1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2019 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * The getcpu wrapper function has been added, which returns the currently
13 used CPU and NUMA node. This function is Linux-specific.
15 * A new convenience target has been added for distribution maintainers
16 to build and install all locales as directories with files. The new
17 target is run by issuing the following command in your build tree:
18 'make localedata/install-locale-files', with an optional DESTDIR
19 to set the install root if you wish to install into a non-default
22 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
24 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
25 for _GNU_SOURCE, to match BSD environments.
27 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
28 indicates that it will abort the transaction prior to entering the kernel
29 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
30 suspended, and this caused some undefined side-effects issues by aborting
31 transactions manually. Glibc avoided it by abort transactions manually on
32 each syscall, but it lead to performance issues on newer kernels where the
33 HTM state is saved and restore lazily (the state being saved even when the
34 process actually does not use HTM).
36 * The functions posix_spawn_file_actions_addchdir_np and
37 posix_spawn_file_actions_addfchdir_np have been added, enabling
38 posix_spawn and posix_spawnp to run the new process in a different
39 directory. These functions are GNU extensions. The function
40 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
43 * The popen and system do not run atfork handlers anymore (BZ#17490).
44 Although it is a possible POSIX violation, the POSIX rationale in
45 pthread_atfork documentation regarding atfork handlers is to handle
46 inconsistent mutex state after a fork call in a multi-threaded process.
47 In both popen and system there is no direct access to user-defined mutexes.
49 * Support for the C-SKY ABIV2 running on Linux has been added. This port
50 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
52 - C-SKY ABIV2 soft-float little-endian
53 - C-SKY ABIV2 hard-float little-endian
55 Deprecated and removed features, and other changes affecting compatibility:
57 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
58 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
60 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
61 in <sys/procfs.h>, has been corrected to match the type actually used by
62 the Linux kernel. This affects the size and layout of that structure on
63 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
65 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
66 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
67 defined in <sys/procfs.h>, has been corrected to match the type actually
68 used by the Linux kernel. This affects the size and layout of those
71 Changes to build and runtime requirements:
73 * Python 3.4 or later is required to build the GNU C Library.
75 * On most architectures, GCC 5 or later is required to build the GNU C
76 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
78 Older GCC versions and non-GNU compilers are still supported when
79 compiling programs that use the GNU C Library.
81 Security related changes:
83 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
84 denial of service due to resource exhaustion when processing getaddrinfo
85 calls with crafted host names. Reported by Guido Vranken.
87 The following bugs are resolved with this release:
89 [The release manager will add the list generated by
90 scripts/list-fixed-bugs.py just before the release.]
97 * The localization data for ISO 14651 is updated to match the 2016
98 Edition 4 release of the standard, this matches data provided by
99 Unicode 9.0.0. This update introduces significant improvements to the
100 collation of Unicode characters. This release deviates slightly from
101 the standard in that the collation element ordering for lowercase and
102 uppercase LATIN script characters is adjusted to ensure that regular
103 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
104 is not matched by [a-z]. With the update many locales have been
105 updated to take advantage of the new collation information. The new
106 collation information has increased the size of the compiled locale
107 archive or binary locales.
109 * The GNU C Library can now be compiled with support for Intel CET, AKA
110 Intel Control-flow Enforcement Technology. When the library is built
111 with --enable-cet, the resulting glibc is protected with indirect
112 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
113 compatible with all existing executables and shared libraries. This
114 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
115 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
116 capable of multi-byte NOPs, like x86-64 processors as well as Intel
117 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
118 x86_64 and x32 on non-CET processors. --enable-cet has been tested
119 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
122 * The GNU C Library now has correct support for ABSOLUTE symbols
123 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
124 relocated incorrectly or in some cases discarded. The GNU linker can
125 make use of the newer semantics, but it must communicate it to the
126 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
127 field) to indicate such support is required.
129 * Unicode 11.0.0 Support: Character encoding, character type info, and
130 transliteration tables are all updated to Unicode 11.0.0, using
131 generator scripts contributed by Mike FABIAN (Red Hat).
133 * <math.h> functions that round their results to a narrower type are added
134 from TS 18661-1:2014 and TS 18661-3:2015:
136 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
139 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
142 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
145 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
148 * Two grammatical forms of month names are now supported for the following
149 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
150 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
151 support two grammatical forms in abbreviated month names: Catalan, Greek,
154 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
155 include the support for two grammatical forms of month names.
157 * Building and running on GNU/Hurd systems now works without out-of-tree
160 * The renameat2 function has been added, a variant of the renameat function
161 which has a flags argument. If the flags are zero, the renameat2 function
162 acts like renameat. If the flag is not zero and there is no kernel
163 support for renameat2, the function will fail with an errno value of
164 EINVAL. This is different from the existing gnulib function renameatu,
165 which performs a plain rename operation in case of a RENAME_NOREPLACE
166 flags and a non-existing destination (and therefore has a race condition
167 that can clobber the destination inadvertently).
169 * The statx function has been added, a variant of the fstatat64
170 function with an additional flags argument. If there is no direct
171 kernel support for statx, glibc provides basic stat support based on
172 the fstatat64 function.
174 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
175 library if installed. libidn2 version 2.0.5 or later is recommended. If
176 libidn2 is not available, internationalized domain names are not encoded
177 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
178 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
179 with an encoding error.) Flags which used to change the IDN encoding and
180 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
181 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
182 deprecated. They no longer have any effect.
184 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
185 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
186 range of ELF gABI expressions including such constructs as
187 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
188 have been further restricted, and where in the past a dynamic string
189 token sequence may have been interpreted as a literal string it will
190 now cause a load failure. These load failures were always considered
191 unspecified behaviour from the perspective of the dynamic loader, and
192 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
193 results in a load failure now.
195 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
196 implementation includes all the standard functions provided by
199 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
200 thrd_detach, thrd_exit, and thrd_join for thread management.
202 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
203 mtx_destroy for mutual exclusion.
205 - call_once for function call synchronization.
207 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
208 cnd_wait for conditional variables.
210 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
212 Application developers must link against libpthread to use ISO C threads.
214 Deprecated and removed features, and other changes affecting compatibility:
216 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
217 installed. Software that was using either header should be updated to
218 use standard <stdio.h> interfaces instead.
220 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
221 This was never required by the C standard, and the macros just expanded
222 to call alternative names for the same functions. If you hoped getc and
223 putc would provide performance improvements over fgetc and fputc, instead
224 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
225 necessary, flockfile and funlockfile.
227 * All stdio functions now treat end-of-file as a sticky condition. If you
228 read from a file until EOF, and then the file is enlarged by another
229 process, you must call clearerr or another function with the same effect
230 (e.g. fseek, rewind) before you can read the additional data. This
231 corrects a longstanding C99 conformance bug. It is most likely to affect
232 programs that use stdio to read interactive input from a terminal.
235 * The macros 'major', 'minor', and 'makedev' are now only available from
236 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
237 headers that happen to include <sys/types.h>. These macros are rarely
238 used, not part of POSIX nor XSI, and their names frequently collide with
239 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
242 <sys/sysmacros.h> is a GNU extension. Portable programs that require
243 these macros should first include <sys/types.h>, and then include
244 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
246 * The tilegx*-*-linux-gnu configurations are no longer supported.
248 * The obsolete function ustat is no longer available to newly linked
249 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
250 function has been deprecated in favor of fstatfs and statfs.
252 * The obsolete function nfsservctl is no longer available to newly linked
253 binaries. This function was specific to systems using the Linux kernel
254 and could not usefully be used with the GNU C Library on systems with
255 version 3.1 or later of the Linux kernel.
257 * The obsolete function name llseek is no longer available to newly linked
258 binaries. This function was specific to systems using the Linux kernel
259 and was not declared in a header. Programs should use the lseek64 name
260 for this function instead.
262 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
263 getaddrinfo and getnameinfo functions have been deprecated. The behavior
264 previously selected by them is now always enabled.
266 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
267 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
268 restriction (rejecting '_' in host names, among other things) has been
269 removed, for increased compatibility with non-IDN name resolution.
271 * The fcntl function now have a Long File Support variant named fcntl64. It
272 is added to fix some Linux Open File Description (OFD) locks usage on non
273 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
274 fcntl and LFS support is handled transparently. Also for Linux, the OFD
275 locks act as a cancellation entrypoint.
277 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
278 ecb_crypt, and des_setparity are no longer available to newly linked
279 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
280 longer installed. These functions encrypted and decrypted data with the
281 DES block cipher, which is no longer considered secure. Software that
282 still uses these functions should switch to a modern cryptography library,
285 * Reflecting the removal of the encrypt and setkey functions above, the
286 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
287 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
290 * The obsolete function fcrypt is no longer available to newly linked
291 binaries. It was just another name for the standard function crypt,
292 and it has not appeared in any header file in many years.
294 * We have tentative plans to hand off maintenance of the passphrase-hashing
295 library, libcrypt, to a separate development project that will, we hope,
296 keep up better with new passphrase-hashing algorithms. We will continue
297 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
298 'crypt_r' should not need to change at all; however, distributions will
299 need to install <crypt.h> and libcrypt from a separate project.
301 In this release, if the configure option --disable-crypt is used, glibc
302 will not install <crypt.h> or libcrypt, making room for the separate
303 project's versions of these files. The plan is to make this the default
304 behavior in a future release.
306 Changes to build and runtime requirements:
308 GNU make 4.0 or later is now required to build glibc.
310 Security related changes:
312 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
313 been fixed by removing the glibc-internal IDNA implementation and using
314 the system-provided libidn2 library instead. Originally reported by Hanno
315 Böck and Christian Weisgerber.
317 CVE-2017-18269: An SSE2-based memmove implementation for the i386
318 architecture could corrupt memory. Reported by Max Horn.
320 CVE-2018-11236: Very long pathname arguments to realpath function could
321 result in an integer overflow and buffer overflow. Reported by Alexey
324 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
325 architecture could write beyond the target buffer, resulting in a buffer
326 overflow. Reported by Andreas Schwab.
328 The following bugs are resolved with this release:
330 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
331 [6889] manual: 'PWD' mentioned but not specified
332 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
333 when __WORDSIZE != 64
334 [13762] regex: re_search etc. should return -2 on memory exhaustion
335 [13888] build: /tmp usage during testing
336 [13932] math: dbl-64 pow unexpectedly slow for some inputs
337 [14092] nptl: Support C11 threads
338 [14095] localedata: Review / update collation data from Unicode / ISO
340 [14508] libc: -Wformat warnings
341 [14553] libc: Namespace pollution loff_t in sys/types.h
342 [14890] libc: Make NT_PRFPREG canonical.
343 [15105] libc: Extra PLT references with -Os
344 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
346 [16335] manual: Feature test macro documentation incomplete and out of
348 [16552] libc: Unify umount implementations in terms of umount2
349 [17082] libc: htons et al.: statement-expressions prevent use on global
350 scope with -O1 and higher
351 [17343] libc: Signed integer overflow in /stdlib/random_r.c
352 [17438] localedata: pt_BR: wrong d_fmt delimiter
353 [17662] libc: please implement binding for the new renameat2 syscall
354 [17721] libc: __restrict defined as /* Ignore */ even in c11
355 [17979] libc: inconsistency between uchar.h and stdint.h
356 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
357 [18023] libc: extend_alloca is broken (questionable pointer comparison,
358 horrible machine code)
359 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
361 [18471] libc: llseek should be a compat symbol
362 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
364 [18991] nss: nss_files skips large entry in database
365 [19239] libc: Including stdlib.h ends up with macros major and minor being
367 [19463] libc: linknamespace failures when compiled with -Os
368 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
369 [19527] locale: Normalized charset name not recognized by setlocale
370 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
371 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
373 [19728] network: out of bounds stack read in libidn function
374 idna_to_ascii_4i (CVE-2016-6261)
375 [19729] network: out of bounds heap read on invalid utf-8 inputs in
376 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
377 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
379 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
380 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
381 [20419] dynamic-link: files with large allocated notes crash in
383 [20530] libc: bswap_16 should use __builtin_bswap16() when available
384 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
385 [20980] manual: CFLAGS environment variable replaces vital options
386 [21163] regex: Assertion failure in pop_fail_stack when executing a
387 malformed regexp (CVE-2015-8985)
388 [21234] manual: use of CFLAGS makes glibc detect no optimization
389 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
390 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
391 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
392 [21508] locale: intl/tst-gettext failure with latest msgfmt
393 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
394 [21812] network: getifaddrs() returns entries with ifa_name == NULL
395 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
397 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
399 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
400 [22247] network: Integer overflow in the decode_digit function in
401 puny_decode.c in libidn (CVE-2017-14062)
402 [22342] nscd: NSCD not properly caching netgroup
403 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
404 [22550] localedata: es_ES locale (and other es_* locales): collation
405 should treat ñ as a primary different character, sync the collation
406 for Spanish with CLDR
407 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
408 by gcc configured with --enable-default-pie
409 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
410 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
411 crossing 2GB threshold (CVE-2017-18269)
412 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
414 [22735] time: Misleading typo in time.h source comment regarding
416 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
417 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
418 `__assert_perror_fail ()' to `__assert_fail_base ()'
419 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
421 [22786] libc: Stack buffer overflow in realpath() if input size is close
422 to SSIZE_MAX (CVE-2018-11236)
423 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
424 through an absolute DT_NEEDED path
425 [22792] build: tcb-offsets.h dependency dropped
426 [22797] libc: pkey_get() uses non-reserved name of argument
427 [22807] libc: PTRACE_* constants missing for powerpc
428 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
429 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
430 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
431 [22848] localedata: ca_ES: update date definitions from CLDR
432 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
433 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
434 [22896] localedata: Update locale data for an_ES
435 [22902] math: float128 test failures with GCC 8
436 [22918] libc: multiple common of `__nss_shadow_database'
437 [22919] libc: sparc32: backtrace yields infinite backtrace with
439 [22926] libc: FTBFS on powerpcspe
440 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
442 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
443 [22947] libc: FAIL: misc/tst-preadvwritev2
444 [22963] localedata: cs_CZ: Add alternative month names
445 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
446 [22996] localedata: change LC_PAPER to en_US in es_BO locale
447 [22998] dynamic-link: execstack tests are disabled when SELinux is
449 [23005] network: Crash in __res_context_send after memory allocation
451 [23007] math: strtod cannot handle -nan
452 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
453 [23036] regex: regex equivalence class regression
454 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
455 [23069] libc: sigaction broken on riscv64-linux-gnu
456 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
457 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
458 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
459 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
460 and libc build with -Os)
461 [23140] localedata: More languages need two forms of month names
462 [23145] libc: _init/_fini aren't marked as hidden
463 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
464 [23171] math: C++ iseqsig for long double converts arguments to double
465 [23178] nscd: sudo will fail when it is run in concurrent with commands
466 that changes /etc/passwd
467 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
469 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
470 [23208] localedata: New locale - Lower Sorbian (dsb)
471 [23233] regex: Memory leak in build_charclass_op function in file
473 [23236] stdio: Harden function pointers in _IO_str_fields
474 [23250] nptl: Offset of __private_ss differs from GCC
475 [23253] math: tgamma test suite failures on i686 with -march=x86-64
476 -mtune=generic -mfpmath=sse
477 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
479 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
480 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
482 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
483 [23277] math: nan function should not have const attribute
484 [23279] math: scanf and strtod wrong for some hex floating-point
485 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
487 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
488 [23303] build: undefined reference to symbol
489 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
490 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
492 [23313] stdio: libio vtables validation and standard file object
494 [23329] libc: The __libc_freeres infrastructure is not properly run across
496 [23349] libc: Various glibc headers no longer compatible with
498 [23351] malloc: Remove unused code related to heap dumps and malloc
500 [23363] stdio: stdio-common/tst-printf.c has non-free license
501 [23396] regex: Regex equivalence regression in single-byte locales
502 [23422] localedata: oc_FR: More updates of locale data
503 [23442] build: New warning with GCC 8
504 [23448] libc: Out of bounds access in IBM-1390 converter
505 [23456] libc: Wrong index_cpu_LZCNT
506 [23458] build: tst-get-cpu-features-static isn't added to tests
507 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
509 [23467] dynamic-link: x86/CET: A property note parser bug
516 * The GNU C Library can now be compiled with support for building static
517 PIE executables (See --enable-static-pie in INSTALL). These static PIE
518 executables are like static executables but can be loaded at any address
519 and provide additional security hardening benefits at the cost of some
520 memory and performance. When the library is built with --enable-static-pie
521 the resulting libc.a is usable with GCC 8 and above to create static PIE
522 executables using the GCC option '-static-pie'. This feature is currently
523 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
524 aarch64 with binutils 2.30 or later.
526 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
527 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
530 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
532 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
534 * In order to support faster and safer process termination the malloc API
535 family of functions will no longer print a failure address and stack
536 backtrace after detecting heap corruption. The goal is to minimize the
537 amount of work done after corruption is detected and to avoid potential
538 security issues in continued process execution. Reducing shutdown time
539 leads to lower overall process restart latency, so there is benefit both
540 from a security and performance perspective.
542 * The abort function terminates the process immediately, without flushing
543 stdio streams. Previous glibc versions used to flush streams, resulting
544 in deadlocks and further data corruption. This change also affects
545 process aborts as the result of assertion failures.
547 * On platforms where long double has the IEEE binary128 format (aarch64,
548 alpha, mips64, riscv, s390 and sparc), the math library now implements
549 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
550 These are the same interfaces added in version 2.26 for some platforms where
551 this format is supported but is not the format of long double.
553 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
554 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
555 implements interfaces for that type, as defined by ISO/IEC TS
556 18661-3:2015. These are corresponding interfaces to those supported for
559 * The math library now implements interfaces for the _Float32, _Float64 and
560 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
561 corresponding interfaces to those supported for _Float128.
563 * glibc now implements the memfd_create and mlock2 functions on Linux.
565 * Support for memory protection keys was added. The <sys/mman.h> header now
566 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
569 * The copy_file_range function was added.
571 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
573 * The ldconfig utility now processes `include' directives using the C/POSIX
574 collation ordering. Previous glibc versions used locale-specific
575 ordering, the change might break systems that relied on that.
577 * Support for two grammatical forms of month names has been added.
578 In a call to strftime, the "%B" and "%b" format specifiers will now
579 produce the grammatical form required when the month is used as part
580 of a complete date. New "%OB" and "%Ob" specifiers produce the form
581 required when the month is named by itself. For instance, in Greek
582 and in many Slavic and Baltic languages, "%B" will produce the month
583 in genitive case, and "%OB" will produce the month in nominative case.
585 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
586 are all valid and will all accept any known form of month
587 name---standalone or complete, abbreviated or full. In a call to
588 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
589 the strings used by "%B" and "%b", respectively. New query
590 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
591 used by "%OB" and "%Ob", respectively.
593 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
594 define the strings for %OB and %Ob, respectively; these have the
595 same syntax as "mon" and "abmon". These arrays are optional; if they
596 are not provided then they have the same content as "mon" and "abmon",
599 These features are provided for locales which define "alt_mon" and/or
600 "ab_alt_mon" in their locale source data. This release includes such
601 alternative month name data for the following languages: Belarusian,
602 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
604 This feature is currently a GNU extension, but it is expected to
605 be added to the next revision of POSIX, and it is also already
606 available on some BSD-derived operating systems.
608 This feature will cause existing statically compiled applications
609 to fail to load locales and fall back to the builtin C/POSIX locales.
610 See notes below for other changes affecting compatibility.
612 * Support for the RISC-V ISA running on Linux has been added. This port
613 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
614 for the following ISA and ABI pairs:
620 Deprecated and removed features, and other changes affecting compatibility:
622 * Statically compiled applications attempting to load locales compiled for the
623 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
624 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
625 support for two grammatical forms of the month names, also extends the locale
626 data binary format. Static applications needing locale support must be
627 recompiled to match the runtime and data they are deployed with. In some
628 distributions there is an upgrade window where dynamically linked applications
629 may use a new library but the old locale data and also fall back to the
630 builtin C/POSIX locales; restarting the application process is sufficient to
633 * Support for statically linked applications which call dlopen is deprecated
634 and will be removed in a future version of glibc. Applications which call
635 dlopen need to be linked dynamically instead.
637 * Support for old programs which use internal stdio data structures and
638 functions is deprecated. This includes programs which use the C++ streams
639 provided by libstdc++ in GCC 2.95. Programs which use the internal
640 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
641 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
642 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
643 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
644 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
645 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
646 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
647 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
648 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
649 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
650 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
651 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
652 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
653 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
654 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
655 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
656 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
657 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
658 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
659 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
660 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
661 future version of glibc. Unlike other symbol removals, these old
662 applications will not be supported using compatibility symbols.
664 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
665 defined by <sys/ptrace.h>.
667 * libm no longer supports SVID error handling (calling a user-provided
668 matherr function on error) or the _LIB_VERSION variable to control error
669 handling. (SVID error handling and the _LIB_VERSION variable still work
670 for binaries linked against older versions of the GNU C Library.) The
671 libieee.a library is no longer provided. math.h no longer defines struct
672 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
673 TLOSS, PLOSS and HUGE.
675 * The libm functions pow10, pow10f and pow10l are no longer supported for
676 new programs. Programs should use the standard names exp10, exp10f and
677 exp10l for these functions instead.
679 * The mcontext_t type is no longer the same as struct sigcontext. On
680 platforms where it was previously the same, this changes the C++ name
681 mangling for interfaces involving this type.
683 * The add-ons mechanism for building additional packages at the same time as
684 glibc has been removed. The --enable-add-ons configure option is now
687 * The --without-fp configure option is now ignored. Whether hardware
688 floating-point instructions are used is now configured based on whether
689 the compiler used at configure time (without any options implied by a
690 --with-cpu= configure option) uses such instructions.
692 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
693 the specified string can be parsed as a domain name.
695 * In the malloc_info output, the <heap> element may contain another <aspace>
696 element, "subheaps", which contains the number of sub-heaps.
698 * The libresolv function p_secstodate is no longer supported for new
701 * The tilepro-*-linux-gnu configuration is no longer supported.
703 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
704 and will be removed in a future release. Software that is still using
705 either header should be updated to use standard <stdio.h> interfaces
708 libio.h was originally the header for a set of supported GNU extensions,
709 but they have not been maintained as such in many years, they are now
710 standing in the way of improvements to stdio, and we don't think there are
711 any remaining external users. _G_config.h was never intended for public
712 use, but predates the bits convention.
714 Changes to build and runtime requirements:
716 * bison version 2.7 or later is required to generate code in the 'intl'
719 Security related changes:
721 CVE-2009-5064: The ldd script would sometimes run the program under
722 examination directly, without preventing code execution through the
723 dynamic linker. (The glibc project disputes that this is a security
724 vulnerability; only trusted binaries must be examined using the ldd
727 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
728 suffered from a one-byte overflow during ~ operator processing (either
729 on the stack or the heap, depending on the length of the user name).
730 Reported by Tim Rühsen.
732 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
733 would sometimes fail to free memory allocated during ~ operator
734 processing, leading to a memory leak and, potentially, to a denial
737 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
738 without GLOB_NOESCAPE, could write past the end of a buffer while
739 unescaping user names. Reported by Tim Rühsen.
741 CVE-2017-17426: The malloc function, when called with an object size near
742 the value SIZE_MAX, would return a pointer to a buffer which is too small,
743 instead of NULL. This was a regression introduced with the new malloc
744 thread cache in glibc 2.26. Reported by Iain Buclaw.
746 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
747 to the allocation of too much memory. (This is not a security bug per se,
748 it is mentioned here only because of the CVE assignment.) Reported by
751 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
752 of the number of search path components. (This is not a security
753 vulnerability per se because no trust boundary is crossed if the fix for
754 CVE-2017-1000366 has been applied, but it is mentioned here only because
755 of the CVE assignment.) Reported by Qualys.
757 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
758 for AT_SECURE or SUID binaries could be used to load libraries from the
761 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
762 succeeds without returning an absolute path due to unexpected behaviour
763 of the Linux kernel getcwd syscall. Reported by halfdog.
765 CVE-2018-6485: The posix_memalign and memalign functions, when called with
766 an object size near the value of SIZE_MAX, would return a pointer to a
767 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
769 CVE-2018-6551: The malloc function, when called with an object size near
770 the value of SIZE_MAX, would return a pointer to a buffer which is too
771 small, instead of NULL.
773 The following bugs are resolved with this release:
775 [866] glob: glob should match dangling symlinks
776 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
777 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
779 [5997] math: Very slow execution of sinf function
780 [10580] localedata: hr_HR: updated locale
781 [10871] locale: 'mon' array should contain both nominative and genitive
783 [12349] localedata: eu_ES: incorrect thousands separator
784 [13605] localedata: shn_MM: new Shan locale
785 [13805] localedata: ru_RU: currency should use ',' as radix point
786 [13953] localedata: km_KH: locale update
787 [13994] localedata: mjw_IN: new locale
788 [14121] build: make writes .mo files in po directory
789 [14333] libc: Fix the race between atexit() and exit()
790 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
791 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
793 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
794 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
795 width Latin characters
796 [15332] localedata: es_CU: locale update
797 [15436] stdio: Don't close or flush stdio streams on abort
798 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
800 [16148] localedata: ca_ES: incorrect thousands separator
801 [16750] dynamic-link: ldd should not try to execute the binaries
803 [16777] localedata: pl_PL: incorrect thousands separator in locale
804 [16905] localedata: hanzi: new collation
805 [17563] localedata: cmn_TW: add hanzi collation
806 [17750] localedata: wrong collation order of diacritics in most locales
807 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
809 [17956] build: Build fails on missing definitions from header file
810 nss/nss.h when Mozilla NSS is used for cryptography
811 [18203] libc: realpath() does not handle unreachable paths correctly
812 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
813 [18812] localedata: kab_DZ: new Kabyle Algeria locale
814 [18822] libc: Internal functions are called via PLT
815 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
816 [19170] libc: __gmon_start__ defined in hppa in crtn.S
817 [19574] libc: glibc should support building static PIE binaries
818 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
820 [19971] glob: glob: Do not skip entries with zero d_ino values
821 [19982] localedata: fr.po: spelling mistake for error code EXDEV
822 [20008] localedata: km_KH: convert to translit_neutral
823 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
824 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
825 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
826 [20482] localedata: de_CH: abbreviated weekdays should be two letters
827 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
829 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
830 they're set, wrong errors returned by gaih_inet when lookup functions
832 [20756] localedata: [PATCH] Use Unicode wise thousands separator
833 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
835 [20952] localedata: yuw_PG: new locale
836 [21084] localedata: charmaps/IBM858: new codepage
837 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
838 [21242] libc: assert gives pedantic warning in old gcc versions
839 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
840 __regcall calling convention
841 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
842 [21326] libc: C99 functions are not declared for C++11 and later if
843 _GNU_SOURCE is not predefined by g++
844 [21457] libc: sys/ucontext.h namespace
845 [21530] libc: tmpfile() should be implemented using O_TMPFILE
846 [21660] math: GCC fails to compile a formula with tgmath.h
847 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
848 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
849 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
850 [21684] math: tgmath.h handling of complex integers
851 [21685] math: tgmath.h handling of bit-fields
852 [21686] math: tgmath.h handling of __int128
853 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
855 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
857 [21750] localedata: column width of characters incompatible with classical
859 [21754] malloc: malloc: Perform as little work as possible after heap
860 consistency check failures
861 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
862 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
863 [21791] string: Unused XXX_chk_XXX functions in libc.a
864 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
866 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
868 [21845] localedata: Added new Locale bho_NP
869 [21853] localedata: Fix abday Which looks same as day in zh_SG
870 [21854] localedata: Added New Locale en_SC
871 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
872 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
873 _dl_runtime_resolve_avx_slow
874 [21885] network: getaddrinfo: gethosts does not release resolver context
875 on memory allocation failure
876 [21899] libc: XPG4.2 sigaction namespace
877 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
879 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
881 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
882 errno value, causing getaddrinfo to fail
883 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
884 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
885 instead of EAI_NODATA
886 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
887 constant PTRACE_SEIZE_DEVEL
888 [21930] math: C-only gcc builtins used in <math.h> isinf
889 [21932] network: Unpaired __resolv_context_get in generic get*_r
891 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
892 [21944] libc: sigval namespace
893 [21951] localedata: Update hanzi collation by stroke
894 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
895 sysdeps/x86_64/fpu/e_expf.S
896 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
897 allocation in syscall loops)
898 [21959] localedata: Fix Country name for xh_ZA
899 [21960] localedata: Fix abmon for bem_ZM
900 [21966] math: AVX2 mathvec functions use FMA without checking
901 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
902 [21971] localedata: Added New Locale for mfe_MU
903 [21972] libc: assert macro requires operator== (int) for its argument type
904 [21973] math: [sparc] libm missing sqrtl compat symbol
905 [21974] libc: Remove __bb_init_func and __bb_exit_func
906 [21982] string: stratcliff.c: error: assuming signed overflow does not
908 [21986] stdio: __guess_grouping is called incorrectly
909 [21987] math: [sparc32] wrong bits/long-double.h installed
910 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
912 [22022] localedata: Missing country_name for mni_IN
913 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
915 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
917 [22026] locale: iconv_open: heap overflow on gconv_init failure
918 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
919 [22035] math: [m68k] bits/math-inline.h macro namespace
920 [22038] localedata: Fix abbreviated weeks and months for Somali
921 [22044] localedata: Remove redundant data for Limburgish Language
922 [22050] malloc: Linking with -lmcheck does not hook
923 __malloc_initialize_hook correctly
924 [22051] libc: zero terminator in the middle of glibc's .eh_frame
925 [22052] malloc: malloc failed to compile with GCC 7 and -O3
926 [22070] localedata: charmaps/UTF-8: wcwidth for
927 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
928 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
929 Jungseong and Jongseong) should be 0
930 [22078] nss: nss_files performance issue in multi mode
931 [22082] math: bits/math-finite.h exp10 condition
932 [22086] libc: pcprofiledump incorrect cross-endian condition
933 [22093] dynamic-link: ld.so no longer searches in .../x86_64
934 [22095] network: Name server address allocation memory leak in resolv.conf
936 [22096] network: __resolv_conf_attach can incorrectly free passed conf
938 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
939 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
941 [22111] malloc: malloc: per thread cache is not returned when thread exits
942 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
943 [22134] libc: [linux] implement fexecve with execveat
944 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
946 [22145] libc: ttyname() gives up too early in the face of namespaces
947 [22146] math: C++ build issue with float128 on x86_64
948 [22153] nptl: nptl: save error code before process termination
949 [22156] libc: [hppa,ia64,microblaze] Executable stack default
950 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
951 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
952 [22165] libc: [hppa] Text relocations in libc.so
953 [22180] libc: destructor registered via __cxa_atexit is called twice
954 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
955 symlinks") cause "make" segfaults
956 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
958 [22207] libc: FAIL: stdlib/test-atexit-race
959 [22225] math: nearbyint arithmetic moved before feholdexcept
960 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
961 [22235] math: iscanonical in C++ and float128
962 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
963 the svid compat wrapper
964 [22244] math: ynf and yn are wrong without the svid compat wrapper
965 [22273] libc: Improper assert in Linux posix_spawn implementation
966 [22284] libc: -pg -pie doesn't work
967 [22292] locale: localedef exits with error 4 when it should be error 1
968 [22294] locale: Allow "" for int_currency_symbol definition in locales.
969 [22295] locale: Don't warn on non-symbolic characters in locale sources in
971 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
972 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
974 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
975 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
976 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
977 [22322] libc: [mips64] wrong bits/long-double.h installed
978 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
979 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
981 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
982 [22343] malloc: Integer overflow in posix_memalign
983 [22347] libc: getrandom() returns the number of bytes that were copied to
984 the buffer even though the comments say "Return 0 on success and -1 on
986 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
987 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
989 [22370] dynamic-link: Incorrect note padding check
990 [22375] libc: malloc returns pointer from tcache_get when should return
991 NULL (CVE-2017-17426)
992 [22377] math: iseqsig, float128 and C++
993 [22382] localedata: Error in tpi_PG locale
994 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
995 inside the ASCII printable range
996 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
998 [22403] localedata: Slash needs escaping in some locales
999 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1000 [22409] network: res_hnok does not accept some host names used on the
1002 [22412] network: res_dnok, res_hnok should perform syntax checks
1003 [22413] network: ns_name_pton ignores syntactically invalid trailing
1005 [22415] stdio: setvbuf can lead to invalid free/segfault
1006 [22432] build: Non-deterministic build
1007 [22439] malloc: malloc_info should compute summary statistics for all sub-
1009 [22442] network: if_nametoindex could report index for the wrong
1010 networking interface
1011 [22446] build: aliasing violation calling readlink in handle_request
1012 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1013 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1014 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1015 __nscd_hash/__nss_hash
1016 [22463] network: p_secstodate overflow handling
1017 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1018 [22478] libc: sigwait can fail with EINTR
1019 [22505] libc: ldconfig processes include directive in locale-specific
1021 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1022 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1023 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1024 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1025 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1026 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1027 same as for Croatian
1028 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1029 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1030 [22577] libc: missing newline after "cannot allocate TLS data structures
1032 [22588] manual: manual/conf.texi: missing underscore in front of
1034 [22593] math: nextafter and nexttoward are declared with const attribute
1035 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1036 [22603] string: ia64 memchr overflows internal pointer check
1037 [22605] libc: SH clone does not set the exit code correctly
1038 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1040 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1041 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1042 modified in case of success
1043 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1044 [22615] manual: manual: ambiguous wording about errno value in case of
1046 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1048 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1049 binaries or if /proc is not mounted (CVE-2017-16997)
1050 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1051 [22630] build: $(no-pie-ldflag) is no longer effective
1052 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1053 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1054 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1055 [22637] nptl: guard size is subtracted from thread stack size instead of
1057 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1058 [22657] localedata: hu_HU: Avoid double space in date
1059 [22660] math: fmax, fmin sNaN handling on alpha
1060 [22664] libc: New warning of GCC8
1061 [22665] math: alpha: ceil and floor raise inexact exceptions
1062 [22666] math: alpha: trunc raise inexact exceptions
1063 [22667] libc: makecontext lacks stack alignment on i386
1064 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1065 [22679] libc: getcwd(3) can succeed without returning an absolute path
1067 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1068 elision and tunables
1069 [22687] math: [powerpc-nofpu] complex long double functions spurious
1071 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1072 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1073 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1075 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1076 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1078 [22701] nis: Incomplete removal of libnsl
1079 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1080 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1081 [22715] dynamic-link: FAIL: elf/tst-audit10
1082 [22719] libc: Backtrace tests fail on hppa
1083 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1084 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1085 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1086 before the first call to crypt_r () in crypt/badsalttest.c
1093 * A per-thread cache has been added to malloc. Access to the cache requires
1094 no locks and therefore significantly accelerates the fast path to allocate
1095 and free small amounts of memory. Refilling an empty cache requires locking
1096 the underlying arena. Performance measurements show significant gains in a
1097 wide variety of user workloads. Workloads were captured using a special
1098 instrumented malloc and analyzed with a malloc simulator. Contributed by
1099 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1101 * Unicode 10.0.0 Support: Character encoding, character type info, and
1102 transliteration tables are all updated to Unicode 10.0.0, using
1103 generator scripts contributed by Mike FABIAN (Red Hat).
1104 These updates cause user visible changes, especially the changes in
1105 wcwidth for many emoji characters cause problems when emoji sequences
1106 are rendered with pango, see for example:
1107 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1109 * Collation of Hungarian has been overhauled and is now consistent with "The
1110 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1113 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1115 - The GNU C Library will now detect when /etc/resolv.conf has been
1116 modified and reload the changed configuration. The new resolver option
1117 “no-reload” (RES_NORELOAD) disables this behavior.
1119 - The GNU C Library now supports an arbitrary number of search domains
1120 (configured using the “search” directive in /etc/resolv.conf);
1121 previously, there was a hard limit of six domains. For backward
1122 compatibility, applications that directly modify the ‘_res’ global
1123 object are still limited to six search domains.
1125 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1126 Library will now randomly pick a name server from the configuration as a
1127 starting point. (Previously, the second name server was always used.)
1129 * The tunables feature is now enabled by default. This allows users to tweak
1130 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1132 * New function reallocarray, which resizes an allocated block (like realloc)
1133 to the product of two sizes, with a guaranteed clean failure upon integer
1134 overflow in the multiplication. Originally from OpenBSD, contributed by
1135 Dennis Wölfing and Rüdiger Sonderfeld.
1137 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1138 These are extended versions of preadv and pwritev, respectively, taking an
1139 additional flags argument. The set of supported flags depends on the
1140 running kernel; full support currently requires kernel 4.7 or later.
1142 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1143 create a new session ID for the spawned process. This feature is
1144 scheduled to be added to the next major revision of POSIX; for the time
1145 being, it is available under _GNU_SOURCE.
1147 * errno.h is now safe to use from C-preprocessed assembly language on all
1148 supported operating systems. In this context, it will only define the
1149 Exxxx constants, as preprocessor macros expanding to integer literals.
1151 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1152 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1153 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1154 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1156 To compile programs that use this feature, the compiler must support
1157 128-bit floating point with the type name _Float128 (as defined by TS
1158 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1159 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1160 must be defined to make the new interfaces visible.
1162 The new functions and macros correspond to those present for other
1163 floating-point types (except for a few obsolescent interfaces not
1164 supported for the new type), with F128 or f128 suffixes; for example,
1165 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1166 printf or scanf formats for the new type; the strfromf128 and strtof128
1167 interfaces should be used instead.
1169 Deprecated and removed features, and other changes affecting compatibility:
1171 * The synchronization that pthread_spin_unlock performs has been changed to
1172 now be equivalent to a C11 atomic store with release memory order to the
1173 spin lock's memory location. Previously, several (but not all)
1174 architectures used stronger synchronization (e.g., containing what is
1175 often called a full barrier). This change can improve performance, but
1176 may affect odd fringe uses of spin locks that depend on the previous
1177 behavior (e.g., using spin locks as atomic variables to try to implement
1178 Dekker's mutual exclusion algorithm).
1180 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1183 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1184 will only be built and installed when the GNU C Library is configured with
1185 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1186 as TIRPC or rpcsvc-proto, to be used.
1188 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1189 libnss_compat, are deprecated, and will not be built or installed by
1192 The NIS(+) support library, libnsl, is also deprecated. By default, a
1193 compatibility shared library will be built and installed, but not headers
1194 or development libraries. Only a few NIS-related programs require this
1195 library. (In particular, the GNU C Library has never required programs
1196 that use 'gethostbyname' to be linked with libnsl.)
1198 Replacement implementations based on TIRPC, which additionally support
1199 IPv6, are available from <https://github.com/thkukuk/>. The configure
1200 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1201 name service modules, to be built and installed.
1203 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1204 support is enabled, the configured recursive resolver must support EDNS.
1205 (Responding to EDNS-enabled queries with responses which are not
1206 EDNS-enabled is fine, but FORMERR responses are not.)
1208 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1209 servers have not supported this opcode for a long time.
1211 * The _res_opcodes variable has been removed from libresolv. It had been
1212 exported by accident.
1214 * <string.h> no longer includes inline versions of any string functions,
1215 as this kind of optimization is better done by the compiler. The macros
1216 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1218 * The nonstandard header <xlocale.h> has been removed. Most programs should
1219 use <locale.h> instead. If you have a specific need for the definition of
1220 locale_t with no other declarations, please contact
1221 libc-alpha@sourceware.org and explain.
1223 * The obsolete header <sys/ultrasound.h> has been removed.
1225 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1227 * The obsolete function cfree has been removed. Applications should use
1230 * The stack_t type no longer has the name struct sigaltstack. This changes
1231 the C++ name mangling for interfaces involving this type.
1233 * The ucontext_t type no longer has the name struct ucontext. This changes
1234 the C++ name mangling for interfaces involving this type.
1236 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1237 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1238 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1239 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1240 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1241 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1242 fpregset_t type no longer has the name struct fpu. This changes the C++
1243 name mangling for interfaces involving those types.
1245 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1246 synced with the kernel:
1248 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1249 are not supported on this architecture and have been removed.
1251 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1252 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1253 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1255 Programs that assume the GET/SETREGS ptrace requests are universally
1256 available will now fail to build, instead of malfunctioning at runtime.
1258 Changes to build and runtime requirements:
1260 * Linux kernel 3.2 or later is required at runtime, on all architectures
1261 supported by that kernel. (This is a change from version 2.25 only for
1264 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1266 * On most architectures, GCC 4.9 or later is required to build the GNU C
1267 Library. On powerpc64le, GCC 6.2 or later is required.
1269 Older GCC versions and non-GNU compilers are still supported when
1270 compiling programs that use the GNU C Library. (We do not know exactly
1271 how old, and some GNU extensions to C may be _de facto_ required. If you
1272 are interested in helping us make this statement less vague, please
1273 contact libc-alpha@sourceware.org.)
1275 Security related changes:
1277 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1278 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1280 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1281 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1283 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1284 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1287 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1288 fixed (CVE-2017-12133).
1290 The following bugs are resolved with this release:
1292 [984] network: Respond to changed resolv.conf in gethostbyname
1293 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1294 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1296 [12189] libc: __stack_chk_fail should not attempt a backtrace
1298 [14096] time: Race condition on timezone/tst-timezone.out
1299 [14172] localedata: az_IR: new locale
1300 [14995] build: glibc fails to build if gold is the default linker, even if
1302 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1303 [16637] network: inet_pton function is accepting IPv6 with bad format
1304 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1305 x86 and other generic code
1306 [16875] localedata: ko_KR: fix lang_name
1307 [17225] localedata: ar_SY: localized month names for May and June are
1309 [17297] localedata: da_DK: wrong date_fmt string
1310 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1311 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1312 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1313 [19066] localedata: ar_SA abbreviated day and month names are in English
1314 [19569] network: resolv: Support an arbitrary number of search domains
1315 [19570] network: Implement random DNS server selection in the stub
1317 [19838] locale: localedef fails on PA-RISC
1318 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1319 order of 0D36 and 0D37
1320 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1322 [20098] libc: FAIL: debug/backtrace-tst on hppa
1323 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1325 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1327 [20313] localedata: Update locale data to Unicode 9.0
1328 [20424] manual: Document how to provide a malloc replacement
1329 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1331 [20686] locale: Add el_GR@euro to SUPPORTED.
1332 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1333 failures consistently
1334 [21015] dynamic-link: Document and fix --enable-bind-now
1335 [21016] nptl: pthread_cond support is broken on hppa
1336 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1337 frame-pointer on i386
1338 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1340 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1341 [21088] libc: Build fails with --enable-static-nss
1342 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1343 generic c code is used
1344 [21109] libc: Tunables broken on big-endian
1345 [21112] math: powf has large ulp errors with base close to 1 and exponent
1347 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1349 [21120] malloc: glibc malloc is incompatible with GCC 7
1350 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1352 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1353 (-0.0) when linking with -lieee
1354 [21171] math: log10, log2 and lgamma return incorrect results
1355 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1356 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1357 [21207] localedata: ce_RU: update weekdays from CLDR
1358 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1359 [21217] localedata: Update months from CLDR-31
1360 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1361 [21243] libc: support_delete_temp_file should issue warning for failed
1363 [21244] libc: support resolv_test_start() socket fd close should be
1365 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1366 new posix_spawn implementation
1367 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1368 leads to lower CPU frequency
1369 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1370 [21261] libc: [sparc64] bits/setjmp.h namespace
1371 [21267] network: [mips] bits/socket.h IOC* namespace
1372 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1373 [21270] libc: mmap64 silently truncates large offset values
1374 [21275] libc: posix_spawn always crashes on ia64 now
1375 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1376 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1377 [21289] libc: Incorrect declaration for 32-bit platforms with
1378 _FILE_OFFSET_BITS=64 causes build error
1379 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1380 not support gethostbyname4_r
1381 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1383 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1385 [21340] libc: Support POSIX_SPAWN_SETSID
1386 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1387 [21359] network: ns_name_pack needs additional byte in destination buffer
1388 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1389 against fragmentation attacks (CVE-2017-12132)
1390 [21369] network: resolv: Remove EDNS fallback
1391 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1393 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1394 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1395 [21393] stdio: Missing dup3 error check in freopen, freopen64
1396 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1397 [21399] localedata: Bad description for U00EC in
1398 localedata/charmaps/CP1254
1399 [21411] malloc: realloc documentation error
1400 [21426] network: sys/socket.h uio.h namespace
1401 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1402 [21445] libc: signal.h bsd_signal namespace
1403 [21455] network: Network headers stdint.h namespace
1404 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1406 [21475] network: resolv: Overlong search path is truncated mid-label
1407 [21511] libc: sigstack namespace
1408 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1409 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1410 bits/syscall.h: No such file or directory
1411 [21517] libc: struct sigaltstack namespace
1412 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1413 [21533] localedata: Update locale data to Unicode 10.0
1415 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1416 junk at end of line, first unrecognized character is `@'
1417 [21538] libc: SIG_HOLD missing for XPG4
1418 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1419 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1420 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1421 [21543] libc: sigevent namespace
1422 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1423 size for general purpose registers in mcontext_t structure
1424 [21550] libc: sigwait namespace
1425 [21552] libc: XPG4 bsd_signal namespace
1426 [21554] libc: sigpause namespace
1427 [21560] libc: sys/wait.h signal.h namespace
1428 [21561] libc: waitid namespace
1429 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1430 '/usr/include/stdlib.h': Permission denied
1431 [21575] libc: sys/wait.h missing struct rusage definition
1432 [21584] libc: sigaltstack etc namespace
1433 [21597] libc: siginterrupt namespace
1434 [21607] math: hppa: FAIL: math/test-tgmath
1435 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1437 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1438 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1439 stack and heap (CVE-2017-1000366)
1440 [21625] libc: wait3 namespace
1441 [21654] nss: Incorrect pointer alignment in NSS group merge result
1443 [21657] network: Parse interface zone id for node-local multicast
1444 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1446 [21666] libc: .symver is used on common symbol
1447 [21668] network: resolv: res_init cross-thread broadcast introduces race
1449 [21687] math: tgmath.h totalorder, totalordermag return type
1450 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1452 [21696] libc: Incorrect assumption of of __cpu_mask in
1453 posix/sched_cpucount.c
1454 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1455 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1457 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1458 leading to relocation crash
1459 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1460 `resolv_conf_matches (resp, conf)' failed.
1461 [21710] localedata: Added Samoan language locale for Samoa
1462 [21711] localedata: Pashto yesstr/nostr locale are missing
1463 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1464 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1465 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1466 [21723] localedata: yesstr/nostr missing for Chinese language locale
1467 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1469 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1471 [21728] localedata: New Locale for Tongan language
1472 [21729] localedata: incorrect LC_NAME fields for hi_IN
1473 [21733] localedata: yesstr and nostr are missing for zh_HK
1474 [21734] localedata: Missing yesstr and nostr are for kw_GB
1475 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1476 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1477 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1478 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1480 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1482 [21749] localedata: Wrong abbreviated day name (“abday”) for
1484 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1485 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1486 [21759] localedata: missing yesstr and nostr for Tigrinya
1487 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1488 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1489 [21767] localedata: Missing Bislama locales
1490 [21768] localedata: Missing yesstr and nostr for aa_ET
1491 [21770] localedata: Missing Field in li_NL
1492 [21778] nptl: Robust mutex may deadlock
1493 [21779] libc: MicroBlaze segfaults when loading libpthread
1494 [21783] localedata: Fix int_select international_call_prefixes
1495 [21784] localedata: Inconsistency in country_isbn
1496 [21788] localedata: Missing Country Postal Abbreviations
1497 [21794] localedata: Added-country_isbn-for-Italy
1498 [21795] localedata: Add/Fix country_isbn for France
1499 [21796] localedata: Added country_isbn for Republic of Korea
1500 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1501 [21799] localedata: Added int_select international_call_prefixes
1502 [21801] localedata: Added int_select international_call_prefixes
1503 [21804] nptl: Double semicolon in thread-shared-types.h
1504 [21807] localedata: LC_ADDRESS fix for pap_CW
1505 [21808] localedata: Fix LC_ADDRESS for pap_AW
1506 [21821] localedata: Added country_name in mai_IN
1507 [21822] localedata: Fix LC_TIME for mai_IN
1508 [21823] localedata: missing yesstr, nostr for sa_IN
1509 [21825] localedata: Fix name_mrs for mag_IN
1510 [21828] localedata: 2.26 changelog should mention user visible changes
1512 [21835] localedata: Added Maithili language locale for Nepal
1513 [21838] localedata: Removed redundant data for the_NP
1514 [21839] localedata: Fix LC_MONETARY for ta_LK
1515 [21844] localedata: Fix Latin characters and Months Sequence.
1516 [21848] localedata: Fix mai_NP Title Name
1521 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1522 24731-2:2010, is supported to enable declarations of functions from that
1523 TR. Note that not all functions from that TR are supported by the GNU C
1526 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1527 18661-1:2014, is supported to enable declarations of functions and macros
1528 from that TS. Note that not all features from that TS are supported by
1531 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1532 18661-4:2015, is supported to enable declarations of functions and macros
1533 from that TS. Note that most features from that TS are not supported by
1536 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1537 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1538 Since the GNU C Library defaults to a much newer revision of POSIX, this
1539 will only affect programs that specifically request an old conformance
1540 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1541 see a change in the visible declarations, but a program compiled with
1542 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1545 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1546 defined by all multithreaded code, but glibc has not required this for
1549 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1550 means that in a future release, the macros “major”, “minor”, and “makedev”
1551 will only be available from <sys/sysmacros.h>.
1553 These macros are not part of POSIX nor XSI, and their names frequently
1554 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1555 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1556 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1559 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1560 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1561 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1563 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1564 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1565 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1566 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1567 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1568 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1569 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1570 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1571 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1572 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1573 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1576 * New <math.h> features are added from TS 18661-1:2014:
1578 - Signaling NaN macros: SNANF, SNAN, SNANL.
1580 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1581 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1582 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1584 - llogb functions: the llogb, llogbf and llogbl functions, and the
1585 FP_LLOGB0 and FP_LLOGBNAN macros.
1587 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1590 - Comparison macros: iseqsig.
1592 - Classification macros: iscanonical, issubnormal, iszero.
1594 - Total order functions: totalorder, totalorderf, totalorderl,
1595 totalordermag, totalordermagf, totalordermagl.
1597 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1599 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1600 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1602 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1603 are added to libc. They convert a floating-point number into string.
1605 * Most of glibc can now be built with the stack smashing protector enabled.
1606 It is recommended to build glibc with --enable-stack-protector=strong.
1607 Implemented by Nick Alcock (Oracle).
1609 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1610 intended to be used instead of memset() to erase sensitive data after use;
1611 the compiler will not optimize out calls to explicit_bzero even if they
1612 are "unnecessary" (in the sense that no _correct_ program can observe the
1613 effects of the memory clear).
1615 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1616 to float instead of double. This does not affect the ABI of any libraries
1617 that are part of the GNU C Library, but may affect the ABI of other
1618 libraries that use this type in their interfaces.
1620 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1621 float_t and double_t types are now defined to long double instead of float
1622 and double. These options are not the default, and this does not affect
1623 the ABI of any libraries that are part of the GNU C Library, but it may
1624 affect the ABI of other libraries that use this type in their interfaces,
1625 if they are compiled or used with those options.
1627 * The getentropy and getrandom functions, and the <sys/random.h> header file
1630 * The buffer size for byte-oriented stdio streams is now limited to 8192
1631 bytes by default. Previously, on Linux, the default buffer size on most
1632 file systems was 4096 bytes (and thus remains unchanged), except on
1633 network file systems, where the buffer size was unpredictable and could be
1634 as large as several megabytes.
1636 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1637 for the Linux quota interface which predates kernel version 2.4.22 has
1640 * The malloc_get_state and malloc_set_state functions have been removed.
1641 Already-existing binaries that dynamically link to these functions will
1642 get a hidden implementation in which malloc_get_state is a stub. As far
1643 as we know, these functions are used only by GNU Emacs and this change
1644 will not adversely affect already-built Emacs executables. Any undumped
1645 Emacs executables, which normally exist only during an Emacs build, should
1646 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1648 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1649 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1650 “no-ip6-dotint” had already been the default, and support for the
1651 “ip6-dotint” option was removed from the Internet in 2006.
1653 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1654 flag from <resolv.h> have been removed. The option relied on a
1655 backwards-incompatible DNS extension which was never deployed on the
1658 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1659 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1660 They were already unimplemented.
1662 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1663 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1664 removed again from the IETF name lookup interface specification in RFC
1665 2553. Applications should use getaddrinfo instead.
1667 * DNSSEC-related declarations and definitions have been removed from the
1668 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1669 decode the data part of DNSSEC record types. Previous versions of glibc
1670 only implemented minimal support for the previous version of DNSSEC, which
1671 is incompatible with the currently deployed version.
1673 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1674 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1675 <arpa/nameser.h> header file because the distinction between RR types and
1676 meta-RR types is not officially standardized, subject to revision, and
1677 thus not suitable for encoding in a macro.
1679 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1680 and rhook members of the res_state type in <resolv.h> have been removed.
1681 The glibc stub resolver did not support these hooks, but the header file
1682 did not reflect that.
1684 * For multi-arch support it is recommended to use a GCC which has
1685 been built with support for GNU indirect functions. This ensures
1686 that correct debugging information is generated for functions
1687 selected by IFUNC resolvers. This support can either be enabled by
1688 configuring GCC with '--enable-gnu-indirect-function', or by
1689 enabling it by default by setting 'default_gnu_indirect_function'
1690 variable for a particular architecture in the GCC source file
1693 * GDB pretty printers have been added for mutex and condition variable
1694 structures in POSIX Threads. When installed and loaded in gdb these pretty
1695 printers show various pthread variables in human-readable form when read
1696 using the 'print' or 'display' commands in gdb.
1698 * Tunables feature added to allow tweaking of the runtime for an application
1699 program. This feature can be enabled with the '--enable-tunables' configure
1700 flag. The GNU C Library manual has details on usage and README.tunables has
1701 instructions on adding new tunables to the library.
1703 * A new version of condition variables functions have been implemented in
1704 the NPTL implementation of POSIX Threads to provide stronger ordering
1707 * A new version of pthread_rwlock functions have been implemented to use a more
1708 scalable algorithm primarily through not using a critical section anymore to
1711 Security related changes:
1713 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1714 have been created with makecontext could fail to terminate due to a
1715 missing .cantunwind annotation. This has been observed to lead to a hang
1716 (denial of service) in some Go applications compiled with gccgo. Reported
1717 by Andreas Schwab. (CVE-2016-6323)
1719 * The DNS stub resolver functions would crash due to a NULL pointer
1720 dereference when processing a query with a valid DNS question type which
1721 was used internally in the implementation. The stub resolver now uses a
1722 question type which is outside the range of valid question type values.
1725 The following bugs are resolved with this release:
1727 [4099] stdio: Overly agressive caching by stream i/o functions.
1728 [7065] build: Support building glibc with -fstack-protector or -fstack-
1730 [9842] localedata: en_CA: incorrect date format
1731 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1732 before it started waiting
1733 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1734 [15765] nptl: sem_open is wrongly a cancellation point
1735 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1736 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1738 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1739 library linked with pthread
1740 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1741 "FIXME: Ingo" issue)
1742 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1743 [17252] libc: getrandom and getentropy syscall
1744 [17730] malloc: thread-local storage is sometimes improperly free()'d
1745 after being __libc_memalign()'d
1746 [18241] stdio: failed fseek on memstream does not set errno and can fail
1748 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1750 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1751 PTHREAD_PRIO_INHERIT mutex on ARM
1752 [18784] network: res_query and related function crash for special record
1753 type queries (CVE-2015-5180)
1754 [19380] math: strtod does not raise "inexact"
1755 [19387] string: Integer overflow in memchr
1756 [19390] string: Integer overflow in strncat
1757 [19398] build: linknamespace tests fail with massively parallel build
1758 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1760 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1761 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1762 [19514] libc: [PATCH] Fix spelling errors in spelling
1763 "implement"/"implementation" in several places
1764 [19582] network: Deprecate RES_USE_INET6
1765 [19673] manual: clog10 docs appear to be erroneous
1766 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1767 causes a segmentation fault
1768 [19826] libc: invalid pointer returned from __tls_get_addr with static
1770 [20016] network: resolv: Remove hooks support from the API
1771 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1773 [20033] math: [x86_64] vectorized math function don't call the __finite
1775 [20116] nptl: use after free in pthread_create
1776 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1777 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1779 [20311] nptl: please install proc_service.h
1780 [20366] libc: Compilation errors in installed headers in strict-compliance
1782 [20370] malloc: malloc: Arena free list management is still racy
1783 (incorrect fix in bug 19243)
1784 [20386] libc: assert (X = 0) does not result in GCC warning
1785 [20432] malloc: malloc: Minimize interface required for interposition
1786 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1787 in _Unwind_Backtrace (CVE-2016-6323)
1788 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1789 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1791 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1792 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1793 [20477] network: resolv: incorrect double-checked locking related to
1795 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1797 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1799 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1800 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1801 cause transition penalty
1802 [20517] math: sparcv9 missing fdiml compat symbol
1803 [20524] manual: strverscmp is inconsistent
1804 [20525] libc: <sys/quota.h> should be based on kernel headers
1805 [20539] math: GCC 7 -static -lm fails to link at -O3
1806 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1808 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1810 [20591] network: Remove obsolete DNSSEC support
1811 [20592] network: DNS resource record type classification macros in
1812 <arpa/nameser.h> are incorrect
1813 [20593] network: Update DNS RR type definitions
1814 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1815 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1817 [20629] network: libresolv: Remove support for bitstring labels
1819 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1820 assertion: "maxidx >= 2"
1821 [20660] math: [arm] Use VSQRT
1822 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1823 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1824 [20689] libc: Test for FMA should also check for AVX.
1825 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1826 [20715] math: iszero macro breaks existing code
1827 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1828 [20728] libc: powerpc: Missing TOC stub in clone
1829 [20729] build: glibc-2.24 fails to build for i486 with -Os
1830 [20750] build: Build fails with default PIE enabled gcc-6.x
1831 [20768] math: [alpha] sqrt fegetenv namespace
1832 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1834 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1835 __FLT_EVAL_METHOD__ is 2
1836 [20790] network: rpcgen buffer overrun in get_prog_declaration
1837 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1838 [20829] libc: crypt snprintf namespace
1839 [20847] libc: tst-vfork3 failure
1840 [20855] math: Default bits/mathdef.h has inappropriate float_t
1841 [20859] math: [sh4] FP_ILOGB0 invalid
1842 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1843 U+20AC), not same as GBK
1844 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1845 [20916] math: pow handling of sNaN arguments
1846 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1847 [20919] math: Bad pow (qNaN, 0) result with -lieee
1848 [20940] math: hypot sNaN handling
1849 [20947] math: fmax, fmin sNaN handling
1850 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1851 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1853 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1854 [20973] nptl: robust mutexes: Lost wake-ups
1855 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1856 [20978] nis: strlen on null pointer in nss_nisplus
1857 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1859 [21014] string: i686 memchr overflows internal pointer check
1860 [21019] libc: [mips] n32 lseek incorrect on overflow
1861 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1862 [21026] libc: [MIPS] readahead syscall is broken on n64
1863 [21028] math: Fallback fesetexceptflag should always succeed
1864 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1865 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1867 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1868 [21061] librt: [microblaze] librt lost clock_* exports
1869 [21073] libc: tunables: insecure environment variables passed to
1870 subprocesses with AT_SECURE
1871 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1875 * The minimum Linux kernel version that this version of the GNU C Library
1876 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1877 kernel version 2.6.32 or later suffices (on architectures that already
1878 required kernel versions more recent than 3.2, those requirements remain
1879 unchanged). Linux 3.2 or later kernel headers are required on all
1882 * The pap_AN locale has been deleted. This has been deprecated for a long
1883 time. It has been replaced by pap_AW & pap_CW, both of which have long
1884 been included in previous releases.
1886 * The readdir_r and readdir64_r functions have been deprecated. It is
1887 recommended to use readdir and readdir64 instead.
1889 * The type “union wait” has been removed. It was deprecated in the early
1890 1990s and never part of POSIX. Application code should use the int type
1891 instead of “union wait”.
1893 * A new NSS action is added to facilitate large distributed system
1894 administration. The action, MERGE, allows remote user stores like LDAP
1895 to be merged into local user stores like /etc/groups in order to provide
1896 easy to use, updated, and managed sets of merged credentials. The new
1897 action can be used by configuring it in /etc/nsswitch.conf:
1898 group: files [SUCCESS=merge] nis
1899 Implemented by Stephen Gallagher (Red Hat).
1901 * The deprecated __malloc_initialize_hook variable has been removed from the
1904 * The long unused localedef --old-style option has been removed. It hasn't
1905 done anything in over 16 years. Scripts using this option can safely
1908 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1909 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1910 return the next representable value in the direction of positive infinity
1911 and the nextdown functions return the next representable value in the
1912 direction of negative infinity. These are currently enabled as GNU
1915 Security related changes:
1917 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1918 could result in a stack overflow when getnetbyname was called with an
1919 overly long name. (CVE-2016-3075)
1921 * Previously, getaddrinfo copied large amounts of address data to the stack,
1922 even after the fix for CVE-2013-4458 has been applied, potentially
1923 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1924 instead. Reported by Michael Petlan. (CVE-2016-3706)
1926 * The glob function suffered from a stack-based buffer overflow when it was
1927 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1928 Reported by Alexander Cherepanov. (CVE-2016-1234)
1930 * The Sun RPC UDP client could exhaust all available stack space when
1931 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1932 alloca plugin for GCC. (CVE-2016-4429)
1934 * The IPv6 name server management code in libresolv could result in a memory
1935 leak for each thread which is created, performs a failing naming lookup,
1936 and exits. Over time, this could result in a denial of service due to
1937 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1939 The following bugs are resolved with this release:
1941 [1170] localedata: ne_NP: update Nepali locale definition file
1942 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1944 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1945 [6796] math: fdim() does not set errno on overflow
1946 [10354] libc: posix_spawn should use vfork() in more cases than presently
1947 [11213] localedata: localedata: add copyright disclaimer to locale files
1948 [12143] localedata: chr_US: new Cherokee locale
1949 [12450] localedata: sgs_LT: new locale
1950 [12676] localedata: ln_CD: new locale
1951 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1953 [13304] math: fma, fmaf, fmal produce wrong results
1954 [14259] build: --localedir arg to configure is ignored
1955 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1956 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1957 [14934] localedata: es_CL: wrong first weekday chilean locale
1958 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1960 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1962 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1963 [15368] nptl: raise() is not async-signal-safe
1964 [15479] math: ceil, floor, round and trunc raise inexact exception
1965 [15578] localedata: kk_KZ: various updates
1966 [16003] localedata: pap_AN: punt old locale
1967 [16137] localedata: iw_IL: punt old locale
1968 [16190] localedata: eo: new esperanto locale
1969 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1970 [16742] malloc: race condition: pthread_atfork() called before first
1971 malloc() results in unexpected locking behaviour/deadlocks
1972 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1974 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1975 [17565] localedata: pt_PT: wrong (work-)week start
1976 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1977 [17950] build: Build fails with -msse
1978 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1979 [18433] libc: posix_spawn does not return correctly upon failure to
1981 [18453] localedata: charmaps/IBM875: incorrect codes
1982 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1984 [18896] localedata: he_IL: improvements for currency
1985 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1986 Romanian locale data
1987 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1989 [19056] libc: Deprecate readdir_r
1990 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1992 [19198] localedata: nl_NL: small improvements for Dutch locales
1993 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1994 nameservers (CVE-2016-5417)
1995 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1997 [19400] locale: Language missing in "iso-639.def", trivial fix in
1999 [19431] malloc: Deadlock between fflush, getdelim, and fork
2000 [19505] libc: Incorrect file descriptor validity checks in
2001 posix_spawn_file_actions_add{open,close,dup2}
2002 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2003 when using RTLD_NEXT
2004 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2005 `intl/{gettextP,loadinfo}.h'
2006 [19534] libc: execle, execlp may use malloc
2007 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2008 [19573] network: res_nclose and __res_maybe_init disagree about name
2009 server initialization, breaking Hesiod
2010 [19575] localedata: Status of GB18030 tables
2011 [19581] localedata: sr_* date_fmt string contains additional newline
2012 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2014 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2015 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2016 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2017 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2019 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2021 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2023 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2024 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2025 [19633] locale: strfmon_l applies global locale to number formatting
2026 [19642] network: Memory leak in getnameinfo
2027 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2028 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2030 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2031 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2033 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2034 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2036 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2038 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2040 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2041 pointers and lengths in error-case.
2042 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2043 report errors on UTF-16 surrogates values.
2044 [19755] nscd: nscd assertion failure in gc
2045 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2046 [19759] libc: mempcpy shouldn't be inlined
2047 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2048 [19765] libc: s390 needs an optimized mempcpy
2049 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2050 NAME_MAX limit assumption (CVE-2016-1234)
2051 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2052 [19787] network: Missing and incorrect truncation checks in getnameinfo
2053 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2055 [19791] network: Assertion failure in res_query.c with un-connectable name
2057 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2058 [19822] math: libm.so install clobbers old version
2059 [19825] network: resolv: send_vc can return uninitialized data in second
2060 response to getaddrinfo
2061 [19830] network: nss_dns: should check RDATA length against buffer length
2062 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2063 confronted with A/AAAA records of invalid size
2064 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2065 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2066 [19853] stdio: Printing IBM long double in decimal with high precision is
2068 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2069 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2070 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2072 [19865] network: Assertion failure or memory leak in
2073 _nss_dns_getcanonname_r
2074 [19868] network: nss_dns: netent code does not skip over non-PTR records
2075 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2077 [19881] string: Improve x86-64 memset
2078 [19907] string: Incorrect memcpy tests
2079 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2081 [19925] libc: termios.h XCASE namespace
2082 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2083 [19929] libc: limits.h NL_NMAX namespace
2084 [19931] stdio: Memory leak in vfprintf
2085 [19957] libc: clone(CLONE_VM) access invalid parent memory
2086 [19963] localedata: en_IL: New locale
2087 [19989] stdio: stdio.h cuserid namespace
2088 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2090 [19996] locale: langinfo.h nl_langinfo_l namespace
2091 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2092 [20010] network: getaddrinfo: Stack overflow in hostent translation
2094 [20012] stdio: libio: fmemopen append mode failure
2095 [20014] stdio: stdio.h namespace for pre-threads POSIX
2096 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2097 [20023] libc: fcntl.h timespec namespace
2098 [20024] math: [x86_64] vectorized sincos trashes the stack
2099 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2100 [20041] time: sys/time.h timespec namespace
2101 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2102 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2103 [20051] libc: ttyslot in wrong header under wrong conditions
2104 [20054] libc: gethostname not declared for XPG4
2105 [20055] libc: termios.h missing tcgetsid for XPG4
2106 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2108 [20073] libc: sys/stat.h fchmod namespace
2109 [20074] libc: stdlib.h rand_r namespace
2110 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2111 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2113 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2114 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2116 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2117 [20119] libc: Wrong mask for processors level type from CPUID
2118 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2119 AS not supporting AVX512
2120 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2122 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2123 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2125 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2126 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2127 [20178] libc: posix_spawn{p} should not call exit
2128 [20191] stdio: libio: vtables hardening
2129 [20195] string: FMA4 detection requires CPUID execution with register
2131 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2132 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2134 [20212] math: acos (sNaN) returns sNaN
2135 [20213] math: asin (sNaN) returns sNaN
2136 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2137 [20218] math: [i386] asinhl (sNaN) returns sNaN
2138 [20219] math: [i386] atanhl (sNaN) returns sNaN
2139 [20222] stdio: fopencookie: Mangle function pointers
2140 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2141 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2142 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2144 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2145 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2146 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2147 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2148 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2149 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2150 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2151 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2152 [20240] math: modf (sNaN) returns sNaN
2153 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2154 [20250] math: frexp (sNaN) returns sNaN
2155 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2156 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2158 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2159 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2160 subscript is below array bounds [-Werror=array-bounds]
2161 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2162 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2164 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2165 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2167 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2168 "invalid" exceptions
2169 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2170 [20316] localedata: id_ID: Februari instead of Pebruari
2171 [20327] string: POWER8 strcasecmp returns incorrect result
2172 [20347] math: Failure: Test: j0_downward (0xap+0)
2173 [20348] libc: FAIL: misc/tst-preadvwritev64
2174 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2175 [20350] libc: There is no test for p{read,write}64
2176 [20357] math: Incorrect cos result for 1.5174239687223976
2177 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2181 * Unicode 8.0.0 Support: Character encoding, character type info, and
2182 transliteration tables are all updated to Unicode 8.0.0, using new
2183 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2184 These updates cause user visible changes, such as the fixes for bugs
2185 89, 16061, and 18568.
2187 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2188 kernel-internal CPU set size. This means that requests that change the
2189 CPU affinity which failed before (for example, an all-ones CPU mask) will
2190 now succeed. Applications that need to determine the effective CPU
2191 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2192 setting it because the kernel can adjust it (and the previous size check
2193 would not detect this in the majority of cases).
2195 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2196 the following new symbols are used: fts64_children, fts64_close,
2197 fts64_open, fts64_read and fts64_set.
2199 * getaddrinfo now detects certain invalid responses on an internal netlink
2200 socket. If such responses are received, an affected process will
2201 terminate with an error message of "Unexpected error <number> on netlink
2202 descriptor <number>" or "Unexpected netlink response of size <number> on
2203 descriptor <number>". The most likely cause for these errors is a
2204 multi-threaded application which erroneously closes and reuses the netlink
2205 file descriptor while it is used by getaddrinfo.
2207 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2208 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2209 unnecessary serialization of memory allocation requests across threads.
2210 The defect is now corrected. Users should see a substantial increase in
2211 the concurent throughput of allocation requests for applications which
2212 trigger this bug. Affected applications typically create and
2213 destroy threads frequently. (Bug 19048 was reported and analyzed by
2216 * There is now a --disable-timezone-tools configure option for disabling the
2217 building and installing of the timezone related utilities (zic, zdump, and
2218 tzselect). This is useful for people who build the timezone data and code
2219 independent of the GNU C Library.
2221 * The obsolete header <regexp.h> has been removed. Programs that require
2222 this header must be updated to use <regex.h> instead.
2224 * The obsolete functions bdflush, create_module, get_kernel_syms,
2225 query_module and uselib are no longer available to newly linked binaries;
2226 the header <sys/kdaemon.h> has been removed. These functions and header
2227 were specific to systems using the Linux kernel and could not usefully be
2228 used with the GNU C Library on systems with version 2.6 or later of the
2231 * Optimized string, wcsmbs and memory functions for IBM z13.
2232 Implemented by Stefan Liebler.
2234 * Newly linked programs that define a variable called signgam will no longer
2235 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2236 require signgam to be set by those functions must ensure that they use the
2237 variable provided by the GNU C Library and declared in <math.h>, without
2238 defining their own copy.
2240 * The minimum GCC version that can be used to build this version of the GNU
2241 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2242 still be used to compile programs using the GNU C Library.
2244 Security related changes:
2246 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2247 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2249 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2250 the pointer guard feature. It is always enabled. Previously,
2251 LD_POINTER_GUARD could be used to disable security hardening in binaries
2252 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2255 * An integer overflow in hcreate and hcreate_r could lead to an
2256 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2258 * The catopen function no longer has unbounded stack usage. Reported by
2259 Max. (CVE-2015-8779)
2261 * The nan, nanf and nanl functions no longer have unbounded stack usage
2262 depending on the length of the string passed as an argument to the
2263 functions. Reported by Joseph Myers. (CVE-2014-9761)
2265 * A stack-based buffer overflow was found in libresolv when invoked from
2266 libnss_dns, allowing specially crafted DNS responses to seize control
2267 of execution flow in the DNS client. The buffer overflow occurs in
2268 the functions send_dg (send datagram) and send_vc (send TCP) for the
2269 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2270 family. The use of AF_UNSPEC triggers the low-level resolver code to
2271 send out two parallel queries for A and AAAA. A mismanagement of the
2272 buffers used for those queries could result in the response of a query
2273 writing beyond the alloca allocated buffer created by
2274 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2275 the overflow. Thanks to the Google Security Team and Red Hat for
2276 reporting the security impact of this issue, and Robert Holiday of
2277 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2279 The following bugs are resolved with this release:
2281 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2282 [887] math: Math library function "logb" and "nextafter" inconsistent
2283 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2284 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2285 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2286 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2288 [4404] localedata: German translation of "Alarm clock" is misleading
2289 [6799] math: nextafter() and nexttoward() doen't set errno on
2290 overflow/underflow errors
2291 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2293 [10432] nis: _nss_nis_setnetgrent assertion failure
2294 [11460] libc: fts has no LFS support
2295 [12926] network: getaddrinfo()/make_request() may spin forever
2296 [13065] nptl: Race condition in pthread barriers
2297 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2298 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2300 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2301 [14912] libc: Rename non-installed bits/*.h headers
2302 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2303 [15367] math: Let gcc use __builtin_isinf
2304 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2305 [15421] math: lgamma wrongly sets signgam for ISO C
2306 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2307 FE_INVALID with argument out of range
2308 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2310 [15786] dynamic-link: ifunc resolver functions can smash function
2312 [15918] math: Unnecessary check for equality in hypotf()
2313 [16061] localedata: Review / update transliteration data
2314 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2316 [16141] time: strptime %z offset restriction
2317 [16171] math: drem should be alias of remainder
2318 [16296] math: fegetround is pure?
2319 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2320 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2322 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2323 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2325 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2326 "Exception "Inexact" set" on ppc32
2327 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2328 [16517] math: Missing underflow exception from tanf/tan/tanl
2329 [16519] math: Missing underflow exception from sinhf
2330 [16520] math: Missing underflow exception from tanhf
2331 [16521] math: Missing underflow exception from exp2
2332 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2334 [16734] stdio: fopen calls mmap to allocate its buffer
2335 [16961] math: nan function incorrect handling of bad sequences
2336 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2337 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2338 [16985] locale: localedef: confusing error message when opening output
2340 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2341 [17197] locale: Redundant shift character in iconv conversion output at
2343 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2344 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2345 [17250] dynamic-link: static linking breaks nss loading
2346 (getaddrinfo/getpwnam/etc...)
2347 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2349 [17441] math: isnan() should use __builtin_isnan() in GCC
2350 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2351 (related to lock elision)
2352 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2353 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2354 [17887] time: strptime should be able to parse "+01:00" style timezones
2355 [17905] libc: catopen() Multiple unbounded stack allocations
2357 [18084] libc: backtrace (..., 0) dumps core on x86
2358 [18086] libc: nice() sets errno to 0 on success
2359 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2360 is too large (CVE-2015-8778)
2361 [18251] dynamic-link: SONAME missing when audit modules provides path
2362 [18265] libc: add attributes for wchar string and memory functions
2363 [18370] math: csqrt missing underflows
2364 [18421] libc: [hppa] read-only segment has dynamic relocations
2365 [18472] libc: Obsolete syscall wrappers should be compat symbols
2366 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2367 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2368 [18525] localedata: Remove locale timezone information
2369 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2370 [18568] localedata: Update locale data to Unicode 8.0
2371 [18589] locale: sort-test.sh fails at random
2372 [18595] math: ctan, ctanh missing underflows
2373 [18604] libc: assert macro-expands its argument
2374 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2375 contains a vector instruction exception.
2376 [18611] math: j1, jn missing errno setting on underflow
2377 [18618] localedata: sync Chechen locale definitions with other *_RU
2379 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2380 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2381 [18665] network: In send_dg, the recvfrom function is NOT always using the
2382 buffer size of a newly created buffer (CVE-2015-7547)
2383 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2385 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2387 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2388 [18699] math: tilegx cproj() for various complex infinities does not yield
2390 [18724] libc: Harden put*ent functions against data injection
2391 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2392 [18755] build: build errors with -DNDEBUG
2393 [18757] stdio: fmemopen fails to set errno on failure
2394 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2396 [18781] libc: openat64 lacks O_LARGEFILE
2397 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2398 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2399 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2400 [18790] math: [ldbl-128ibm] tanhl inaccurate
2401 [18795] libc: stpncpy fortification misses buffer lengths that are
2402 statically too large
2403 [18796] build: build fails for --disable-mathvec
2404 [18803] math: hypot missing underflows
2405 [18820] stdio: fmemopen may leak memory on failure
2406 [18823] math: csqrt spurious underflows
2407 [18824] math: fma spurious underflows
2408 [18825] math: pow missing underflows
2409 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2410 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2411 [18870] build: sem_open.c fails to compile with missing symbol
2413 [18872] stdio: Fix memory leak in printf_positional
2414 [18873] libc: posix_fallocate overflow check ineffective
2415 [18875] math: Excess precision leads incorrect libm
2416 [18877] libc: arm: mmap offset regression
2417 [18887] libc: memory corruption when using getmntent on blank lines
2418 [18918] localedata: hu_HU: change time to HH:MM:SS format
2419 [18921] libc: Regression: extraneous stat() and fstat() performed by
2421 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2422 binaries (CVE-2015-8777)
2423 [18951] math: tgamma missing underflows
2424 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2426 [18953] localedata: lt_LT: change currency symbol to the euro
2427 [18956] math: powf inaccuracy
2428 [18961] math: [i386] exp missing underflows
2429 [18966] math: [i386] exp10 missing underflows
2430 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2431 [18969] build: multiple string test failures due to missing locale
2433 [18970] libc: Reference of pthread_setcancelstate in libc.a
2434 [18977] math: float / long double Bessel functions not in XSI POSIX
2435 [18980] math: i386 libm functions return with excess range and precision
2436 [18981] math: i386 scalb*, ldexp return with excess range and precision
2437 [18982] stdio: va_list and vprintf
2438 [18985] time: Passing out of range data to strftime() causes a segfault
2440 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2441 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2442 [19012] locale: iconv_open leaks memory on error path
2443 [19016] math: clog, clog10 inaccuracy
2444 [19018] nptl: Mangle function pointers in tls_dtor_list
2445 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2446 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2447 [19048] malloc: malloc: arena free list can become cyclic, increasing
2449 [19049] math: [powerpc] erfc incorrect zero sign
2450 [19050] math: [powerpc] log* incorrect zero sign
2451 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2452 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2453 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2454 [19074] network: Data race in _res_hconf_reorder_addrs
2455 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2456 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2457 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2459 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2461 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2462 [19086] manual: posix_fallocate64 documented argument order is wrong.
2463 [19088] math: lround, llround missing exceptions close to overflow
2465 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2466 [19095] math: dbl-64 lrint incorrect for 64-bit long
2467 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2468 [19124] dynamic-link: ld.so failed to build with older assmebler
2469 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2470 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2471 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2472 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2473 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2474 pthread_setaffinity_np
2475 [19156] math: [ldbl-128] j0l spurious underflows
2476 [19164] nptl: tst-getcpu fails with many possible CPUs
2477 [19168] math: math/test-ildoubl and math/test-ldouble failure
2478 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2479 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2481 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2482 (FE_NOMASK_ENV) do not clear SSE exceptions
2483 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2484 _int_new_arena/reused_arena
2485 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2486 [19201] math: dbl-64 remainder incorrect sign of zero result
2487 [19205] math: bits/math-finite.h conditions do not match math.h and
2489 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2490 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2491 for C99-based standards
2492 [19212] libc: features.h not -Wundef clean
2493 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2495 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2496 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2497 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2499 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2500 "inexact" exceptions
2501 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2503 [19242] libc: strtol incorrect in Turkish locales
2504 [19243] malloc: reused_arena can pick an arena on the free list, leading
2505 to an assertion failure and reference count corruption
2506 [19253] time: tzset() ineffective when temporary TZ did not include DST
2508 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2509 [19270] math: [hppa] Shared libm missing __isnanl
2510 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2511 MAP_HUGETLB and MAP_STACK defines
2512 [19313] nptl: Wrong __cpu_mask for x32
2513 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2514 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2515 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2516 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2517 [19363] time: x32: times() return value wrongly truncates/sign extends
2519 [19367] dynamic-link: Improve branch prediction on Silvermont
2520 [19369] network: Default domain name not reset by res_ninit when "search"
2521 / "domain" entry is removed from resolv.conf
2522 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2524 [19385] localedata: bg_BG: time separator should be colon, not comma
2525 [19408] libc: linux personality syscall wrapper may erroneously return an
2526 error on 32-bit architectures
2527 [19415] libc: dladdr returns wrong names on hppa
2528 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2529 IBM905, IBM907, and IBM909
2530 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2531 [19443] build: build failures with -DDEBUG
2532 [19451] build: Make check fails on test-double-vlen2
2533 [19462] libc: Glibc failed to build with -Os
2534 [19465] math: Wrong code with -Os
2535 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2537 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2539 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2540 [19486] math: S390: Math tests fail with "Exception Inexact set".
2541 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2542 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2544 [19590] math: Fail to build shared objects that use libmvec.so functions.
2548 * The following bugs are resolved with this release:
2550 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2551 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2552 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2553 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2554 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2555 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2556 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2557 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2558 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2559 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2560 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2561 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2562 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2563 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2564 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2565 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2566 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2567 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2568 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2569 18657, 18676, 18694, 18696, 18887.
2571 * Cache information can be queried via sysconf() function on s390 e.g. with
2572 _SC_LEVEL1_ICACHE_SIZE as argument.
2574 * A buffer overflow in gethostbyname_r and related functions performing DNS
2575 requests has been fixed. If the NSS functions were called with a
2576 misaligned buffer, the buffer length change due to pointer alignment was
2577 not taken into account. This could result in application crashes or,
2578 potentially arbitrary code execution, using crafted, but syntactically
2579 valid DNS responses. (CVE-2015-1781)
2581 * The time zone file parser has been made more robust against crafted time
2582 zone files, avoiding heap buffer overflows related to the processing of
2583 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2584 large time zone data files. Overly long time zone specifiers in the TZ
2585 variable no longer result in stack overflows and crashes.
2587 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2588 for LD and GD on x86 and x86-64, has been implemented. You will need
2589 binutils-2.24 or later to enable this optimization.
2591 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2592 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2593 Hat). These updates cause user visible changes, such as the fix for bug
2596 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2597 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2598 condition in some applications.
2600 * Added vector math library named libmvec with the following vectorized x86_64
2601 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2603 The library can be disabled with --disable-mathvec. Use of the functions is
2604 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2605 Shared library libmvec.so is linked in as needed when using -lm (no need to
2606 specify -lmvec explicitly for not static builds).
2607 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2609 * A new fmemopen implementation has been added with the goal of POSIX
2610 compliance. The new implementation fixes the following long-standing
2611 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2612 old implementation is still present for use be by existing binaries.
2614 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2615 releases. It has been fixed to match 2.19 and older, but binaries built
2616 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2618 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2619 Contributed by Roland McGrath (Google).
2621 * The header <regexp.h> is deprecated, and will be removed in a future
2622 release. Use of this header will trigger a deprecation warning.
2623 Application developers should update their code to use <regex.h> instead.
2625 This header was formerly part of SUS, but was deprecated in 1994 and
2626 removed from the standard in 2001. Also, the glibc implementation
2627 leaks memory. See BZ#18681 for more details.
2631 * The following bugs are resolved with this release:
2633 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2634 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2635 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2636 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2637 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2638 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2639 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2640 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2641 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2642 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2645 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2646 too little memory for the to-be-scanned arguments and overflow the
2647 allocated buffer. The implementation now correctly computes the required
2648 buffer size when using malloc, and switches to malloc from alloca as
2651 * A new semaphore algorithm has been implemented in generic C code for all
2652 machines. Previous custom assembly implementations of semaphore were
2653 difficult to reason about or ensure that they were safe. The new version
2654 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2655 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2656 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2658 * Port to Altera Nios II has been contributed by Mentor Graphics.
2660 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2661 implementations for powerpc64/powerpc64le.
2662 Implemented by Adhemerval Zanella (IBM).
2664 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2665 and powerpc64le. This may improve lock scaling of existing programs on
2666 HTM capable systems. The lock elision code is only enabled with
2667 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2668 powerpc will issue a transaction abort on every syscall to avoid side
2669 effects being visible outside transactions.
2671 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2672 AArch64. Contributed by ARM Ltd.
2674 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2676 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2677 under certain input conditions resulting in the execution of a shell for
2678 command substitution when the application did not request it. The
2679 implementation now checks WRDE_NOCMD immediately before executing the
2680 shell and returns the error WRDE_CMDSUB as expected.
2682 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2683 processing format strings with a large number of format specifiers.
2685 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2686 infinite loop if the DNS response contained a PTR record of an unexpected
2689 * The minimum GCC version that can be used to build this version of the GNU
2690 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2691 still be used to compile programs using the GNU C Library.
2693 * The GNU C Library is now built with -Werror by default. This can be
2694 disabled by configuring with --disable-werror.
2696 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2698 * The obsolete sigvec function has been removed. This was the original
2699 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2700 programs have been using instead for about 25 years. Of course, ABI
2701 compatibility for old binaries using sigvec remains intact.
2703 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2704 with newer versions of bison.
2706 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2707 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2708 registers overlay two consecutive single-precision registers. MIPS32R2
2709 introduced a new FPU mode (FR=1) where double-precision registers extend the
2710 corresponding single-precision registers which is incompatible with the
2711 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2712 require the use of FR=1 making a transition necessary. New o32 ABI
2713 extensions enable users to migrate over time from the original o32 ABI
2714 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2715 tracks the ABI of any loaded object and verifies that new objects are
2716 compatible. Mode transitions will also be requested as required and
2717 unsupportable objects will be rejected. The ABI checks include both soft and
2718 hard float ABIs for o32, n32 and n64.
2720 GCC 5 with GNU binutils 2.25 onwards:
2721 It is strongly recommended that all o32 system libraries are built using the
2722 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2723 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2724 compiler using --with-fp-32=xx to set this by default.
2728 * The following bugs are resolved with this release:
2730 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2731 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2732 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2733 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2734 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2735 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2736 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2737 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2738 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2739 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2740 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2741 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2742 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2743 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2744 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2746 * Reverted change of ABI data structures for s390 and s390x:
2747 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2748 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2749 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2750 are preserved pointing straight to the same implementation as the old ones.
2751 Given that, new callers will simply provide a too-big buffer to these
2752 functions. Any applications/libraries out there that embed jmp_buf or
2753 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2754 against 2.19 headers will have to rebuilt again. This is necessary in any
2755 case to revert the breakage in their ABI caused by the glibc change.
2757 * Support for file description locks is added to systems running the
2758 Linux kernel. The standard file locking interfaces are extended to
2759 operate on file descriptions, not file descriptors, via the use of
2760 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2761 are associated with an open file instead of a process.
2763 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2765 * The minimum Linux kernel version that this version of the GNU C Library
2766 can be used with is 2.6.32.
2768 * Running the testsuite no longer terminates as soon as a test fails.
2769 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2770 with PASS or FAIL lines for individual tests. A summary of the results is
2771 printed, including a list of failing lists, and "make check" exits with
2772 error status if there were any unexpected failures. "make check
2773 stop-on-test-failure=y" may be used to keep the old behavior.
2775 * The am33 port, which had not worked for several years, has been removed
2778 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2779 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2780 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2781 conflicted with POSIX (support for which was removed in 2.19), the
2782 interfaces those macros enabled remain available when compiling with
2783 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2784 test macros defined.
2786 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2788 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2789 This may improve lock scaling of existing programs on TX capable systems.
2790 The lock elision code is only built with --enable-lock-elision=yes and
2791 then requires a GCC version supporting the TX builtins. With lock elision
2792 default mutexes are elided via __builtin_tbegin, if the cpu supports
2793 transactions. By default lock elision is not enabled and the elision code
2796 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2797 copy the path argument. This allowed programs to cause posix_spawn to
2798 deference a dangling pointer, or use an unexpected pathname argument if
2799 the string was modified after the posix_spawn_file_actions_addopen
2802 * All supported architectures now use the main glibc sysdeps directory
2803 instead of some being in a separate "ports" directory (which was
2804 distributed separately before glibc 2.17).
2806 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2807 On configurations that support it (all Linux configurations), it's now
2808 used regardless of the --enable-add-ons switch to configure. It is no
2809 longer possible to build such configurations without pthreads support.
2811 * Locale names, including those obtained from environment variables (LANG
2812 and the LC_* variables), are more tightly checked for proper syntax.
2813 setlocale will now fail (with EINVAL) for locale names that are overly
2814 long, contain slashes without starting with a slash, or contain ".." path
2815 components. (CVE-2014-0475) Previously, some valid locale names were
2816 silently replaced with the "C" locale when running in AT_SECURE mode
2817 (e.g., in a SUID program). This is no longer necessary because of the
2820 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2821 with application code using Intel MPX instructions. (With all previous
2822 versions, the MPX register state could be clobbered when making calls
2823 into or out of a shared library.) Note that while the new dynamic
2824 linker is compatible with all known x86 hardware whether or not it
2825 supports Intel MPX, some x86 instruction-set emulators might fail to
2826 handle the new instruction encodings. This is known to affect Valgrind
2827 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2828 release), and might affect other tools that do instruction emulation.
2830 * Support for loadable gconv transliteration modules has been removed.
2831 The support for transliteration modules has been non-functional for
2832 over a decade, and the removal is prompted by security defects. The
2833 normal gconv conversion modules are still supported. Transliteration
2834 with //TRANSLIT is still possible, and the //IGNORE specifier
2835 continues to be supported. (CVE-2014-5119)
2837 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2838 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2839 resulting a denial-of-service security vulnerability in applications which
2840 use functions related to iconv. (CVE-2014-6040)
2844 * The following bugs are resolved with this release:
2846 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2847 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2848 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2849 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2850 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2851 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2852 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2853 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2854 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2855 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2856 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2857 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2858 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2859 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2860 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2861 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2862 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2863 16431, 16453, 16474, 16506, 16510, 16529
2865 * Slovenian translations for glibc messages have been contributed by the
2866 Translation Project's Slovenian team of translators.
2868 * The public headers no longer use __unused nor __block. This change is to
2869 support compiling programs that are derived from BSD sources and use
2870 __unused internally, and to support compiling with Clang's -fblock
2871 extension which uses __block.
2873 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2874 large collation sequences to optimize multiple passes. This cache
2875 computation may overflow for large collation sequences and may cause a
2876 stack or buffer overflow. This is now fixed to use a slower algorithm
2877 which does not use a cache if there is an integer overflow.
2879 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2880 rules for large collation sequences to optimize multiple passes and falls
2881 back to alloca if malloc fails, resulting in a possible stack overflow.
2882 The implementation now falls back to an uncached collation sequence lookup
2885 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2886 initialized for static applications resulting in the security feature
2887 being disabled. The pointer guard is now correctly initialized to a
2888 random value for static applications. Existing static applications need
2889 to be recompiled to take advantage of the fix (bug 15754).
2891 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2892 to the d_name member of struct dirent, or omit the terminating NUL
2893 character. (Bugzilla #14699).
2895 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2896 aligned_alloc functions could allocate too few bytes or corrupt the
2897 heap when passed very large allocation size values (Bugzilla #15855,
2900 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2901 for AF_INET6 has been fixed (Bugzilla #16072).
2903 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2904 pap_CW, quz_PE, the_NP.
2906 * Substantially revised locales: gd_GB, ht_HT
2908 * The LC_ADDRESS field was updated to support country_car for almost all
2911 * ISO 1427 definitions were updated.
2913 * ISO 3166 definitions were updated.
2915 * The localedef utility now supports --big-endian and --little-endian
2916 command-line options to generate locales for a different system from that
2917 for which the C library was built.
2919 * Binary locale files now only depend on the endianness of the system for
2920 which they are generated and not on other properties of that system. As a
2921 consequence, binary files generated with new localedef may be incompatible
2922 with old versions of the GNU C Library, and binary files generated with
2923 old localedef may be incompatible with this version of the GNU C Library,
2924 in the following circumstances:
2926 + Locale files may be incompatible on m68k systems.
2928 + Locale archive files (but not separate files for individual locales) may
2929 be incompatible on systems where plain "char" is signed.
2931 * The configure option --disable-versioning has been removed. Builds with
2932 --disable-versioning had not worked for several years.
2934 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2935 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2937 * SystemTap probes for malloc have been introduced.
2939 * SystemTap probes for slow multiple precision fallback paths of
2940 transcendental functions have been introduced.
2942 * Support for powerpc64le has been added.
2944 * The soft-float powerpc port now supports e500 processors.
2946 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2948 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2949 set of header declarations that are enabled by default, even when other
2950 feature test macros or compiler options such as -std=c99 would otherwise
2951 disable some of those declarations.
2953 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2954 conflict with POSIX. The libbsd-compat library (which was a dummy library
2955 that did nothing) has also been removed.
2957 * Preliminary documentation about Multi-Thread, Async-Signal and
2958 Async-Cancel Safety has been added.
2960 * Change of ABI data structures for s390 and s390x:
2961 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2962 allow for future hardware extensions. All library functions that accept or
2963 return these structures were versioned in order to provide backwards
2964 compatibility with existing code. However, not all cases can be handled
2965 with interface versioning. If an application embeds either structure into
2966 another structure and that structure is passed to another compilation unit
2967 which expects the newer larger-sized object then any access to the new
2968 fields will result in undefined behaviour. Similarly any access to fields
2969 that were moved by the enlarging of the structures will cause undefined
2970 behaviour. To fix the undefined behaviour all objects that are part of the
2971 final in-memory image must agree on the size of structures, and this may
2972 require recompilation.
2976 * The following bugs are resolved with this release:
2978 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2979 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2980 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2981 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2982 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2983 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2984 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2985 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2986 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2987 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2988 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2989 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2990 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2993 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2994 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2995 Distributions can re-enable building and using pt_chown via the new configure
2996 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2997 considerable security risks and should only be used if the distribution
2998 understands and accepts the risks.
3000 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3003 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3004 fixed (Bugzilla #15330).
3006 * Add support for calling C++11 thread_local object destructors on thread
3007 and program exit. This needs compiler support for offloading C++11
3008 destructor calls to glibc.
3010 * Improved worst case performance of libm functions with double inputs and
3013 * Support for priority inherited mutexes in pthread condition variables on
3014 non-x86 architectures.
3016 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3018 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3020 * Optimized string functions for ARM. Implemented by Will Newton and
3023 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3025 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3028 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3029 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3031 * Added a benchmark framework to track performance of functions in glibc.
3033 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3034 It is based on draft TS 18661 and currently enabled as a GNU extension.
3036 * On Linux, the clock function now uses the clock_gettime system call
3037 for improved precision, rather than old times system call.
3039 * Added support for version-3 tz format files. This is needed when using
3040 the tz database release 2013e or later, and affects a few unusual cases --
3041 currently only TZ='America/Godthab' for time stamps after 2037.
3043 * Added new API functions pthread_getattr_default_np and
3044 pthread_setattr_default_np to get and set the default pthread
3045 attributes of a process.
3047 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3048 This may improve lock scaling of existing programs on TSX capable systems.
3049 When the --enable-lock-elision=yes parameter is specified at configure
3050 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3053 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3054 availability). Implemented by Ryan S. Arnold.
3056 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3060 * The following bugs are resolved with this release:
3062 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3063 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3064 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3065 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3066 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3067 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3068 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3069 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3070 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3071 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3072 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3073 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3074 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3075 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3076 14879, 14889, 14893, 14898, 14914.
3078 * Optimization of memcpy for MIPS.
3080 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3081 EMFILE has been fixed (Bugzilla #14889).
3083 * The add-on ports collection is now distributed in the "ports" subdirectory
3084 of the main GNU C Library distribution, rather than separately.
3086 * Port to ARM AArch64 contributed by Linaro.
3088 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3089 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3091 Implemented by Andreas Krebbel.
3093 * The new function secure_getenv allows secure access to the environment,
3094 returning NULL if running in a SUID/SGID process. This function replaces
3095 the internal function __secure_getenv.
3097 * SystemTap static probes have been added into the dynamic linker.
3098 Implemented by Gary Benson.
3100 * Optimizations of string functions strstr, strcasestr and memmem.
3101 Implemented by Maxim Kuvyrkov.
3103 * The minimum Linux kernel version that this version of the GNU C Library
3104 can be used with is 2.6.16.
3106 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3107 powerpc POWER7. Implemented by Will Schmidt.
3109 * New configure option --disable-nscd builds the C library such that it
3110 never attempts to contact the Name Service Caching Daemon (nscd).
3111 New configure option --disable-build-nscd avoids building nscd itself;
3112 this is the default if --disable-nscd is used.
3114 * Improved support for cross-compilation, including cross-testing and
3115 bootstrap builds without a previously built glibc.
3117 * Several testsuite tests are now able to test multiple IFUNC variants of an
3118 interface, rather than just testing the one that would be chooen by
3121 * New configure options --with-bugurl and --with-pkgversion, for
3122 distributors to use to embed their bug-reporting and package version
3123 information in --help and --version output.
3125 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3126 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3127 allows creation of chroots without the procfs mounted on /proc.
3129 * The `crypt' function now fails if passed salt bytes that violate the
3130 specification for those values. On Linux, the `crypt' function will
3131 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3132 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3133 when the mode is enabled.
3135 * The `clock_*' suite of functions (declared in <time.h>) is now available
3136 directly in the main C library. Previously it was necessary to link with
3137 -lrt to use these functions. This change has the effect that a
3138 single-threaded program that uses a function such as `clock_gettime' (and
3139 is not linked with -lrt) will no longer implicitly load the pthreads
3140 library at runtime and so will not suffer the overheads associated with
3141 multi-thread support in other code such as the C++ runtime library.
3143 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3148 * The following bugs are resolved with this release:
3150 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3151 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3152 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3153 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3154 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3155 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3156 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3157 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3158 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3159 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3160 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3161 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3162 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3163 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3164 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3165 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3166 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3167 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3168 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3169 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3170 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3171 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3174 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3175 configuring glibc with:
3176 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3177 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3178 Implemented by H.J. Lu.
3182 + define static_assert
3184 + do not declare gets
3186 + declare at_quick_exit and quick_exit also for ISO C11
3188 + aligned_alloc. NB: The code is deliberately allows the size parameter
3189 to not be a multiple of the alignment. This is a moronic requirement
3190 in the standard but it is only a requirement on the caller, not the
3193 + timespec_get added
3195 + uchar.h support added
3197 + CMPLX, CMPLXF, CMPLXL added
3199 Implemented by Ulrich Drepper.
3201 * Support for the IA-64 has been moved to ports.
3203 * Remove support for anything but ELF binary format
3205 * Checking versions of poll, ppoll added.
3206 Implemented by Ulrich Drepper.
3208 * More generic and 64-bit performance optimizations to math functions.
3209 Implemented by Ulrich Drepper.
3211 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3212 headers and functions available at compile time as they were before
3213 version 2.14. This option will be removed at some time in the future
3214 after the TI-RPC library becomes fully sufficient for the needs of
3215 existing applications.
3217 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3218 Note that glibc is not expected to work with any Linux kernel version
3221 * New header <sys/auxv.h> and function getauxval allowing easy access to
3222 the AT_* key-value pairs passed from the Linux kernel. The header also
3223 defines the HWCAP_* bits associated with the AT_HWCAP key.
3225 * New locales: mag_IN
3227 * New configure option --enable-systemtap builds SystemTap static probes
3228 into libc for setjmp and longjmp and into libpthread for various operations.
3229 So far the setjmp/longjmp probes and some of the libpthread probes are
3230 provided only for i*86 and x86_64.
3231 Implemented by Roland McGrath and Rayson Ho.
3233 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3235 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3238 * More optimized functions for SPARC. Implemented by David S. Miller.
3240 * Improved support for cross-compilation, especially bootstrap builds
3241 without a previously built glibc.
3243 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3244 Chris Metcalf from Tilera.
3246 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3247 now supported for ARM processors.
3249 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3250 as the name of the dynamic linker, to distinguish it from the
3251 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3253 * Support for CRIS has been removed from ports.
3255 * A new class of installed header has been documented for low-level
3256 platform-specific functionality. PowerPC added the first instance with a
3257 function to provide time base register access. Contributed by Tulio
3258 Magno Quites Machado Filho.
3260 * ix86 configurations now install header files that are consistent with
3261 what x86-64 configurations install. These same header files can be used
3262 for -m32, -m64, or -mx32 builds.
3263 Contributed by H.J. Lu.
3265 * Math library bug fixes. A thorough audit of all open math library bugs was
3266 conducted by Joseph Myers. Significant progress was made on many math
3267 library bugs resulting in more accurate exceptions and function results.
3268 Many thanks to all those that contributed including Andreas Jaeger for his
3269 patch review and work on the x87 trigonometric instruction issues.
3271 * Timezone data is no longer installed. Timezone-related binaries and scripts
3272 will continue to be installed. Users should obtain their timezone data from
3273 their distribution provider or from the tzdata package at
3274 <ftp://munnari.oz.au/pub/>.
3278 * The following bugs are resolved with this release:
3280 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3281 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3282 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3283 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3284 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3285 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3286 13484, 13506, 13515, 13523, 13524, 13538, 13540
3288 * New program pldd to list loaded object of a process
3289 Implemented by Ulrich Drepper.
3291 * Add nss_db support back to glibc. No more dependency on Berkeley db
3292 and support for initgroups lookups.
3293 Implemented by Ulrich Drepper.
3295 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3296 Contributed by HJ Lu.
3298 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3299 Contributed by HJ Lu.
3301 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3302 on x86-32 and x86-64.
3303 Contributed by Liubov Dmitrieva.
3305 * Optimized strchr and strrchr for SSE on x86-32.
3306 Contributed by Liubov Dmitrieva.
3308 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3309 for x86-64 and x86-32.
3310 Contributed by Liubov Dmitrieva.
3312 * New interfaces: scandirat, scandirat64
3313 Implemented by Ulrich Drepper.
3315 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3316 Implemented by Ulrich Drepper.
3318 * nscd now also caches the netgroup database.
3319 Implemented by Ulrich Drepper.
3321 * Integrate libm with gcc's -ffinite-math-only option.
3322 Implemented by Ulrich Drepper.
3324 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3325 to math functions. Implemented by Ulrich Drepper.
3327 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3328 Implemented by Ulrich Drepper.
3330 * New Linux interfaces: process_vm_readv, process_vm_writev
3332 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3333 Implemented by Ulrich Drepper.
3335 * Optimized nearbyint and strcasecmp for PPC.
3336 Implemented by Adhemerval Zanella.
3338 * New locales: bho_IN, unm_US, es_CU, ta_LK
3342 * The following bugs are resolved with this release:
3344 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3345 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3346 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3347 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3348 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3349 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3350 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3351 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3352 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3353 12795, 12811, 12813, 12814, 12841
3355 * The RPC implementation in libc is obsoleted. Old programs keep working
3356 but new programs cannot be linked with the routines in libc anymore.
3357 Programs in need of RPC functionality must be linked against TI-RPC.
3358 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3360 Visible changes of this change include (obviously) the inability to link
3361 programs using RPC functions without referencing the TI-RPC library and the
3362 removal of the RPC headers from the glibc headers.
3363 Implemented by Ulrich Drepper.
3365 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3366 syncfs, setns, sendmmsg
3368 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3369 yue_HK, lij_IT, mhr_RU
3371 * New encodings: CP770, CP771, CP772, CP773, CP774
3373 * New program sotruss to trace calls through PLTs
3374 Implemented by Ulrich Drepper.
3376 * The malloc hook implementation is marked deprecated and will be removed
3377 from the default implementation in the next version. The design never
3378 worked ever since the introduction of threads. Even programs which do
3379 not create threads themselves can use multiple threads created internally.
3383 * The following bugs are resolved with this release:
3385 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3386 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3387 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3388 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3391 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3393 * POWER7 optimizations: memset, memcmp, strncmp
3395 * New optimized string functions for x86-64: strnlen (SSE2),
3396 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3397 Implemented by Ulrich Drepper.
3401 * The following bugs are resolved with this release:
3403 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3404 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3405 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3406 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3407 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3408 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3409 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3410 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3412 * New interfaces: pthread_getname_np, pthread_setname_np
3414 * New Linux interface: recvmmsg
3416 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3418 * The dynamic linker now recognizes supported ABI versions from the
3419 EI_ABIVERSION field in the ELF header.
3420 Implemented by Ulrich Drepper.
3422 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3423 will not be used to fill in password fields in the passwd.byname replies.
3424 Instead it is used to synthesize the shadow.byname table, should it be
3425 missing. This is a useful mode in some installations involving Solaris.
3426 Implemented by Ulrich Drepper.
3428 * New locales: kok_IN, sq_MK, cv_RU
3432 * The following bugs are resolved with this release:
3434 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3435 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3436 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3437 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3438 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3439 10789, 10817, 10823, 10840, 10847
3441 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3443 Implemented by Ulrich Drepper.
3445 * Checking version of longjmp added that fails if an uninitialized stack
3446 frame would be created. Implemented by Ulrich Drepper.
3448 * STT_GNU_IFUNC is now supported in static executables.
3449 Implemented by H.J. Lu.
3451 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3453 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3454 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3455 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3456 Contributed by H.J. Lu.
3458 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3459 Implemented by Ulrich Drepper.
3461 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3463 Contributed by H.J. Lu.
3465 * Support for fma instruction in AVX on x86-64.
3466 Implemented by H.J. Lu and Ulrich Drepper.
3468 * AVX support in x86-64 auditing support in ld.so.
3469 Implemented by H.J. Lu.
3471 * STB_GNU_UNIQUE support added.
3472 Implemented by Ulrich Drepper.
3474 * Implement second fallback mode for optimized DNS lookup for even more
3475 broken environments. If two requests from the same port are not
3476 handled correctly close the socket and open a new one before sending
3477 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3478 can be used to select this mode right away, instead of rediscovering the
3479 necessity is every process again.
3480 Implemented by Ulrich Drepper.
3482 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3483 Implemented by Adam Tkac.
3485 * Optimized iconv conversions for S390x.
3486 Implemented by Andreas Krebbel.
3488 * Using condvars with PI mutexes is now more efficient due to kernel
3489 support for requeueing to PI futexes. NPTL support added for x86-64.
3490 Implemented by Ulrich Drepper.
3492 * New locales: ps_AF, my_MM
3496 * The following bugs are resolved with this release:
3498 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3499 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3500 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3501 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3502 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3504 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3505 Implemented by Ulrich Drepper.
3507 * New Linux interfaces: accept4, fallocate, fallocate64.
3508 Implemented by Ulrich Drepper.
3510 * Correct declarations of string function when used in C++ code. This
3511 could lead to compile errors for invalid C++ code.
3513 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3516 * New POSIX 2008 interface: psiginfo
3517 Implemented by Ulrich Drepper.
3519 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3520 Implemented by Ulrich Drepper.
3522 * Support for selecting between multiple function definitions at runtime
3523 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3525 * The libcrypt library can now use the hash function implementations in
3526 NSS. Implemented by Ulrich Drepper.
3528 * The malloc implementation can be compiled to be less memory efficient
3529 but higher performing in multi-threaded programs.
3530 Implemented by Ulrich Drepper.
3532 * New locales: nan_TW@latin, ks_IN
3534 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3535 Implemented by Ulrich Drepper.
3537 * Extended printf hook support. It is possible to use user-defined types
3538 and extend existing format specifiers.
3539 Implemented by Ulrich Drepper.
3541 * Handling for group shadow files has been added.
3542 Implemented by Ulrich Drepper.
3544 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3545 servers (the case, e.g., for some people using the built-in DNS
3546 server in ADSL modems/routers). There is a once-per-process timeout
3547 in case of a broken server. To avoid it, users can run nscd or put
3548 'options single-request' in /etc/resolv.conf.
3549 Implemented by Ulrich Drepper.
3553 * The following bugs are resolved with this release:
3555 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3556 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3557 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3558 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3559 6968, 6974, 6980, 6995, 7008, 7009, 7029
3561 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3562 up at the same time. Implemented by Ulrich Drepper.
3564 * TLS descriptors for LD and GD on x86 and x86-64.
3565 Implemented by Alexandre Oliva.
3567 * getaddrinfo now handles DCCP and UDPlite.
3568 Implemented by Ulrich Drepper.
3570 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3571 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3572 Implemented by Ulrich Drepper.
3574 * New implementation of memmem, strstr, and strcasestr which is O(n).
3575 Implemented by Eric Blake.
3577 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3579 * Implement "e" option for popen to open file descriptor with the
3580 close-on-exec flag set. Implemented by Ulrich Drepper.
3582 * Implement "b" mode for fmemopen. In this mode writes writes don't
3583 implicitly add a NUL byte and seeks from the end of the buffer really
3584 use the buffer end, not the string length as the basis.
3585 Implemented by Ulrich Drepper.
3587 * Many functions, exported and internal, now atomically set the close-on-exec
3588 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3590 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3592 Implemented by Pravin Satpute.
3594 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3598 * The following bugs are resolved with this release:
3600 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3601 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3602 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3603 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3604 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3605 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3606 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3608 * New locales: bo_CN, bo_IN, shs_CA.
3610 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3612 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3613 Implemented by Pravin Satpute.
3615 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3617 * nscd caches now all timeouts for DNS entries
3618 Implemented by Ulrich Drepper.
3620 * nscd is more efficient and wakes up less often.
3621 Implemented by Ulrich Drepper.
3623 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3624 vdprintf, and obstack_vprintf.
3625 Implemented by Jakub Jelinek.
3627 * Faster memset for x86-64.
3628 Implemented by Harsha Jagasia and H.J. Lu.
3630 * Faster memcpy on x86.
3631 Implemented by Ulrich Drepper.
3633 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3634 Implemented by Ulrich Drepper.
3636 * Faster sqrt and sqrtf implemention for some PPC variants.
3637 Implemented by Steven Munroe.
3641 * The following bugs are resolved with this release:
3643 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3644 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3645 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3646 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3647 5071, 5103, 5104, 5112, 5113, 5184, 5186
3649 * More checking functions: fread, fread_unlocked, open*, mq_open.
3650 Implemented by Jakub Jelinek and Ulrich Drepper.
3652 * Extend fortification to C++. Implemented by Jakub Jelinek.
3654 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3655 by not recognizing 'a' as a modifier when those specs are requested.
3656 Implemented by Jakub Jelinek.
3658 * PPC optimizations to math and string functions.
3659 Implemented by Steven Munroe.
3661 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3662 options to be passed. Implemented by Ulrich Drepper.
3664 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3666 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3667 Implemented by Ulrich Drepper.
3669 * Handle private futexes in the NPTL implementation.
3670 Implemented by Jakub Jelinek and Ulrich Drepper.
3672 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3673 Implemented by Roland McGrath and Ulrich Drepper.
3675 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3677 * SHA-256 and SHA-512 based password encryption.
3678 Implemented by Ulrich Drepper.
3680 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3681 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3684 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3685 Implemented by Ulrich Drepper.
3689 * The following bugs are resolved with this release:
3691 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3692 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3693 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3694 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3695 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3696 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3697 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3700 * New Linux interfaces: epoll_pwait, sched_getcpu.
3702 * New generic interfaces: strerror_l.
3704 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3708 * The following bugs are resolved with this release:
3710 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3711 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3712 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3713 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3714 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3715 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3716 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3717 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3718 3225, 3251, 3252, 3253, 3273
3720 * For Linux, the sorting of addresses returned by getaddrinfo now also
3721 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3722 Implemented by Ulrich Drepper.
3724 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3725 Implemented by Ulrich Drepper.
3727 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3729 * New iconv module for MIK. Contributed by Alexander Shopov.
3731 * For sites with broken group and/or passwd database, the auto-propagate
3732 option of nscd can prevent creating ID lookup entries from the results
3733 of a name lookup and vice versa. This usually is no problem but some
3734 site might have problems with the default behavior.
3735 Implemented by Ulrich Drepper.
3737 * Iterating over entire database in NIS can be slow. With the
3738 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3739 to trade time for memory. The entire database will be read at once.
3740 Implemented by Ulrich Drepper.
3742 * The interfaces introduced in RFC 3542 have been implemented by
3745 * Support for the new ELF hash table format was added by Ulrich Drepper.
3747 * Support for priority inheritance mutexes added by Jakub Jelinek and
3750 * Support for priority protected mutexes added by Jakub Jelinek.
3752 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3753 or_IN, csb_PL, fy_NL, sr_ME.
3757 * More overflow detection functions.
3759 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3760 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3762 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3763 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3764 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3765 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3766 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3767 by Masahide Washizawa.
3769 * It is now possible to install audit libraries for the dynamic linker using
3770 LD_AUDIT. Implemented by Ulrich Drepper.
3772 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3773 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3774 For a libc and libpthread that works well on Linux 2.4 kernels, we
3775 recommend using the stable 2.3 branch.
3777 * The new function open_wmemstream acts like open_memstream,
3778 but uses a wchar_t wide character buffer.
3780 * The new function ppoll is an improved version of poll, similar to pselect.
3782 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3783 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3784 renameat, unlinkat, linkat, symlinkat, readlinkat.
3786 * New Linux kernel system calls: unshare,
3787 inotify_init, inotify_add_watch, inotify_rm_watch.
3789 * The euidaccess function is now also known by the alias eaccess,
3790 for compatibility with some other systems.
3792 * Timezone data updated to 2006b version.
3796 * The following bugs are resolved with this release:
3798 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3799 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3800 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3801 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3802 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3803 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3805 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3807 * As of this release, GCC 4 can be used to compile the C Library.
3809 * Timezone data updated to 2005m version.
3813 * The following bugs are resolved with this release:
3815 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3816 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3817 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3818 777, 787, 821, 822, 823, 825
3820 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3824 * Support for RFC 3678. Real implementations exist only for Linux so far.
3825 Implemented by Ulrich Drepper.
3827 * nscd can now cache entries persistently. Expiring entries are reloaded.
3828 For speedups the cache can be shared in memory with client processes.
3829 Implemented by Ulrich Drepper.
3831 * nscd can now perform SELinux checks.
3832 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3834 * getaddrinfo queries are now cached. Canonical name lookup is performed
3836 Implemented by Ulrich Drepper.
3838 * The nothrow function attribute is used when headers are used by gcc when
3839 compiling C code. This can avoid significant amounts of exception
3842 * The malloc functions perform more error checking and are stricter when
3843 it comes to reacting on errors. The default action is to terminate
3844 the process after showing an error message. Implemented by Ulrich Drepper.
3846 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3847 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3849 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3850 namespaces using the new function dlmopen(). This feature is of course,
3851 like most other dynamic loading functionality, not available in statically
3852 linked applications. Implemented by Ulrich Drepper.
3854 * Low-overhead boundary checking variants of string and some stdio functions
3855 were added. These are to be used in conjunction with a gcc patch by
3856 Jakub Jelinek which adds calls to these functions if possible.
3857 Implemented by Jakub Jelinek and Ulrich Drepper.
3859 * Old code for several operating systems and machine architectures that
3860 have not been in working condition in a long time have been removed from
3861 the main source tree maintained by the GNU C Library's maintainers.
3862 These files are now reside in the separate `ports' source module
3863 that is usable as an add-on when building the library.
3867 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3868 interrogate the dynamic linker, compatible with the Solaris interface.
3870 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3871 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3873 * getifaddrs now uses the netlink interface on Linux to get its information.
3874 Implemented by Thorsten Kukuk.
3876 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3877 Implemented by Ulrich Drepper.
3879 * support for non-executable stacks on x86 has been added. Changes mostly
3882 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3885 * getaddrinfo now performs destination address selection according to
3890 * Thread-safe interfaces for many functions that access locale data
3891 were added in version 2.3, but these features were omitted from NEWS.
3892 Many functions have variants with an `_l' suffix that take a `locale_t'
3893 object as a parameter rather than consulting the current locale.
3894 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3895 create and maintain `locale_t' objects. Additionally, the new function
3896 `uselocale' sets "the current locale" (as used by functions not so
3897 parameterized) set for an individual thread. These features were added
3898 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3900 * The functions getresuid, getresgid, setresuid, and setresgid, which
3901 have long been available on Linux, are now declared in <unistd.h>
3902 and are now also available on the Hurd.
3904 * ELF thread-local storage support (TLS) now works on x86-64.
3906 * The new dynamic string token $LIB is expanded in shared library names.
3907 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3909 * Aldy Hernandez contributed complete software floating point support for
3910 PowerPC machines with no FPU.
3912 * fexecve is implemented on Linux.
3914 * The `btowc' function should work at least twice as fast due to
3915 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3917 * With approriate thread add-ons cancelable functions are now implemented
3918 in libc.so as well. No need to call the function in libpthread. This
3919 change allowed to finally disable the incorrect and expensive handling
3920 of weak definition in ld.so.
3922 * Yet more PLT entries in libc.so have been removed. We finally arrived
3923 at the bare minimum. Startup times improved appropriately.
3925 * Support for the new Linux/x86 system call interface was added. The
3926 AT_SYSINFO auxiliary vector entry is recognized and handled.
3930 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3933 * iconv (the program and the interface) now accepts empty names (excluding
3934 options like //TRANSLIT) to mean "use charset of current locale".
3936 * localedef can now transliterate characters in strings which are not in
3937 the provided charmap. The information from the input locale is used.
3939 * Prelinking support was added for ELF targets. This requires additional
3940 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3942 * Read-only stdio streams now use mmap to speed up operation by eliminating
3943 copying and buffer underflows. To use add 'm' to the mode string of
3944 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3946 * The malloc functions were completely rewritten by Wolfram Gloger based
3947 on Doug Lea's malloc-2.7.0.c.
3949 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3950 implementation of regex.
3952 * Bruno Haible upgraded the iconv and locale implementation to support
3955 * Contents of the LC_* and LANG environment variables in the CEN style are
3956 not recognized anymore. It never was used. Change by Ulrich Drepper.
3958 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3959 thread-local storage (TLS) ABI on some platforms.
3960 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3962 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3963 EUC-JISX0213, and TSCII.
3965 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3966 BSD-compatible interface for getting all network interface addresses.
3967 Implementation for IPv4 by Roland McGrath.
3969 * Loading of locale data is faster due to the introduction of a locale
3970 archive. Implemented by Roland McGrath and Ulrich Drepper.
3972 * Startup times are significantly reduced by not using exported functions
3973 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3976 * Steven Munroe contributed a port to PowerPC64/Linux.
3980 * The Hurd now uses the GNU libio implementation of stdio rather than the
3981 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3983 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3984 with the same behavior as the Linux system call of the same name.
3988 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3989 128-bit long double format.
3991 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3992 IBM1161, and IBM1162 charsets.
3994 * Andreas Jaeger contributed a port to x86-64/Linux.
3996 * Peter Bruin contributed a port to PowerPC/Hurd.
3998 * libc functions using I/O streams now can handle wide-oriented streams
4001 * optimizations in the dynamic linker. Binaries created by recent binutils
4002 versions start up quicker due to reduced time spent on relocations.
4004 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4008 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4009 asinh, atanh, j0 for the 128-bit long double format.
4011 * Bruno Haible updated all the code handling Unicode in some form to
4012 support Unicode 3.1.
4014 * Speed of regex for single-byte locales is back to previous levels.
4015 Patch by Isamu Hasegawa.
4017 * Alpha, SPARC, and IA-64 now also using floating stacks.
4019 * Startup time of internationalized applications greatly improved through
4020 iconv cache. Use iconvconfig to generate the cache file.
4021 Contributed by Ulrich Drepper.
4023 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4024 severe performance problems. Patch by David Mosberger.
4026 * The Hurd port got a lot more functionality like AIO, various stdio
4027 extensions, etc. Mainly done by Roland McGrath.
4029 * mtrace can now lookup symbols in shared libraries.
4033 * Intel's IA-64 math library is largely integrated. It provides fast and
4034 accurate implementatations for most basic and standard math functions
4035 in float, double, and long double format.
4037 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4038 and asin for the 96-bit long double format and asin, log, tan for the
4039 128-bit long double format.
4041 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4042 The basic double functions exist today. Contributed by Abraham Ziv
4043 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4044 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4046 * An asynchronous name lookup library was added. The interface is designed
4047 after POSIX AIO. The proposal was circulated beforehand to get comments.
4048 No negative ones came in. Implemented by Ulrich Drepper.
4050 * Port to S390/64bit contributed by Martin Schwidefsky
4051 <schwidefsky@de.ibm.com>.
4053 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4054 of functions for Linux/IA-64.
4056 * The RPC code is now thread safe. Threads can now use the same service
4057 of different services at the same time. Patch by Eric Norum
4058 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4060 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4061 family of functions for Linux/S390.
4063 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4064 of functions for Linux/x86.
4066 * Port to Linux/CRIS contributed by Axis Communications.
4070 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4071 we can now check for namespace violations and missing declarations. The
4072 result is that almost all headers are now Unix-compliant (as defined in
4073 the upcoming XPG6). The negative side is that some programs might need
4074 corrections, too, if they depend on the incorrect form of the headers in
4075 previous versions which defined too many symbols and included too many
4078 * regex now handles multibyte character sets correctly.
4079 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4081 * iconv (the program) does now conform to the upcoming XPG6 and handles
4082 charmaps. Instead of the charset names the path of charmaps can be
4083 provided and the conversion happens based on this data.
4084 Contributed by Ulrich Drepper.
4086 * The locale program now provides detailed information about the installed
4091 only lists the names of the supported locales
4095 provides details such as country, language, and codeset name.
4096 Contributed by Ulrich Drepper.
4100 * The gencat program now parses the input file according to the charset
4101 selected by the LC_CTYPE category. This is important for stateful
4102 character sets. To make generating catalogs easier there is a way
4103 to overwrite the charset selected by the locale: before the first
4104 message or $ quote line the catalog can contain a line like
4106 $ codeset=ISO-8859-2
4108 to select the charset (ISO-8859-2 in this case).
4110 Implemented by Shinya Hanataka and Ulrich Drepper.
4112 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4113 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4114 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4116 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4117 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4119 * The LANGUAGE environment variable is now ignored unless the locale is
4120 changed from the default "C" locale.
4122 * The usual bug fixes.
4126 * Greg McGary added runtime support for bounds checking using gcc's
4127 new -fbounded-pointers option. ix86 target is complete. PowerPC
4130 * Thorsten Kukuk added secure mode support to nscd.
4132 * The Berkeley db libraries have been removed.
4134 Related, the nss_db module is now in a separate package since it
4135 obviously requires a database library being available.
4137 * Wide character I/O streams implemented by Ulrich Drepper.
4139 * Functions from the extended socket API added by Ulrich Drepper.
4141 * Functions feenableexcept and fedisableexcept to control the
4142 behaviour of individual exceptions have been added by Andreas Jaeger.
4144 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4146 * The resolver code has been updated from bind 8.2.3-T5B which supports
4147 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4150 This change could in some situations effect backward compatibility. Since
4151 now `_res' is a thread-local instead of a global variable, modifying it
4152 in one thread does not have any effect in other threads.
4154 The resolver library was also extended to allow IPv6 as the transport
4155 protocol for the requests. This work was done by Stig Venaas.
4157 * Compatibility code for K&R C compilers has been removed from the
4158 header files. A ISO C compiler is needed to use the library
4159 (conforming to either C89 or C99 standard).
4161 * Complete rewrite of the localedef program to support multibyte character
4162 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4163 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4164 Implemented by Ulrich Drepper.
4166 Bruno Haible significantly improved the generation and use of the data
4167 structures for the wide character tables.
4169 * Plural handling in gettext implemented by Ulrich Drepper.
4171 * The utmp daemon has been removed.
4173 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4175 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4178 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4180 * POSIX barriers implemented by Kaz Kylheku.
4182 * POSIX spawn function family implemented by Ulrich Drepper.
4184 * POSIX spinlocks are now available.
4186 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4188 * the configure option --enable-kernel=X.Y.Z allows to strip out
4189 compatibility for kernel versions before X.Y.Z. This is currently only
4190 implemented for Linux.
4192 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4193 field sin6_scope_id. This means that all programs using IPv6 should be
4194 recompiled. Don't expect binary compatibility with previous glibc
4197 * various conversion modules for IBM character sets contributed by
4200 * IA-64 port by Jes Sorensen and HJ Lu.
4214 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4216 * New xdr functions are added; some rpc functions are now 64bit clean.
4218 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4220 * Fixed known incompatibilities with glibc 2.0.
4222 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4224 * Optimized a number of functions (especially the ELF dynamic loader).
4226 * Update timezone data files.
4228 * lots of charmaps corrections
4230 * some new locale definitions and charmaps
4235 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4236 conform to POSIX member type requirements. Symbol versions have been
4237 adjusted as needed within the library, and for direct use by applications,
4238 but there is potential for problems if third-party libraries use
4239 struct timeval as part of their interface. This does not present
4240 a problem for X and other "essential" system libraries.
4242 * An additional locale model to support C++ Standard Library locale
4243 model and probably more was implemented by Ulrich Drepper.
4245 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4248 * Miles Bader provided the `argp' function family to support hierachical
4249 command line argument parsing, layered on top of getopt.
4251 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4253 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4256 * scanf recognizes the %a and %A format for scanning floating point numbers.
4258 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4259 information and interfaces for the available integer types.
4261 * about 130 new math functions were added to implement the ISO C9x math
4264 * the new header <complex.h> contains definitions of the complex math
4265 functions from ISO C 9X.
4267 * the new header <tgmath.h> defines generic macros to use complex or
4268 real valued functions.
4270 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4272 * Andreas Jaeger provided a test suite for the math library.
4274 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4276 * Ulrich Drepper added character set conversion functions (iconv).
4278 * Optimized string functions have been added.
4280 * The localedata addon is now part of glibc.
4282 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4284 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4285 daemon for NSS (nscd).
4287 Missing a better place here are some numbers on improvements. Under
4288 Linux 2.1.125 un-tar-ing the kernel sources takes
4292 using local files 12.19s 6.88s 22.91s
4294 using NIS 13.92s 8.91s 26.34s
4296 using NIS & nscd 10.37s 7.34s 25.30s
4298 using NIS+ 27.57s 30.37s 640.46s
4300 using NIS+ & nscd 10.25s 7.83s 26.51s
4302 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4304 Keep in mind that non-namelookup related operations dominate above times.
4305 It was just a common complain that using NIS+ unpacking the kernel is
4308 [1] The old nscd implementation is not available anymore since it was
4309 distributed with glibc up to version 2.0.98 and thus is now replaced.
4311 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4313 * Mark Kettenis provided a Hesiod NSS module.
4315 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4316 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4318 * rcmd can now handle netgroups (Dick Streefland).
4320 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4323 * Support for the IPv6 protocol has been added to the socket API, as per the
4324 latest draft standards.
4326 * Support for Linux 2.2 has been added.
4328 * Interface changes relative to the latest 2.0.x release:
4329 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4330 addseverity NEW: Unix98
4331 alphasort64 NEW: LFS
4332 argp_err_exit_status NEW: argp, GNU ext
4333 argp_error NEW: argp, GNU ext
4334 argp_failure NEW: argp, GNU ext
4335 argp_help NEW: argp, GNU ext
4336 argp_parse NEW: argp, GNU ext
4337 argp_program_bug_address NEW: argp, GNU ext
4338 argp_program_version NEW: argp, GNU ext
4339 argp_program_version_hook NEW: argp, GNU ext
4340 argp_state_help NEW: argp, GNU ext
4341 argp_usage NEW: argp, GNU ext
4342 authdes_create NEW: Secure RPC
4343 authdes_getucred NEW: Secure RPC
4344 authdes_pk_create NEW: Secure RPC
4345 backtrace NEW: GNU ext.
4346 backtrace_symbols NEW: GNU ext.
4347 backtrace_symbols_fd NEW: GNU ext.
4349 cacosf NEW: ISO C 9x
4350 cacosh NEW: ISO C 9x
4351 cacoshf NEW: ISO C 9x
4352 cacoshl NEW: ISO C 9x
4353 cacosl NEW: ISO C 9x
4360 casinf NEW: ISO C 9x
4361 casinh NEW: ISO C 9x
4362 casinhf NEW: ISO C 9x
4363 casinhl NEW: ISO C 9x
4364 casinl NEW: ISO C 9x
4366 catanf NEW: ISO C 9x
4367 catanh NEW: ISO C 9x
4368 catanhf NEW: ISO C 9x
4369 catanhl NEW: ISO C 9x
4370 catanl NEW: ISO C 9x
4371 cbc_crypt NEW: Secure RPC
4375 ccoshf NEW: ISO C 9x
4376 ccoshl NEW: ISO C 9x
4382 cimagf NEW: ISO C 9x
4383 cimagl NEW: ISO C 9x
4384 clearerr_locked REMOVED
4385 clntunix_create NEW: sunrpc ext
4387 clog10 NEW: ISO C 9x
4388 clog10f NEW: ISO C 9x
4389 clog10l NEW: ISO C 9x
4399 cprojf NEW: ISO C 9x
4400 cprojl NEW: ISO C 9x
4402 crealf NEW: ISO C 9x
4403 creall NEW: ISO C 9x
4408 csinhf NEW: ISO C 9x
4409 csinhl NEW: ISO C 9x
4412 csqrtf NEW: ISO C 9x
4413 csqrtl NEW: ISO C 9x
4417 ctanhf NEW: ISO C 9x
4418 ctanhl NEW: ISO C 9x
4420 des_setparity NEW: Secure RPC
4421 ecb_crypt NEW: Secure RPC
4422 endutxent NEW: Unix98
4424 exp10f NEW: ISO C 9x
4425 exp10l NEW: ISO C 9x
4429 fattach NEW: STREAMS
4430 fdetach NEW: STREAMS
4434 feclearexcept NEW: ISO C 9x
4435 fegetenv NEW: ISO C 9x
4436 fegetexceptflag NEW: ISO C 9x
4437 fegetround NEW: ISO C 9x
4438 feholdexcept NEW: ISO C 9x
4440 feraiseexcept NEW: ISO C 9x
4441 ferror_locked REMOVED
4442 fesetenv NEW: ISO C 9x
4443 fesetexceptflag NEW: ISO C 9x
4444 fesetround NEW: ISO C 9x
4445 fetestexcept NEW: ISO C 9x
4446 feupdateenv NEW: ISO C 9x
4447 fflush_locked REMOVED
4451 fgets_unlocked NEW: GNU ext.
4452 fileno_locked REMOVED
4464 fputc_locked REMOVED
4465 fputs_unlocked NEW: GNU ext.
4466 fread_unlocked NEW: GNU ext.
4471 fstatvfs NEW: Unix98
4475 ftruncate64 NEW: LFS
4477 fwrite_unlocked NEW: GNU ext.
4478 gai_strerror NEW: IPv6
4482 getchar_locked REMOVED
4484 getdate_err NEW: Unix98
4485 getdate_r NEW: GNU ext.
4487 getnameinfo NEW: IPv6
4488 getnetname NEW: Secure RPC
4489 getpmsg NEW: STREAMS
4490 getpt NEW: Unix98 PTY
4491 getrlimit64 NEW: LFS
4492 getutxent NEW: Unix98
4493 getutxid NEW: Unix98
4494 getutxline NEW: Unix98
4495 glob64 NEW: GNU ext.
4496 globfree64 NEW: GNU ext.
4497 gnu_get_libc_release NEW: GNU ext.
4498 gnu_get_libc_version NEW: GNU ext.
4499 grantpt NEW: Unix98 PTY
4500 host2netname NEW: Secure RPC
4502 iconv_close NEW: iconv
4503 iconv_open NEW: iconv
4504 if_freenameindex NEW: IPv6
4505 if_indextoname NEW: IPv6
4506 if_nameindex NEW: IPv6
4507 if_nametoindex NEW: IPv6
4508 in6addr_any NEW: IPv6
4509 in6addr_loopback NEW: IPv6
4510 inet6_isipv4mapped NEW: IPv6
4511 isastream NEW: STREAMS
4512 iswblank NEW: GNU ext.
4513 key_decryptsession NEW: Secure RPC
4514 key_decryptsession_pk NEW: Secure RPC
4515 key_encryptsession NEW: Secure RPC
4516 key_encryptsession_pk NEW: Secure RPC
4517 key_gendes NEW: Secure RPC
4518 key_get_conv NEW: Secure RPC
4519 key_secretkey_is_set NEW: Secure RPC
4520 key_setnet NEW: Secure RPC
4521 key_setsecret NEW: Secure RPC
4522 llrint NEW: ISO C 9x
4523 llrintf NEW: ISO C 9x
4524 llrintl NEW: ISO C 9x
4525 llround NEW: ISO C 9x
4526 llroundf NEW: ISO C 9x
4527 llroundl NEW: ISO C 9x
4532 lrintf NEW: ISO C 9x
4533 lrintl NEW: ISO C 9x
4534 lround NEW: ISO C 9x
4535 lroundf NEW: ISO C 9x
4536 lroundl NEW: ISO C 9x
4538 makecontext NEW: Unix98
4539 mempcpy NEW: GNU ext.
4542 modify_ldt NEW: kernel
4546 nearbyint NEW: ISO C 9x
4547 nearbyintf NEW: ISO C 9x
4548 nearbyintl NEW: ISO C 9x
4549 netname2host NEW: Secure RPC
4550 netname2user NEW: Secure RPC
4551 nexttoward NEW: ISO C 9x
4552 nexttowardf NEW: ISO C 9x
4553 nexttowardl NEW: ISO C 9x
4557 passwd2des NEW: Secure RPC
4559 pow10f NEW: GNU ext.
4560 pow10l NEW: GNU ext.
4563 printf_size NEW: GNU ext.
4564 printf_size_info NEW: GNU ext.
4565 profil_counter REMOVED
4566 pthread_mutexattr_getkind_np REPLACED
4567 pthread_mutexattr_setkind_np REPLACED
4568 ptsname NEW: Unix98 PTY
4569 ptsname_r NEW: Unix98 PTY
4571 putchar_locked REMOVED
4572 putgrent NEW: GNU ext.
4574 putpmsg NEW: STREAMS
4575 pututxline NEW: Unix98
4579 readdir64_r NEW: LFS
4580 remquo NEW: ISO C 9x
4581 remquof NEW: ISO C 9x
4582 remquol NEW: ISO C 9x
4584 roundf NEW: ISO C 9x
4585 roundl NEW: ISO C 9x
4587 scalbln NEW: ISO C 9x
4588 scalblnf NEW: ISO C 9x
4589 scalblnl NEW: ISO C 9x
4591 sendfile NEW: kernel
4592 setcontext NEW: Unix98
4593 setrlimit64 NEW: LFS
4594 setutxent NEW: Unix98
4596 sigignore NEW: Unix98
4597 sigqueue NEW: POSIX.1b
4598 sigrelse NEW: Unix98
4599 sigset NEW: POSIX.1b
4600 sigtimedwait NEW: POSIX.1b
4601 sigwaitinfo NEW: POSIX.1b
4602 sincos NEW: GNU ext.
4603 sincosf NEW: GNU ext.
4604 sincosl NEW: GNU ext.
4608 strcasestr NEW: GNU ext.
4609 strtoimax NEW: ISO C 9x
4610 strtoumax NEW: ISO C 9x
4611 strverscmp NEW: GNU ext.
4612 svcauthdes_stats NEW: Secure RPC
4613 svcunix_create NEW: sunrpc etx
4614 svcunixfd_create NEW: sunrpc ext
4615 swapcontext NEW: Unix98
4616 tcgetsid NEW: Unix98 PTY
4617 tdestroy NEW: GNU ext.
4618 tgamma NEW: ISO C 9x
4619 tgammaf NEW: ISO C 9x
4620 tgammal NEW: ISO C 9x
4624 truncf NEW: ISO C 9x
4625 truncl NEW: ISO C 9x
4627 unlockpt NEW: Unix98 PTY
4628 updwtmpx NEW: Unix98
4629 user2netname NEW: Secure RPC
4630 utmpxname NEW: Unix98
4631 versionsort NEW: GNU ext.
4632 versionsort64 NEW: GNU ext.
4634 wcscasecmp NEW: GNU ext.
4635 wcsncasecmp NEW: GNU ext.
4636 wcsnlen NEW: GNU ext.
4637 wcstoimax NEW: ISO C 9x
4638 wcstoll NEW: ISO C 9x
4639 wcstoull NEW: ISO C 9x
4640 wcstoumax NEW: ISO C 9x
4642 wordexp NEW: POSIX.2
4643 wordfree NEW: POSIX.2
4644 write_profiling REMOVED
4645 xdecrypt NEW: Secure RPC
4646 xdr_authdes_cred NEW: Secure RPC
4647 xdr_authdes_verf NEW: Secure RPC
4648 xdr_cryptkeyarg NEW: Secure RPC
4649 xdr_cryptkeyarg2 NEW: Secure RPC
4650 xdr_cryptkeyres NEW: Secure RPC
4651 xdr_getcredres NEW: Secure RPC
4652 xdr_key_netstarg NEW: Secure RPC
4653 xdr_key_netstres NEW: Secure RPC
4654 xdr_keybuf NEW: Secure RPC
4655 xdr_keystatus NEW: Secure RPC
4656 xdr_netnamestr NEW: Secure RPC
4657 xdr_sizeof NEW: Secure RPC
4658 xdr_unixcred NEW: sunrpc ext
4659 xencrypt NEW: Secure RPC
4660 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4671 * inet_ntoa is thread-safe
4673 * updwtmp is moved from libutil to libc
4675 * rewrite of cbrt function
4677 * update of timezone data
4691 * add atoll function
4693 * fix complex problems in Berkeley DB code
4695 * fix math functions
4699 * fixed lots of header problems (especially Linux/GNU specific)
4701 * dynamic loader preserves all registers
4703 * Roland McGrath provided support for handling of auxiliary objects in
4704 the ELF dynamic loader.
4706 * support for parallel builds is improved
4710 * GNU extensions are no longer declared by default. To enable them you
4711 must define the macro `_GNU_SOURCE' in your program or compile with
4714 * The library has changed from using GNU ld symbol aliases to using weak
4715 symbols where available. The ELF object file format supports weak
4716 symbols; GNU ld also supports weak symbols in the a.out format. (There
4717 is also now support for other GNU ld extensions in ELF. Use the
4718 `--with-elf' option to configure to indicate you have ELF, and
4719 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4720 of many files which contained only symbol aliases, reducing the size of
4721 the source and the compiled library; many other files were renamed to
4722 less cryptic names previously occupied by the symbol alias files.
4723 There is a new header file <elf.h> for programs which operate on
4724 files in the ELF format.
4726 * Converted to Autoconf version 2, so `configure' has more options.
4727 Run `configure --help' to see the details.
4729 * The library can now be configured to build profiling, highly-optimized
4730 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4731 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4732 `configure' enable building these extra libraries. The shared library is
4733 built by default when using both ELF and GNU ld. When shared libraries
4734 are enabled, the new library `-ldl' is available for arbitrary run-time
4735 loading of shared objects; its interface is defined in <dlfcn.h>. The
4736 new header file <link.h> gives access to the internals of the run-time
4737 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4738 application of same name on other systems and it provides information
4739 about dynamically linked binaries.
4741 * The C library now provides the run-time support code for profiling
4742 executables compiled with `-pg'. Programs can control the profiling code
4743 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4744 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4745 the support for this file format was contributed by David Mosberger-Tang.
4747 * The math code has been replaced with a math library based on fdlibm from
4748 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4749 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4750 functions. The math functions now reside in a separate library, so
4751 programs using them will need to use `-lm' their linking commands.
4753 * John C. Bowman contributed optimized ix87 assembler inline functions.
4755 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4756 mechanism similar to that found in Solaris 2. This is now used for the
4757 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4758 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4759 file controls what services are used for each individual database. This
4760 works by loading shared libraries with names specified in `nsswitch.conf',
4761 so service modules can be changed or added at any time without even
4762 relinking any program. Currently there are the file, db, and NIS based
4763 NSS services available.
4765 * The new functions `strtoq' and `strtouq' parse integer values from
4766 strings, like `strtol' and `strtoul', but they return `long long int' and
4767 `unsigned long long int' values, respectively (64-bit quantities).
4769 * The new functions `strtof' and `strtold' parse floating-point values from
4770 strings, like `strtod', but they return `float' and `long double' values,
4771 respectively (on some machines `double' and `long double' are the same).
4773 * Ulrich Drepper has contributed new implementations of the floating-point
4774 printing and reading code used in the `printf' family of functions and
4775 `strtod', `strtof', and `strtold'. These new functions are perfectly
4776 accurate, and much faster than the old ones.
4778 * The implementation of the POSIX locale model was completely rewritten by
4779 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4780 to compile the POSIX locale definition.
4782 * The former dummy implementations of the strcoll and strxfrm function are
4783 now replaced by fully functional code contributed by Ulrich Drepper. The
4784 collation information comes from the POSIX locale definitions.
4786 * The new header <langinfo.h> defines an interface for accessing
4787 various locale-dependent data (using the locale chosen with `setlocale').
4789 * Ulrich Drepper has contributed a new suite of functions for operation on
4790 wide-character and multibyte-character strings, in <wchar.h>;
4791 and classification and case conversion of wide characters, in <wctype.h>.
4792 These new functions are conforming to the ISO C, Amendement 1 specification.
4794 * There is now a second implementation of the standard I/O library available.
4795 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4796 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4798 * You can now use positional parameter specifications in format strings
4799 for the `printf' and `scanf' families of functions. For example,
4800 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4801 ``Number 6, Mr Jones''. This is mainly useful when providing different
4802 format strings for different languages, whose grammars may dictate
4803 different orderings of the values being printed. To support this
4804 feature, the interface for `register_printf_handler' has changed; see
4805 the header file <printf.h> for details.
4807 * The `printf' and `scanf' families of functions now understand a new
4808 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4809 to group numbers as indicated by the locale; for `scanf' and friends, this
4810 says to accept as valid only a number with all the proper grouping
4811 separators in the right places. In the default "C" locale, numbers are
4812 not grouped; but locales for specific countries will define the usual
4813 conventions (i.e. separate thousands with `,' in the US locale).
4815 * The pgrp functions have been regularized, slightly incompatibly but much
4816 less confusingly. The core functions are now `getpgid' and `setpgid',
4817 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4818 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4819 provided for compatibility. There is no longer an incompatible `getpgrp'
4820 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4822 * The new header file <fts.h> and suite of functions simplify programs that
4823 operate on directory trees. This code comes from 4.4 BSD.
4825 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4826 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4827 NSS scheme used in glibc.
4829 * There is a new malloc debugging hook `__memalign_hook'.
4831 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4832 `unsigned int' in <sys/types.h>. These are for compatibility only and
4833 their use is discouraged.
4835 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4836 done differently, so that it works even without GNU ld.
4838 * New function `euidaccess' checks allowed access to a file like `access',
4839 but using the effective IDs instead of the real IDs.
4841 * The time zone data files have been updated for the latest and greatest
4842 local time conventions of the countries of the world.
4844 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4847 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4848 for formatting floating-point numbers. They are provided only for
4849 compatibility; new programs should use `sprintf' instead. There are
4850 also equivalent function for the `long double' floating-point type and
4851 all functions also exist in a reentrant form.
4853 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4854 functions for maintaining the login-record files (primarily of use to
4855 system programs such as `login'), and convenient functions for
4856 allocating and initializing a pseudo-terminal (pty) device.
4858 * Ulrich Drepper has contributed new support for System V style
4859 shared memory and IPC on systems that support it.
4861 * Ulrich Drepper has contributed several miscellaneous new functions found
4862 in System V: The `hsearch' family of functions provide an effective
4863 implementation of hash tables; `a64l' and `l64a' provide a very simple
4864 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4867 * Ulrich Drepper has contributed new reentrant counterparts for the
4868 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4870 * Ulrich Drepper has contributed new, highly-optimized versions of several
4871 string functions for the i486/Pentium family of processors.
4873 * Ulrich Drepper has updated the Linux-specific code, based largely
4874 on work done in Hongjiu Lu's version of GNU libc for Linux.
4875 The GNU library now supports Linux versions 2.0.10 and later,
4876 using the ELF object file format (i[3456]86-*-linux).
4878 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4880 * David Mosberger-Tang and Richard Henderson have ported the C library
4881 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4882 dynamic linking support for ELF/Alpha.
4884 * Richard Henderson contributed several Alpha optimized assembler function
4885 for arithmetic and string handling.
4887 * Ulrich Drepper has contributed a new set of message catalog functions to
4888 support multiple languages using the <libintl.h> interface, for use with
4889 his new package GNU gettext. Translation volunteers have contributed
4890 catalogs of the library's messages in Spanish, German, and Korean.
4892 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4893 program and the `catgets' function for reading the catalog files it
4894 creates. (The <libintl.h> interface is preferred; we include the
4895 <nl_types.h> interface using `catgets' only for source compatibility with
4896 programs already written to use it.)
4898 * New header file <values.h> gives SVID-compatible names for <limits.h>
4901 * Various new macros, declarations, and small header files for compatibility
4904 * New function `group_member' is a convenient way to check if a process has
4905 a given effective group ID.
4907 * When using GCC 2.7 and later, the socket functions are now declared in a
4908 special way so that passing an argument of type `struct sockaddr_in *',
4909 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4910 `struct sockaddr *' type, does not generate a type-clash warning.
4912 * New function `error' declared in header file <error.h> is a convenient
4913 function for printing error messages and optionally exiting; this is the
4914 canonical function used in GNU programs. The new functions `err', `warn',
4915 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4916 doing the same thing.
4918 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4919 POSIX.2 `glob' function to do ~ and {...} expansion.
4921 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4922 `clearenv' which is used in POSIX.9 is also available.
4924 * New function `getsid' returns session ID number on systems that support it.
4926 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4927 files <db.h> and <mpool.h> provide a rich set of functions for several
4928 types of simple databases stored in memory and in files, and <ndbm.h> is
4929 an old `ndbm'-compatible interface using the `db' functions. Link with
4930 `-ldb' to get these functions.
4932 * New macro `strdupa' copies a string like `strdup', but uses local stack
4933 space from `alloca' instead of dynamic heap space from `malloc'.
4935 * New function `strnlen' is like `strlen' but searches only a given maximum
4936 number of characters for the null terminator. `stpncpy', `strndup' and
4937 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4940 * New function `statfs' in header <sys/statfs.h>.
4942 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4943 provide convenient functions for operating on blocks of null-terminated
4946 * A new suite of functions in <utmp.h> handle all the details of reading
4947 and writing the utmp file.
4949 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4952 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4953 completely POSIX compliant and also implemented the extended functionality
4954 to handle alternate digit representation and alternate era date formats.
4956 * Ulrich Drepper provided an implementation of the `strptime' function
4957 defined in XPG4.2 which transforms a string into a `struct tm' value.
4959 * Paul Eggert provided the tzselect shell script as part of the timezone
4960 code. The shell script makes it easy to select the correct timezone
4963 * The implementation of the malloc family of functions is completely replaced
4964 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4965 The implementation uses the mmap function (if available) and it is
4966 optimized for the use in multi threaded programs.
4968 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4969 functions. This new functionality is usable by specifying a special
4970 salt string and it is compatible with implementation on *BSD systems.
4972 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4973 `getsubopt' to handle second level command line options, `bsd_signal'
4974 to access BSD style `signal' functionality, the obsolete `regexp' style
4977 * the `lchown' function is available on system which support this
4980 * The implementation of the shadow password handling function was contributed
4983 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4985 * POSIX.1g support was added. The <sys/select.h> header is available,
4986 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4987 implementation of `getaddrinfo'.
4991 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4993 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4994 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4996 * The new function `daemon' from 4.4 BSD is useful for server programs that
4997 want to put themselves in the background.
4999 * Joel Sherrill has contributed support for several standalone boards that
5000 run without an operating system.
5002 * `printf', `scanf' and friends now accept a `q' type modifier for long
5003 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5005 * All of the code taken from BSD (notably most of the math and networking
5006 routines) has been updated from the BSD 4.4-Lite release.
5008 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5010 * The new functions `getdomainname' and `setdomainname' fetch or change the
5011 YP/NIS domain name. These are system calls which exist on systems which
5014 * The time zone data files have been updated for the latest international
5017 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5018 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5022 * The C library now includes support for Sun RPC, from Sun's free
5023 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5024 are included. (There is still no support for YP.)
5026 * Tom Quinn has contributed a port of the C library to SGI machines running
5027 Irix 4 (mips-sgi-irix4).
5029 * The new `lockf' function is a simplified interface to the locking
5030 facilities of `fcntl', included for compatibility.
5032 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5034 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5036 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5039 * The error message printed by `assert' for a failed assertion now includes
5040 the name of the program (if using GNU ld) and the name of the calling
5041 function (with versions of GCC that support this).
5043 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5045 * The library now includes the <sys/mman.h> header file and memory
5046 management functions `mmap', `munmap', `mprotect', `msync', and
5047 `madvise', on systems that support those facilities.
5049 * The interface for `mcheck' has changed slightly: the function called to
5050 abort the program when an allocation inconsistency is detected now takes
5051 an argument that indicates the type of failure. The new function
5052 `mprobe' lets you request a consistency check for a particular block at
5053 any time (checks are normally done only when you call `free' or `realloc'
5056 * It is now possible to easily cross-compile the C library, building on one
5057 system a library to run on another machine and/or operating system. All
5058 you need to do is set the variable `HOST_CC' in `configparms' to the
5059 native compiler for programs to run on the machine you are building on (a
5060 few generator programs are used on Unix systems); set `CC' to the
5063 * The new function `fexecve' (only implemented on the GNU system) executes
5064 a program file given a file descriptor already open on the file.
5068 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5069 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5071 * You can set the variable `libprefix' in `configparms' to specify a prefix
5072 to be prepended to installed library files; this makes it easy to install
5073 the GNU C library to be linked as `-lgnuc' or whatever.
5075 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5076 copies a limited number of characters from a string, and returns the
5077 address of the last character written.
5079 * You no longer need to check for whether the installed `stddef.h' is
5080 compatible with the GNU C library. configure now checks for you.
5082 * You can now define a per-stream `fileno' function to convert the
5083 stream's cookie into an integral file descriptor.
5085 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5086 allocates zero bytes of storage, and returns a unique pointer which
5087 you can pass to `realloc' or `free'. The behavior is undefined if
5088 you dereference this pointer.
5090 * The C library now runs on Sony NEWS m68k machines running either
5091 NewsOS 3 or NewsOS 4.
5093 * The new `syscall' function is a system-dependent primitive function
5094 for invoking system calls. It has the canonical behavior on Unix
5095 systems, including unreliable return values for some calls (such as
5096 `pipe', `fork' and `getppid').
5098 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5099 to `EAGAIN', which is the preferred name. On systems whose kernels
5100 use two distinct codes, the C library now translates EWOULDBLOCK to
5101 EAGAIN in every system call function.
5105 * The GNU C Library Reference Manual is now distributed with the library.
5106 `make dvi' will produce a DVI file of the printed manual.
5107 `make info' will produce Info files that you can read on line using C-h i
5108 in Emacs or the `info' program.
5109 Please send comments on the manual to bug-glibc-manual@gnu.org.
5111 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5113 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5115 * Jason Merrill has contributed a port to the Sequent Symmetry running
5116 Dynix version 3 (i386-sequent-dynix).
5118 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5119 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5121 * New function `memory_warnings' lets you arrange to get warnings when
5122 malloc is running out of memory to allocate, like Emacs gives you.
5124 * The C library now contains the relocating allocator used in Emacs 19 for
5125 its editing buffers. This allocator (ralloc) minimizes allocation
5126 overhead and fragmentation by moving allocated regions around whenever it
5127 needs to. You always refer to a ralloc'd region with a "handle" (a
5128 pointer to a pointer--an object of type `void **').
5130 * There is a new `printf' format: `%m' gives you the string corresponding
5131 to the error code in `errno'.
5133 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5134 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5135 pass, the `a' modifier says to fill in a `char **' you pass with a
5138 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5139 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5140 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5142 * `mkstemp' is a traditional Unix function to atomically create and open a
5143 uniquely-named temporary file.
5147 * The standard location for the file that says what the local timezone is
5148 has changed again. It is now `/usr/local/etc/localtime' (or more
5149 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5151 * The distribution no longer contains any files with names longer than 14
5154 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5155 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5157 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5161 * The configuration process has changed quite a bit. The `configure'
5162 script is now used just like the configuration scripts for other GNU
5163 packages. The `sysdeps' directory hierarchy is much rearranged.
5164 The file `INSTALL' explains the new scheme in detail.
5166 * The header files no longer need to be processed into ANSI C and
5167 traditional C versions. There is just one set of files to install, and
5168 it will work with ANSI or old C compilers (including `gcc -traditional').
5170 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5171 MIPS DECStation running Ultrix 4.
5173 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5174 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5175 made itself into a shared library.
5177 * Yet further improved support for the i386, running 4.3 BSD-like systems
5178 (such as Mach 3 with the Unix single-server), or System V.
5180 * New function `strncasecmp' to do case-insensitive string comparison
5181 with limited length.
5183 * New function `strsep' is a reentrant alternative to `strtok'.
5185 * New functions `scandir' and `alphasort' for searching directories.
5187 * New function `setenv' is a better interface to `putenv'.
5189 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5190 function for traversing a directory tree.
5192 * The GNU obstack package is now also part of the C library.
5193 The new function `open_obstack_stream' creates a stdio stream that
5194 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5195 formatted output directly to an obstack.
5197 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5198 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5200 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5202 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5203 things to your strings.
5205 * There are some new test programs: test-fseek, testmb, and testrand.
5207 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5208 These ports are not finished, but are a good starting place for really
5209 supporting those systems.
5211 * `/etc/localtime' is now the standard location for the file that says what
5212 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5213 This follows the general principle that `/etc' is the place for all local
5214 configuration files.
5216 * The C library header files now use `extern "C"' when used by the C++
5217 compiler, so the C library should now work with C++ code.
5219 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5220 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5223 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5224 function which uses the `merge sort' algorithm, and is said to be
5225 significantly faster than the old GNU `qsort' function. Merge sort is
5226 now the standard `qsort' function. The new algorithm can require a lot
5227 of temporary storage; so, the old sorting function is called when the
5228 required storage is not available.
5230 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5231 provides the Unix `crypt' function, plus some other entry points.
5233 * The code and header files taken from 4.4 BSD have been updated with the
5234 latest files released from Berkeley.
5236 ----------------------------------------------------------------------
5237 Copyright information:
5239 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5241 Permission is granted to anyone to make or distribute verbatim copies
5242 of this document as received, in any medium, provided that the
5243 copyright notice and this permission notice are preserved,
5244 thus giving the recipient permission to redistribute in turn.
5246 Permission is granted to distribute modified versions
5247 of this document, or of portions of it,
5248 under the above conditions, provided also that they
5249 carry prominent notices stating who last changed them.
5252 version-control: never