locale program now provides detailed information about the installed
4285 only lists the names of the supported locales
4289 provides details such as country, language, and codeset name.
4290 Contributed by Ulrich Drepper.
4294 * The gencat program now parses the input file according to the charset
4295 selected by the LC_CTYPE category. This is important for stateful
4296 character sets. To make generating catalogs easier there is a way
4297 to overwrite the charset selected by the locale: before the first
4298 message or $ quote line the catalog can contain a line like
4300 $ codeset=ISO-8859-2
4302 to select the charset (ISO-8859-2 in this case).
4304 Implemented by Shinya Hanataka and Ulrich Drepper.
4306 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4307 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4308 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4310 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4311 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4313 * The LANGUAGE environment variable is now ignored unless the locale is
4314 changed from the default "C" locale.
4316 * The usual bug fixes.
4320 * Greg McGary added runtime support for bounds checking using gcc's
4321 new -fbounded-pointers option. ix86 target is complete. PowerPC
4324 * Thorsten Kukuk added secure mode support to nscd.
4326 * The Berkeley db libraries have been removed.
4328 Related, the nss_db module is now in a separate package since it
4329 obviously requires a database library being available.
4331 * Wide character I/O streams implemented by Ulrich Drepper.
4333 * Functions from the extended socket API added by Ulrich Drepper.
4335 * Functions feenableexcept and fedisableexcept to control the
4336 behaviour of individual exceptions have been added by Andreas Jaeger.
4338 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4340 * The resolver code has been updated from bind 8.2.3-T5B which supports
4341 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4344 This change could in some situations effect backward compatibility. Since
4345 now `_res' is a thread-local instead of a global variable, modifying it
4346 in one thread does not have any effect in other threads.
4348 The resolver library was also extended to allow IPv6 as the transport
4349 protocol for the requests. This work was done by Stig Venaas.
4351 * Compatibility code for K&R C compilers has been removed from the
4352 header files. A ISO C compiler is needed to use the library
4353 (conforming to either C89 or C99 standard).
4355 * Complete rewrite of the localedef program to support multibyte character
4356 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4357 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4358 Implemented by Ulrich Drepper.
4360 Bruno Haible significantly improved the generation and use of the data
4361 structures for the wide character tables.
4363 * Plural handling in gettext implemented by Ulrich Drepper.
4365 * The utmp daemon has been removed.
4367 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4369 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4372 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4374 * POSIX barriers implemented by Kaz Kylheku.
4376 * POSIX spawn function family implemented by Ulrich Drepper.
4378 * POSIX spinlocks are now available.
4380 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4382 * the configure option --enable-kernel=X.Y.Z allows to strip out
4383 compatibility for kernel versions before X.Y.Z. This is currently only
4384 implemented for Linux.
4386 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4387 field sin6_scope_id. This means that all programs using IPv6 should be
4388 recompiled. Don't expect binary compatibility with previous glibc
4391 * various conversion modules for IBM character sets contributed by
4394 * IA-64 port by Jes Sorensen and HJ Lu.
4408 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4410 * New xdr functions are added; some rpc functions are now 64bit clean.
4412 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4414 * Fixed known incompatibilities with glibc 2.0.
4416 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4418 * Optimized a number of functions (especially the ELF dynamic loader).
4420 * Update timezone data files.
4422 * lots of charmaps corrections
4424 * some new locale definitions and charmaps
4429 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4430 conform to POSIX member type requirements. Symbol versions have been
4431 adjusted as needed within the library, and for direct use by applications,
4432 but there is potential for problems if third-party libraries use
4433 struct timeval as part of their interface. This does not present
4434 a problem for X and other "essential" system libraries.
4436 * An additional locale model to support C++ Standard Library locale
4437 model and probably more was implemented by Ulrich Drepper.
4439 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4442 * Miles Bader provided the `argp' function family to support hierachical
4443 command line argument parsing, layered on top of getopt.
4445 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4447 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4450 * scanf recognizes the %a and %A format for scanning floating point numbers.
4452 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4453 information and interfaces for the available integer types.
4455 * about 130 new math functions were added to implement the ISO C9x math
4458 * the new header <complex.h> contains definitions of the complex math
4459 functions from ISO C 9X.
4461 * the new header <tgmath.h> defines generic macros to use complex or
4462 real valued functions.
4464 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4466 * Andreas Jaeger provided a test suite for the math library.
4468 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4470 * Ulrich Drepper added character set conversion functions (iconv).
4472 * Optimized string functions have been added.
4474 * The localedata addon is now part of glibc.
4476 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4478 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4479 daemon for NSS (nscd).
4481 Missing a better place here are some numbers on improvements. Under
4482 Linux 2.1.125 un-tar-ing the kernel sources takes
4486 using local files 12.19s 6.88s 22.91s
4488 using NIS 13.92s 8.91s 26.34s
4490 using NIS & nscd 10.37s 7.34s 25.30s
4492 using NIS+ 27.57s 30.37s 640.46s
4494 using NIS+ & nscd 10.25s 7.83s 26.51s
4496 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4498 Keep in mind that non-namelookup related operations dominate above times.
4499 It was just a common complain that using NIS+ unpacking the kernel is
4502 [1] The old nscd implementation is not available anymore since it was
4503 distributed with glibc up to version 2.0.98 and thus is now replaced.
4505 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4507 * Mark Kettenis provided a Hesiod NSS module.
4509 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4510 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4512 * rcmd can now handle netgroups (Dick Streefland).
4514 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4517 * Support for the IPv6 protocol has been added to the socket API, as per the
4518 latest draft standards.
4520 * Support for Linux 2.2 has been added.
4522 * Interface changes relative to the latest 2.0.x release:
4523 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4524 addseverity NEW: Unix98
4525 alphasort64 NEW: LFS
4526 argp_err_exit_status NEW: argp, GNU ext
4527 argp_error NEW: argp, GNU ext
4528 argp_failure NEW: argp, GNU ext
4529 argp_help NEW: argp, GNU ext
4530 argp_parse NEW: argp, GNU ext
4531 argp_program_bug_address NEW: argp, GNU ext
4532 argp_program_version NEW: argp, GNU ext
4533 argp_program_version_hook NEW: argp, GNU ext
4534 argp_state_help NEW: argp, GNU ext
4535 argp_usage NEW: argp, GNU ext
4536 authdes_create NEW: Secure RPC
4537 authdes_getucred NEW: Secure RPC
4538 authdes_pk_create NEW: Secure RPC
4539 backtrace NEW: GNU ext.
4540 backtrace_symbols NEW: GNU ext.
4541 backtrace_symbols_fd NEW: GNU ext.
4543 cacosf NEW: ISO C 9x
4544 cacosh NEW: ISO C 9x
4545 cacoshf NEW: ISO C 9x
4546 cacoshl NEW: ISO C 9x
4547 cacosl NEW: ISO C 9x
4554 casinf NEW: ISO C 9x
4555 casinh NEW: ISO C 9x
4556 casinhf NEW: ISO C 9x
4557 casinhl NEW: ISO C 9x
4558 casinl NEW: ISO C 9x
4560 catanf NEW: ISO C 9x
4561 catanh NEW: ISO C 9x
4562 catanhf NEW: ISO C 9x
4563 catanhl NEW: ISO C 9x
4564 catanl NEW: ISO C 9x
4565 cbc_crypt NEW: Secure RPC
4569 ccoshf NEW: ISO C 9x
4570 ccoshl NEW: ISO C 9x
4576 cimagf NEW: ISO C 9x
4577 cimagl NEW: ISO C 9x
4578 clearerr_locked REMOVED
4579 clntunix_create NEW: sunrpc ext
4581 clog10 NEW: ISO C 9x
4582 clog10f NEW: ISO C 9x
4583 clog10l NEW: ISO C 9x
4593 cprojf NEW: ISO C 9x
4594 cprojl NEW: ISO C 9x
4596 crealf NEW: ISO C 9x
4597 creall NEW: ISO C 9x
4602 csinhf NEW: ISO C 9x
4603 csinhl NEW: ISO C 9x
4606 csqrtf NEW: ISO C 9x
4607 csqrtl NEW: ISO C 9x
4611 ctanhf NEW: ISO C 9x
4612 ctanhl NEW: ISO C 9x
4614 des_setparity NEW: Secure RPC
4615 ecb_crypt NEW: Secure RPC
4616 endutxent NEW: Unix98
4618 exp10f NEW: ISO C 9x
4619 exp10l NEW: ISO C 9x
4623 fattach NEW: STREAMS
4624 fdetach NEW: STREAMS
4628 feclearexcept NEW: ISO C 9x
4629 fegetenv NEW: ISO C 9x
4630 fegetexceptflag NEW: ISO C 9x
4631 fegetround NEW: ISO C 9x
4632 feholdexcept NEW: ISO C 9x
4634 feraiseexcept NEW: ISO C 9x
4635 ferror_locked REMOVED
4636 fesetenv NEW: ISO C 9x
4637 fesetexceptflag NEW: ISO C 9x
4638 fesetround NEW: ISO C 9x
4639 fetestexcept NEW: ISO C 9x
4640 feupdateenv NEW: ISO C 9x
4641 fflush_locked REMOVED
4645 fgets_unlocked NEW: GNU ext.
4646 fileno_locked REMOVED
4658 fputc_locked REMOVED
4659 fputs_unlocked NEW: GNU ext.
4660 fread_unlocked NEW: GNU ext.
4665 fstatvfs NEW: Unix98
4669 ftruncate64 NEW: LFS
4671 fwrite_unlocked NEW: GNU ext.
4672 gai_strerror NEW: IPv6
4676 getchar_locked REMOVED
4678 getdate_err NEW: Unix98
4679 getdate_r NEW: GNU ext.
4681 getnameinfo NEW: IPv6
4682 getnetname NEW: Secure RPC
4683 getpmsg NEW: STREAMS
4684 getpt NEW: Unix98 PTY
4685 getrlimit64 NEW: LFS
4686 getutxent NEW: Unix98
4687 getutxid NEW: Unix98
4688 getutxline NEW: Unix98
4689 glob64 NEW: GNU ext.
4690 globfree64 NEW: GNU ext.
4691 gnu_get_libc_release NEW: GNU ext.
4692 gnu_get_libc_version NEW: GNU ext.
4693 grantpt NEW: Unix98 PTY
4694 host2netname NEW: Secure RPC
4696 iconv_close NEW: iconv
4697 iconv_open NEW: iconv
4698 if_freenameindex NEW: IPv6
4699 if_indextoname NEW: IPv6
4700 if_nameindex NEW: IPv6
4701 if_nametoindex NEW: IPv6
4702 in6addr_any NEW: IPv6
4703 in6addr_loopback NEW: IPv6
4704 inet6_isipv4mapped NEW: IPv6
4705 isastream NEW: STREAMS
4706 iswblank NEW: GNU ext.
4707 key_decryptsession NEW: Secure RPC
4708 key_decryptsession_pk NEW: Secure RPC
4709 key_encryptsession NEW: Secure RPC
4710 key_encryptsession_pk NEW: Secure RPC
4711 key_gendes NEW: Secure RPC
4712 key_get_conv NEW: Secure RPC
4713 key_secretkey_is_set NEW: Secure RPC
4714 key_setnet NEW: Secure RPC
4715 key_setsecret NEW: Secure RPC
4716 llrint NEW: ISO C 9x
4717 llrintf NEW: ISO C 9x
4718 llrintl NEW: ISO C 9x
4719 llround NEW: ISO C 9x
4720 llroundf NEW: ISO C 9x
4721 llroundl NEW: ISO C 9x
4726 lrintf NEW: ISO C 9x
4727 lrintl NEW: ISO C 9x
4728 lround NEW: ISO C 9x
4729 lroundf NEW: ISO C 9x
4730 lroundl NEW: ISO C 9x
4732 makecontext NEW: Unix98
4733 mempcpy NEW: GNU ext.
4736 modify_ldt NEW: kernel
4740 nearbyint NEW: ISO C 9x
4741 nearbyintf NEW: ISO C 9x
4742 nearbyintl NEW: ISO C 9x
4743 netname2host NEW: Secure RPC
4744 netname2user NEW: Secure RPC
4745 nexttoward NEW: ISO C 9x
4746 nexttowardf NEW: ISO C 9x
4747 nexttowardl NEW: ISO C 9x
4751 passwd2des NEW: Secure RPC
4753 pow10f NEW: GNU ext.
4754 pow10l NEW: GNU ext.
4757 printf_size NEW: GNU ext.
4758 printf_size_info NEW: GNU ext.
4759 profil_counter REMOVED
4760 pthread_mutexattr_getkind_np REPLACED
4761 pthread_mutexattr_setkind_np REPLACED
4762 ptsname NEW: Unix98 PTY
4763 ptsname_r NEW: Unix98 PTY
4765 putchar_locked REMOVED
4766 putgrent NEW: GNU ext.
4768 putpmsg NEW: STREAMS
4769 pututxline NEW: Unix98
4773 readdir64_r NEW: LFS
4774 remquo NEW: ISO C 9x
4775 remquof NEW: ISO C 9x
4776 remquol NEW: ISO C 9x
4778 roundf NEW: ISO C 9x
4779 roundl NEW: ISO C 9x
4781 scalbln NEW: ISO C 9x
4782 scalblnf NEW: ISO C 9x
4783 scalblnl NEW: ISO C 9x
4785 sendfile NEW: kernel
4786 setcontext NEW: Unix98
4787 setrlimit64 NEW: LFS
4788 setutxent NEW: Unix98
4790 sigignore NEW: Unix98
4791 sigqueue NEW: POSIX.1b
4792 sigrelse NEW: Unix98
4793 sigset NEW: POSIX.1b
4794 sigtimedwait NEW: POSIX.1b
4795 sigwaitinfo NEW: POSIX.1b
4796 sincos NEW: GNU ext.
4797 sincosf NEW: GNU ext.
4798 sincosl NEW: GNU ext.
4802 strcasestr NEW: GNU ext.
4803 strtoimax NEW: ISO C 9x
4804 strtoumax NEW: ISO C 9x
4805 strverscmp NEW: GNU ext.
4806 svcauthdes_stats NEW: Secure RPC
4807 svcunix_create NEW: sunrpc etx
4808 svcunixfd_create NEW: sunrpc ext
4809 swapcontext NEW: Unix98
4810 tcgetsid NEW: Unix98 PTY
4811 tdestroy NEW: GNU ext.
4812 tgamma NEW: ISO C 9x
4813 tgammaf NEW: ISO C 9x
4814 tgammal NEW: ISO C 9x
4818 truncf NEW: ISO C 9x
4819 truncl NEW: ISO C 9x
4821 unlockpt NEW: Unix98 PTY
4822 updwtmpx NEW: Unix98
4823 user2netname NEW: Secure RPC
4824 utmpxname NEW: Unix98
4825 versionsort NEW: GNU ext.
4826 versionsort64 NEW: GNU ext.
4828 wcscasecmp NEW: GNU ext.
4829 wcsncasecmp NEW: GNU ext.
4830 wcsnlen NEW: GNU ext.
4831 wcstoimax NEW: ISO C 9x
4832 wcstoll NEW: ISO C 9x
4833 wcstoull NEW: ISO C 9x
4834 wcstoumax NEW: ISO C 9x
4836 wordexp NEW: POSIX.2
4837 wordfree NEW: POSIX.2
4838 write_profiling REMOVED
4839 xdecrypt NEW: Secure RPC
4840 xdr_authdes_cred NEW: Secure RPC
4841 xdr_authdes_verf NEW: Secure RPC
4842 xdr_cryptkeyarg NEW: Secure RPC
4843 xdr_cryptkeyarg2 NEW: Secure RPC
4844 xdr_cryptkeyres NEW: Secure RPC
4845 xdr_getcredres NEW: Secure RPC
4846 xdr_key_netstarg NEW: Secure RPC
4847 xdr_key_netstres NEW: Secure RPC
4848 xdr_keybuf NEW: Secure RPC
4849 xdr_keystatus NEW: Secure RPC
4850 xdr_netnamestr NEW: Secure RPC
4851 xdr_sizeof NEW: Secure RPC
4852 xdr_unixcred NEW: sunrpc ext
4853 xencrypt NEW: Secure RPC
4854 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4865 * inet_ntoa is thread-safe
4867 * updwtmp is moved from libutil to libc
4869 * rewrite of cbrt function
4871 * update of timezone data
4885 * add atoll function
4887 * fix complex problems in Berkeley DB code
4889 * fix math functions
4893 * fixed lots of header problems (especially Linux/GNU specific)
4895 * dynamic loader preserves all registers
4897 * Roland McGrath provided support for handling of auxiliary objects in
4898 the ELF dynamic loader.
4900 * support for parallel builds is improved
4904 * GNU extensions are no longer declared by default. To enable them you
4905 must define the macro `_GNU_SOURCE' in your program or compile with
4908 * The library has changed from using GNU ld symbol aliases to using weak
4909 symbols where available. The ELF object file format supports weak
4910 symbols; GNU ld also supports weak symbols in the a.out format. (There
4911 is also now support for other GNU ld extensions in ELF. Use the
4912 `--with-elf' option to configure to indicate you have ELF, and
4913 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4914 of many files which contained only symbol aliases, reducing the size of
4915 the source and the compiled library; many other files were renamed to
4916 less cryptic names previously occupied by the symbol alias files.
4917 There is a new header file <elf.h> for programs which operate on
4918 files in the ELF format.
4920 * Converted to Autoconf version 2, so `configure' has more options.
4921 Run `configure --help' to see the details.
4923 * The library can now be configured to build profiling, highly-optimized
4924 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4925 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4926 `configure' enable building these extra libraries. The shared library is
4927 built by default when using both ELF and GNU ld. When shared libraries
4928 are enabled, the new library `-ldl' is available for arbitrary run-time
4929 loading of shared objects; its interface is defined in <dlfcn.h>. The
4930 new header file <link.h> gives access to the internals of the run-time
4931 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4932 application of same name on other systems and it provides information
4933 about dynamically linked binaries.
4935 * The C library now provides the run-time support code for profiling
4936 executables compiled with `-pg'. Programs can control the profiling code
4937 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4938 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4939 the support for this file format was contributed by David Mosberger-Tang.
4941 * The math code has been replaced with a math library based on fdlibm from
4942 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4943 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4944 functions. The math functions now reside in a separate library, so
4945 programs using them will need to use `-lm' their linking commands.
4947 * John C. Bowman contributed optimized ix87 assembler inline functions.
4949 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4950 mechanism similar to that found in Solaris 2. This is now used for the
4951 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4952 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4953 file controls what services are used for each individual database. This
4954 works by loading shared libraries with names specified in `nsswitch.conf',
4955 so service modules can be changed or added at any time without even
4956 relinking any program. Currently there are the file, db, and NIS based
4957 NSS services available.
4959 * The new functions `strtoq' and `strtouq' parse integer values from
4960 strings, like `strtol' and `strtoul', but they return `long long int' and
4961 `unsigned long long int' values, respectively (64-bit quantities).
4963 * The new functions `strtof' and `strtold' parse floating-point values from
4964 strings, like `strtod', but they return `float' and `long double' values,
4965 respectively (on some machines `double' and `long double' are the same).
4967 * Ulrich Drepper has contributed new implementations of the floating-point
4968 printing and reading code used in the `printf' family of functions and
4969 `strtod', `strtof', and `strtold'. These new functions are perfectly
4970 accurate, and much faster than the old ones.
4972 * The implementation of the POSIX locale model was completely rewritten by
4973 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4974 to compile the POSIX locale definition.
4976 * The former dummy implementations of the strcoll and strxfrm function are
4977 now replaced by fully functional code contributed by Ulrich Drepper. The
4978 collation information comes from the POSIX locale definitions.
4980 * The new header <langinfo.h> defines an interface for accessing
4981 various locale-dependent data (using the locale chosen with `setlocale').
4983 * Ulrich Drepper has contributed a new suite of functions for operation on
4984 wide-character and multibyte-character strings, in <wchar.h>;
4985 and classification and case conversion of wide characters, in <wctype.h>.
4986 These new functions are conforming to the ISO C, Amendement 1 specification.
4988 * There is now a second implementation of the standard I/O library available.
4989 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4990 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4992 * You can now use positional parameter specifications in format strings
4993 for the `printf' and `scanf' families of functions. For example,
4994 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4995 ``Number 6, Mr Jones''. This is mainly useful when providing different
4996 format strings for different languages, whose grammars may dictate
4997 different orderings of the values being printed. To support this
4998 feature, the interface for `register_printf_handler' has changed; see
4999 the header file <printf.h> for details.
5001 * The `printf' and `scanf' families of functions now understand a new
5002 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
5003 to group numbers as indicated by the locale; for `scanf' and friends, this
5004 says to accept as valid only a number with all the proper grouping
5005 separators in the right places. In the default "C" locale, numbers are
5006 not grouped; but locales for specific countries will define the usual
5007 conventions (i.e. separate thousands with `,' in the US locale).
5009 * The pgrp functions have been regularized, slightly incompatibly but much
5010 less confusingly. The core functions are now `getpgid' and `setpgid',
5011 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
5012 argument) and BSD `setpgrp' (identical to `setpgid') functions are
5013 provided for compatibility. There is no longer an incompatible `getpgrp'
5014 with an argument declared under _BSD_SOURCE; no BSD code uses it.
5016 * The new header file <fts.h> and suite of functions simplify programs that
5017 operate on directory trees. This code comes from 4.4 BSD.
5019 * The resolver code has been updated from the BIND 4.9.5-P1 release.
5020 Parts of the code were heavily modified by Ulrich Drepper to fit in the
5021 NSS scheme used in glibc.
5023 * There is a new malloc debugging hook `__memalign_hook'.
5025 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
5026 `unsigned int' in <sys/types.h>. These are for compatibility only and
5027 their use is discouraged.
5029 * The `-lmcheck' library to enable standard malloc debugging hooks is now
5030 done differently, so that it works even without GNU ld.
5032 * New function `euidaccess' checks allowed access to a file like `access',
5033 but using the effective IDs instead of the real IDs.
5035 * The time zone data files have been updated for the latest and greatest
5036 local time conventions of the countries of the world.
5038 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
5041 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
5042 for formatting floating-point numbers. They are provided only for
5043 compatibility; new programs should use `sprintf' instead. There are
5044 also equivalent function for the `long double' floating-point type and
5045 all functions also exist in a reentrant form.
5047 * The new auxiliary library `-lutil' from 4.4 BSD contains various
5048 functions for maintaining the login-record files (primarily of use to
5049 system programs such as `login'), and convenient functions for
5050 allocating and initializing a pseudo-terminal (pty) device.
5052 * Ulrich Drepper has contributed new support for System V style
5053 shared memory and IPC on systems that support it.
5055 * Ulrich Drepper has contributed several miscellaneous new functions found
5056 in System V: The `hsearch' family of functions provide an effective
5057 implementation of hash tables; `a64l' and `l64a' provide a very simple
5058 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
5061 * Ulrich Drepper has contributed new reentrant counterparts for the
5062 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
5064 * Ulrich Drepper has contributed new, highly-optimized versions of several
5065 string functions for the i486/Pentium family of processors.
5067 * Ulrich Drepper has updated the Linux-specific code, based largely
5068 on work done in Hongjiu Lu's version of GNU libc for Linux.
5069 The GNU library now supports Linux versions 2.0.10 and later,
5070 using the ELF object file format (i[3456]86-*-linux).
5072 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
5074 * David Mosberger-Tang and Richard Henderson have ported the C library
5075 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
5076 dynamic linking support for ELF/Alpha.
5078 * Richard Henderson contributed several Alpha optimized assembler function
5079 for arithmetic and string handling.
5081 * Ulrich Drepper has contributed a new set of message catalog functions to
5082 support multiple languages using the <libintl.h> interface, for use with
5083 his new package GNU gettext. Translation volunteers have contributed
5084 catalogs of the library's messages in Spanish, German, and Korean.
5086 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
5087 program and the `catgets' function for reading the catalog files it
5088 creates. (The <libintl.h> interface is preferred; we include the
5089 <nl_types.h> interface using `catgets' only for source compatibility with
5090 programs already written to use it.)
5092 * New header file <values.h> gives SVID-compatible names for <limits.h>
5095 * Various new macros, declarations, and small header files for compatibility
5098 * New function `group_member' is a convenient way to check if a process has
5099 a given effective group ID.
5101 * When using GCC 2.7 and later, the socket functions are now declared in a
5102 special way so that passing an argument of type `struct sockaddr_in *',
5103 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
5104 `struct sockaddr *' type, does not generate a type-clash warning.
5106 * New function `error' declared in header file <error.h> is a convenient
5107 function for printing error messages and optionally exiting; this is the
5108 canonical function used in GNU programs. The new functions `err', `warn',
5109 and friends in header file <err.h> are the canonical 4.4 BSD interface for
5110 doing the same thing.
5112 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
5113 POSIX.2 `glob' function to do ~ and {...} expansion.
5115 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
5116 `clearenv' which is used in POSIX.9 is also available.
5118 * New function `getsid' returns session ID number on systems that support it.
5120 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
5121 files <db.h> and <mpool.h> provide a rich set of functions for several
5122 types of simple databases stored in memory and in files, and <ndbm.h> is
5123 an old `ndbm'-compatible interface using the `db' functions. Link with
5124 `-ldb' to get these functions.
5126 * New macro `strdupa' copies a string like `strdup', but uses local stack
5127 space from `alloca' instead of dynamic heap space from `malloc'.
5129 * New function `strnlen' is like `strlen' but searches only a given maximum
5130 number of characters for the null terminator. `stpncpy', `strndup' and
5131 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
5134 * New function `statfs' in header <sys/statfs.h>.
5136 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
5137 provide convenient functions for operating on blocks of null-terminated
5140 * A new suite of functions in <utmp.h> handle all the details of reading
5141 and writing the utmp file.
5143 * An implementation of the NIS/YP(tm) based NSS service was contributed by
5146 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
5147 completely POSIX compliant and also implemented the extended functionality
5148 to handle alternate digit representation and alternate era date formats.
5150 * Ulrich Drepper provided an implementation of the `strptime' function
5151 defined in XPG4.2 which transforms a string into a `struct tm' value.
5153 * Paul Eggert provided the tzselect shell script as part of the timezone
5154 code. The shell script makes it easy to select the correct timezone
5157 * The implementation of the malloc family of functions is completely replaced
5158 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
5159 The implementation uses the mmap function (if available) and it is
5160 optimized for the use in multi threaded programs.
5162 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
5163 functions. This new functionality is usable by specifying a special
5164 salt string and it is compatible with implementation on *BSD systems.
5166 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
5167 `getsubopt' to handle second level command line options, `bsd_signal'
5168 to access BSD style `signal' functionality, the obsolete `regexp' style
5171 * the `lchown' function is available on system which support this
5174 * The implementation of the shadow password handling function was contributed
5177 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5179 * POSIX.1g support was added. The <sys/select.h> header is available,
5180 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5181 implementation of `getaddrinfo'.
5185 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5187 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5188 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5190 * The new function `daemon' from 4.4 BSD is useful for server programs that
5191 want to put themselves in the background.
5193 * Joel Sherrill has contributed support for several standalone boards that
5194 run without an operating system.
5196 * `printf', `scanf' and friends now accept a `q' type modifier for long
5197 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5199 * All of the code taken from BSD (notably most of the math and networking
5200 routines) has been updated from the BSD 4.4-Lite release.
5202 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5204 * The new functions `getdomainname' and `setdomainname' fetch or change the
5205 YP/NIS domain name. These are system calls which exist on systems which
5208 * The time zone data files have been updated for the latest international
5211 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5212 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5216 * The C library now includes support for Sun RPC, from Sun's free
5217 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5218 are included. (There is still no support for YP.)
5220 * Tom Quinn has contributed a port of the C library to SGI machines running
5221 Irix 4 (mips-sgi-irix4).
5223 * The new `lockf' function is a simplified interface to the locking
5224 facilities of `fcntl', included for compatibility.
5226 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5228 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5230 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5233 * The error message printed by `assert' for a failed assertion now includes
5234 the name of the program (if using GNU ld) and the name of the calling
5235 function (with versions of GCC that support this).
5237 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5239 * The library now includes the <sys/mman.h> header file and memory
5240 management functions `mmap', `munmap', `mprotect', `msync', and
5241 `madvise', on systems that support those facilities.
5243 * The interface for `mcheck' has changed slightly: the function called to
5244 abort the program when an allocation inconsistency is detected now takes
5245 an argument that indicates the type of failure. The new function
5246 `mprobe' lets you request a consistency check for a particular block at
5247 any time (checks are normally done only when you call `free' or `realloc'
5250 * It is now possible to easily cross-compile the C library, building on one
5251 system a library to run on another machine and/or operating system. All
5252 you need to do is set the variable `HOST_CC' in `configparms' to the
5253 native compiler for programs to run on the machine you are building on (a
5254 few generator programs are used on Unix systems); set `CC' to the
5257 * The new function `fexecve' (only implemented on the GNU system) executes
5258 a program file given a file descriptor already open on the file.
5262 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5263 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5265 * You can set the variable `libprefix' in `configparms' to specify a prefix
5266 to be prepended to installed library files; this makes it easy to install
5267 the GNU C library to be linked as `-lgnuc' or whatever.
5269 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5270 copies a limited number of characters from a string, and returns the
5271 address of the last character written.
5273 * You no longer need to check for whether the installed `stddef.h' is
5274 compatible with the GNU C library. configure now checks for you.
5276 * You can now define a per-stream `fileno' function to convert the
5277 stream's cookie into an integral file descriptor.
5279 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5280 allocates zero bytes of storage, and returns a unique pointer which
5281 you can pass to `realloc' or `free'. The behavior is undefined if
5282 you dereference this pointer.
5284 * The C library now runs on Sony NEWS m68k machines running either
5285 NewsOS 3 or NewsOS 4.
5287 * The new `syscall' function is a system-dependent primitive function
5288 for invoking system calls. It has the canonical behavior on Unix
5289 systems, including unreliable return values for some calls (such as
5290 `pipe', `fork' and `getppid').
5292 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5293 to `EAGAIN', which is the preferred name. On systems whose kernels
5294 use two distinct codes, the C library now translates EWOULDBLOCK to
5295 EAGAIN in every system call function.
5299 * The GNU C Library Reference Manual is now distributed with the library.
5300 `make dvi' will produce a DVI file of the printed manual.
5301 `make info' will produce Info files that you can read on line using C-h i
5302 in Emacs or the `info' program.
5303 Please send comments on the manual to bug-glibc-manual@gnu.org.
5305 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5307 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5309 * Jason Merrill has contributed a port to the Sequent Symmetry running
5310 Dynix version 3 (i386-sequent-dynix).
5312 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5313 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5315 * New function `memory_warnings' lets you arrange to get warnings when
5316 malloc is running out of memory to allocate, like Emacs gives you.
5318 * The C library now contains the relocating allocator used in Emacs 19 for
5319 its editing buffers. This allocator (ralloc) minimizes allocation
5320 overhead and fragmentation by moving allocated regions around whenever it
5321 needs to. You always refer to a ralloc'd region with a "handle" (a
5322 pointer to a pointer--an object of type `void **').
5324 * There is a new `printf' format: `%m' gives you the string corresponding
5325 to the error code in `errno'.
5327 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5328 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5329 pass, the `a' modifier says to fill in a `char **' you pass with a
5332 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5333 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5334 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5336 * `mkstemp' is a traditional Unix function to atomically create and open a
5337 uniquely-named temporary file.
5341 * The standard location for the file that says what the local timezone is
5342 has changed again. It is now `/usr/local/etc/localtime' (or more
5343 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5345 * The distribution no longer contains any files with names longer than 14
5348 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5349 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5351 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5355 * The configuration process has changed quite a bit. The `configure'
5356 script is now used just like the configuration scripts for other GNU
5357 packages. The `sysdeps' directory hierarchy is much rearranged.
5358 The file `INSTALL' explains the new scheme in detail.
5360 * The header files no longer need to be processed into ANSI C and
5361 traditional C versions. There is just one set of files to install, and
5362 it will work with ANSI or old C compilers (including `gcc -traditional').
5364 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5365 MIPS DECStation running Ultrix 4.
5367 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5368 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5369 made itself into a shared library.
5371 * Yet further improved support for the i386, running 4.3 BSD-like systems
5372 (such as Mach 3 with the Unix single-server), or System V.
5374 * New function `strncasecmp' to do case-insensitive string comparison
5375 with limited length.
5377 * New function `strsep' is a reentrant alternative to `strtok'.
5379 * New functions `scandir' and `alphasort' for searching directories.
5381 * New function `setenv' is a better interface to `putenv'.
5383 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5384 function for traversing a directory tree.
5386 * The GNU obstack package is now also part of the C library.
5387 The new function `open_obstack_stream' creates a stdio stream that
5388 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5389 formatted output directly to an obstack.
5391 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5392 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5394 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5396 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5397 things to your strings.
5399 * There are some new test programs: test-fseek, testmb, and testrand.
5401 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5402 These ports are not finished, but are a good starting place for really
5403 supporting those systems.
5405 * `/etc/localtime' is now the standard location for the file that says what
5406 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5407 This follows the general principle that `/etc' is the place for all local
5408 configuration files.
5410 * The C library header files now use `extern "C"' when used by the C++
5411 compiler, so the C library should now work with C++ code.
5413 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5414 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5417 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5418 function which uses the `merge sort' algorithm, and is said to be
5419 significantly faster than the old GNU `qsort' function. Merge sort is
5420 now the standard `qsort' function. The new algorithm can require a lot
5421 of temporary storage; so, the old sorting function is called when the
5422 required storage is not available.
5424 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5425 provides the Unix `crypt' function, plus some other entry points.
5427 * The code and header files taken from 4.4 BSD have been updated with the
5428 latest files released from Berkeley.
5430 ----------------------------------------------------------------------
5431 Copyright information:
5433 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5435 Permission is granted to anyone to make or distribute verbatim copies
5436 of this document as received, in any medium, provided that the
5437 copyright notice and this permission notice are preserved,
5438 thus giving the recipient permission to redistribute in turn.
5440 Permission is granted to distribute modified versions
5441 of this document, or of portions of it,
5442 under the above conditions, provided also that they
5443 carry prominent notices stating who last changed them.
5446 version-control: never