1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2018 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 * A new convenience target has been added for distribution maintainers
13 to build and install all locales as directories with files. The new
14 target is run by issuing the following command in your build tree:
15 'make localedata/install-locale-files', with an optional DESTDIR
16 to set the install root if you wish to install into a non-default
19 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
21 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
22 for _GNU_SOURCE, to match BSD environments.
24 Deprecated and removed features, and other changes affecting compatibility:
26 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
27 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
29 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
30 in <sys/procfs.h>, has been corrected to match the type actually used by
31 the Linux kernel. This affects the size and layout of that structure on
32 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
34 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
35 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
36 defined in <sys/procfs.h>, has been corrected to match the type actually
37 used by the Linux kernel. This affects the size and layout of those
40 Changes to build and runtime requirements:
42 [Add changes to build and runtime requirements here]
44 Security related changes:
46 [Add security related changes here]
48 The following bugs are resolved with this release:
50 [The release manager will add the list generated by
51 scripts/list-fixed-bugs.py just before the release.]
58 * The localization data for ISO 14651 is updated to match the 2016
59 Edition 4 release of the standard, this matches data provided by
60 Unicode 9.0.0. This update introduces significant improvements to the
61 collation of Unicode characters. This release deviates slightly from
62 the standard in that the collation element ordering for lowercase and
63 uppercase LATIN script characters is adjusted to ensure that regular
64 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
65 is not matched by [a-z]. With the update many locales have been
66 updated to take advantage of the new collation information. The new
67 collation information has increased the size of the compiled locale
68 archive or binary locales.
70 * The GNU C Library can now be compiled with support for Intel CET, AKA
71 Intel Control-flow Enforcement Technology. When the library is built
72 with --enable-cet, the resulting glibc is protected with indirect
73 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
74 compatible with all existing executables and shared libraries. This
75 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
76 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
77 capable of multi-byte NOPs, like x86-64 processors as well as Intel
78 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
79 x86_64 and x32 on non-CET processors. --enable-cet has been tested
80 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
83 * The GNU C Library now has correct support for ABSOLUTE symbols
84 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
85 relocated incorrectly or in some cases discarded. The GNU linker can
86 make use of the newer semantics, but it must communicate it to the
87 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
88 field) to indicate such support is required.
90 * Unicode 11.0.0 Support: Character encoding, character type info, and
91 transliteration tables are all updated to Unicode 11.0.0, using
92 generator scripts contributed by Mike FABIAN (Red Hat).
94 * <math.h> functions that round their results to a narrower type are added
95 from TS 18661-1:2014 and TS 18661-3:2015:
97 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
100 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
103 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
106 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
109 * Two grammatical forms of month names are now supported for the following
110 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
111 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
112 support two grammatical forms in abbreviated month names: Catalan, Greek,
115 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
116 include the support for two grammatical forms of month names.
118 * Building and running on GNU/Hurd systems now works without out-of-tree
121 * The renameat2 function has been added, a variant of the renameat function
122 which has a flags argument. If the flags are zero, the renameat2 function
123 acts like renameat. If the flag is not zero and there is no kernel
124 support for renameat2, the function will fail with an errno value of
125 EINVAL. This is different from the existing gnulib function renameatu,
126 which performs a plain rename operation in case of a RENAME_NOREPLACE
127 flags and a non-existing destination (and therefore has a race condition
128 that can clobber the destination inadvertently).
130 * The statx function has been added, a variant of the fstatat64
131 function with an additional flags argument. If there is no direct
132 kernel support for statx, glibc provides basic stat support based on
133 the fstatat64 function.
135 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
136 library if installed. libidn2 version 2.0.5 or later is recommended. If
137 libidn2 is not available, internationalized domain names are not encoded
138 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
139 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
140 with an encoding error.) Flags which used to change the IDN encoding and
141 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
142 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
143 deprecated. They no longer have any effect.
145 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
146 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
147 range of ELF gABI expressions including such constructs as
148 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
149 have been further restricted, and where in the past a dynamic string
150 token sequence may have been interpreted as a literal string it will
151 now cause a load failure. These load failures were always considered
152 unspecified behaviour from the perspective of the dynamic loader, and
153 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
154 results in a load failure now.
156 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
157 implementation includes all the standard functions provided by
160 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
161 thrd_detach, thrd_exit, and thrd_join for thread management.
163 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
164 mtx_destroy for mutual exclusion.
166 - call_once for function call synchronization.
168 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
169 cnd_wait for conditional variables.
171 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
173 Application developers must link against libpthread to use ISO C threads.
175 Deprecated and removed features, and other changes affecting compatibility:
177 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
178 installed. Software that was using either header should be updated to
179 use standard <stdio.h> interfaces instead.
181 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
182 This was never required by the C standard, and the macros just expanded
183 to call alternative names for the same functions. If you hoped getc and
184 putc would provide performance improvements over fgetc and fputc, instead
185 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
186 necessary, flockfile and funlockfile.
188 * All stdio functions now treat end-of-file as a sticky condition. If you
189 read from a file until EOF, and then the file is enlarged by another
190 process, you must call clearerr or another function with the same effect
191 (e.g. fseek, rewind) before you can read the additional data. This
192 corrects a longstanding C99 conformance bug. It is most likely to affect
193 programs that use stdio to read interactive input from a terminal.
196 * The macros 'major', 'minor', and 'makedev' are now only available from
197 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
198 headers that happen to include <sys/types.h>. These macros are rarely
199 used, not part of POSIX nor XSI, and their names frequently collide with
200 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
203 <sys/sysmacros.h> is a GNU extension. Portable programs that require
204 these macros should first include <sys/types.h>, and then include
205 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
207 * The tilegx*-*-linux-gnu configurations are no longer supported.
209 * The obsolete function ustat is no longer available to newly linked
210 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
211 function has been deprecated in favor of fstatfs and statfs.
213 * The obsolete function nfsservctl is no longer available to newly linked
214 binaries. This function was specific to systems using the Linux kernel
215 and could not usefully be used with the GNU C Library on systems with
216 version 3.1 or later of the Linux kernel.
218 * The obsolete function name llseek is no longer available to newly linked
219 binaries. This function was specific to systems using the Linux kernel
220 and was not declared in a header. Programs should use the lseek64 name
221 for this function instead.
223 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
224 getaddrinfo and getnameinfo functions have been deprecated. The behavior
225 previously selected by them is now always enabled.
227 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
228 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
229 restriction (rejecting '_' in host names, among other things) has been
230 removed, for increased compatibility with non-IDN name resolution.
232 * The fcntl function now have a Long File Support variant named fcntl64. It
233 is added to fix some Linux Open File Description (OFD) locks usage on non
234 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
235 fcntl and LFS support is handled transparently. Also for Linux, the OFD
236 locks act as a cancellation entrypoint.
238 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
239 ecb_crypt, and des_setparity are no longer available to newly linked
240 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
241 longer installed. These functions encrypted and decrypted data with the
242 DES block cipher, which is no longer considered secure. Software that
243 still uses these functions should switch to a modern cryptography library,
246 * Reflecting the removal of the encrypt and setkey functions above, the
247 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
248 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
251 * The obsolete function fcrypt is no longer available to newly linked
252 binaries. It was just another name for the standard function crypt,
253 and it has not appeared in any header file in many years.
255 * We have tentative plans to hand off maintenance of the passphrase-hashing
256 library, libcrypt, to a separate development project that will, we hope,
257 keep up better with new passphrase-hashing algorithms. We will continue
258 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
259 'crypt_r' should not need to change at all; however, distributions will
260 need to install <crypt.h> and libcrypt from a separate project.
262 In this release, if the configure option --disable-crypt is used, glibc
263 will not install <crypt.h> or libcrypt, making room for the separate
264 project's versions of these files. The plan is to make this the default
265 behavior in a future release.
267 Changes to build and runtime requirements:
269 GNU make 4.0 or later is now required to build glibc.
271 Security related changes:
273 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
274 been fixed by removing the glibc-internal IDNA implementation and using
275 the system-provided libidn2 library instead. Originally reported by Hanno
276 Böck and Christian Weisgerber.
278 CVE-2017-18269: An SSE2-based memmove implementation for the i386
279 architecture could corrupt memory. Reported by Max Horn.
281 CVE-2018-11236: Very long pathname arguments to realpath function could
282 result in an integer overflow and buffer overflow. Reported by Alexey
285 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
286 architecture could write beyond the target buffer, resulting in a buffer
287 overflow. Reported by Andreas Schwab.
289 The following bugs are resolved with this release:
291 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
292 [6889] manual: 'PWD' mentioned but not specified
293 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
294 when __WORDSIZE != 64
295 [13762] regex: re_search etc. should return -2 on memory exhaustion
296 [13888] build: /tmp usage during testing
297 [13932] math: dbl-64 pow unexpectedly slow for some inputs
298 [14092] nptl: Support C11 threads
299 [14095] localedata: Review / update collation data from Unicode / ISO
301 [14508] libc: -Wformat warnings
302 [14553] libc: Namespace pollution loff_t in sys/types.h
303 [14890] libc: Make NT_PRFPREG canonical.
304 [15105] libc: Extra PLT references with -Os
305 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
307 [16335] manual: Feature test macro documentation incomplete and out of
309 [16552] libc: Unify umount implementations in terms of umount2
310 [17082] libc: htons et al.: statement-expressions prevent use on global
311 scope with -O1 and higher
312 [17343] libc: Signed integer overflow in /stdlib/random_r.c
313 [17438] localedata: pt_BR: wrong d_fmt delimiter
314 [17662] libc: please implement binding for the new renameat2 syscall
315 [17721] libc: __restrict defined as /* Ignore */ even in c11
316 [17979] libc: inconsistency between uchar.h and stdint.h
317 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
318 [18023] libc: extend_alloca is broken (questionable pointer comparison,
319 horrible machine code)
320 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
322 [18471] libc: llseek should be a compat symbol
323 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
325 [18991] nss: nss_files skips large entry in database
326 [19239] libc: Including stdlib.h ends up with macros major and minor being
328 [19463] libc: linknamespace failures when compiled with -Os
329 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
330 [19527] locale: Normalized charset name not recognized by setlocale
331 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
332 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
334 [19728] network: out of bounds stack read in libidn function
335 idna_to_ascii_4i (CVE-2016-6261)
336 [19729] network: out of bounds heap read on invalid utf-8 inputs in
337 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
338 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
340 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
341 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
342 [20419] dynamic-link: files with large allocated notes crash in
344 [20530] libc: bswap_16 should use __builtin_bswap16() when available
345 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
346 [20980] manual: CFLAGS environment variable replaces vital options
347 [21163] regex: Assertion failure in pop_fail_stack when executing a
348 malformed regexp (CVE-2015-8985)
349 [21234] manual: use of CFLAGS makes glibc detect no optimization
350 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
351 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
352 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
353 [21508] locale: intl/tst-gettext failure with latest msgfmt
354 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
355 [21812] network: getifaddrs() returns entries with ifa_name == NULL
356 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
358 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
360 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
361 [22247] network: Integer overflow in the decode_digit function in
362 puny_decode.c in libidn (CVE-2017-14062)
363 [22342] nscd: NSCD not properly caching netgroup
364 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
365 [22550] localedata: es_ES locale (and other es_* locales): collation
366 should treat ñ as a primary different character, sync the collation
367 for Spanish with CLDR
368 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
369 by gcc configured with --enable-default-pie
370 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
371 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
372 crossing 2GB threshold (CVE-2017-18269)
373 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
375 [22735] time: Misleading typo in time.h source comment regarding
377 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
378 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
379 `__assert_perror_fail ()' to `__assert_fail_base ()'
380 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
382 [22786] libc: Stack buffer overflow in realpath() if input size is close
383 to SSIZE_MAX (CVE-2018-11236)
384 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
385 through an absolute DT_NEEDED path
386 [22792] build: tcb-offsets.h dependency dropped
387 [22797] libc: pkey_get() uses non-reserved name of argument
388 [22807] libc: PTRACE_* constants missing for powerpc
389 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
390 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
391 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
392 [22848] localedata: ca_ES: update date definitions from CLDR
393 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
394 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
395 [22896] localedata: Update locale data for an_ES
396 [22902] math: float128 test failures with GCC 8
397 [22918] libc: multiple common of `__nss_shadow_database'
398 [22919] libc: sparc32: backtrace yields infinite backtrace with
400 [22926] libc: FTBFS on powerpcspe
401 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
403 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
404 [22947] libc: FAIL: misc/tst-preadvwritev2
405 [22963] localedata: cs_CZ: Add alternative month names
406 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
407 [22996] localedata: change LC_PAPER to en_US in es_BO locale
408 [22998] dynamic-link: execstack tests are disabled when SELinux is
410 [23005] network: Crash in __res_context_send after memory allocation
412 [23007] math: strtod cannot handle -nan
413 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
414 [23036] regex: regex equivalence class regression
415 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
416 [23069] libc: sigaction broken on riscv64-linux-gnu
417 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
418 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
419 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
420 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
421 and libc build with -Os)
422 [23140] localedata: More languages need two forms of month names
423 [23145] libc: _init/_fini aren't marked as hidden
424 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
425 [23171] math: C++ iseqsig for long double converts arguments to double
426 [23178] nscd: sudo will fail when it is run in concurrent with commands
427 that changes /etc/passwd
428 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
430 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
431 [23208] localedata: New locale - Lower Sorbian (dsb)
432 [23233] regex: Memory leak in build_charclass_op function in file
434 [23236] stdio: Harden function pointers in _IO_str_fields
435 [23250] nptl: Offset of __private_ss differs from GCC
436 [23253] math: tgamma test suite failures on i686 with -march=x86-64
437 -mtune=generic -mfpmath=sse
438 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
440 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
441 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
443 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
444 [23277] math: nan function should not have const attribute
445 [23279] math: scanf and strtod wrong for some hex floating-point
446 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
448 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
449 [23303] build: undefined reference to symbol
450 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
451 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
453 [23313] stdio: libio vtables validation and standard file object
455 [23329] libc: The __libc_freeres infrastructure is not properly run across
457 [23349] libc: Various glibc headers no longer compatible with
459 [23351] malloc: Remove unused code related to heap dumps and malloc
461 [23363] stdio: stdio-common/tst-printf.c has non-free license
462 [23396] regex: Regex equivalence regression in single-byte locales
463 [23422] localedata: oc_FR: More updates of locale data
464 [23442] build: New warning with GCC 8
465 [23448] libc: Out of bounds access in IBM-1390 converter
466 [23456] libc: Wrong index_cpu_LZCNT
467 [23458] build: tst-get-cpu-features-static isn't added to tests
468 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
470 [23467] dynamic-link: x86/CET: A property note parser bug
477 * The GNU C Library can now be compiled with support for building static
478 PIE executables (See --enable-static-pie in INSTALL). These static PIE
479 executables are like static executables but can be loaded at any address
480 and provide additional security hardening benefits at the cost of some
481 memory and performance. When the library is built with --enable-static-pie
482 the resulting libc.a is usable with GCC 8 and above to create static PIE
483 executables using the GCC option '-static-pie'. This feature is currently
484 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
485 aarch64 with binutils 2.30 or later.
487 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
488 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
491 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
493 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
495 * In order to support faster and safer process termination the malloc API
496 family of functions will no longer print a failure address and stack
497 backtrace after detecting heap corruption. The goal is to minimize the
498 amount of work done after corruption is detected and to avoid potential
499 security issues in continued process execution. Reducing shutdown time
500 leads to lower overall process restart latency, so there is benefit both
501 from a security and performance perspective.
503 * The abort function terminates the process immediately, without flushing
504 stdio streams. Previous glibc versions used to flush streams, resulting
505 in deadlocks and further data corruption. This change also affects
506 process aborts as the result of assertion failures.
508 * On platforms where long double has the IEEE binary128 format (aarch64,
509 alpha, mips64, riscv, s390 and sparc), the math library now implements
510 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
511 These are the same interfaces added in version 2.26 for some platforms where
512 this format is supported but is not the format of long double.
514 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
515 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
516 implements interfaces for that type, as defined by ISO/IEC TS
517 18661-3:2015. These are corresponding interfaces to those supported for
520 * The math library now implements interfaces for the _Float32, _Float64 and
521 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
522 corresponding interfaces to those supported for _Float128.
524 * glibc now implements the memfd_create and mlock2 functions on Linux.
526 * Support for memory protection keys was added. The <sys/mman.h> header now
527 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
530 * The copy_file_range function was added.
532 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
534 * The ldconfig utility now processes `include' directives using the C/POSIX
535 collation ordering. Previous glibc versions used locale-specific
536 ordering, the change might break systems that relied on that.
538 * Support for two grammatical forms of month names has been added.
539 In a call to strftime, the "%B" and "%b" format specifiers will now
540 produce the grammatical form required when the month is used as part
541 of a complete date. New "%OB" and "%Ob" specifiers produce the form
542 required when the month is named by itself. For instance, in Greek
543 and in many Slavic and Baltic languages, "%B" will produce the month
544 in genitive case, and "%OB" will produce the month in nominative case.
546 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
547 are all valid and will all accept any known form of month
548 name---standalone or complete, abbreviated or full. In a call to
549 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
550 the strings used by "%B" and "%b", respectively. New query
551 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
552 used by "%OB" and "%Ob", respectively.
554 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
555 define the strings for %OB and %Ob, respectively; these have the
556 same syntax as "mon" and "abmon". These arrays are optional; if they
557 are not provided then they have the same content as "mon" and "abmon",
560 These features are provided for locales which define "alt_mon" and/or
561 "ab_alt_mon" in their locale source data. This release includes such
562 alternative month name data for the following languages: Belarusian,
563 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
565 This feature is currently a GNU extension, but it is expected to
566 be added to the next revision of POSIX, and it is also already
567 available on some BSD-derived operating systems.
569 This feature will cause existing statically compiled applications
570 to fail to load locales and fall back to the builtin C/POSIX locales.
571 See notes below for other changes affecting compatibility.
573 * Support for the RISC-V ISA running on Linux has been added. This port
574 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
575 for the following ISA and ABI pairs:
581 Deprecated and removed features, and other changes affecting compatibility:
583 * Statically compiled applications attempting to load locales compiled for the
584 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
585 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
586 support for two grammatical forms of the month names, also extends the locale
587 data binary format. Static applications needing locale support must be
588 recompiled to match the runtime and data they are deployed with. In some
589 distributions there is an upgrade window where dynamically linked applications
590 may use a new library but the old locale data and also fall back to the
591 builtin C/POSIX locales; restarting the application process is sufficient to
594 * Support for statically linked applications which call dlopen is deprecated
595 and will be removed in a future version of glibc. Applications which call
596 dlopen need to be linked dynamically instead.
598 * Support for old programs which use internal stdio data structures and
599 functions is deprecated. This includes programs which use the C++ streams
600 provided by libstdc++ in GCC 2.95. Programs which use the internal
601 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
602 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
603 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
604 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
605 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
606 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
607 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
608 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
609 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
610 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
611 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
612 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
613 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
614 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
615 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
616 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
617 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
618 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
619 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
620 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
621 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
622 future version of glibc. Unlike other symbol removals, these old
623 applications will not be supported using compatibility symbols.
625 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
626 defined by <sys/ptrace.h>.
628 * libm no longer supports SVID error handling (calling a user-provided
629 matherr function on error) or the _LIB_VERSION variable to control error
630 handling. (SVID error handling and the _LIB_VERSION variable still work
631 for binaries linked against older versions of the GNU C Library.) The
632 libieee.a library is no longer provided. math.h no longer defines struct
633 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
634 TLOSS, PLOSS and HUGE.
636 * The libm functions pow10, pow10f and pow10l are no longer supported for
637 new programs. Programs should use the standard names exp10, exp10f and
638 exp10l for these functions instead.
640 * The mcontext_t type is no longer the same as struct sigcontext. On
641 platforms where it was previously the same, this changes the C++ name
642 mangling for interfaces involving this type.
644 * The add-ons mechanism for building additional packages at the same time as
645 glibc has been removed. The --enable-add-ons configure option is now
648 * The --without-fp configure option is now ignored. Whether hardware
649 floating-point instructions are used is now configured based on whether
650 the compiler used at configure time (without any options implied by a
651 --with-cpu= configure option) uses such instructions.
653 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
654 the specified string can be parsed as a domain name.
656 * In the malloc_info output, the <heap> element may contain another <aspace>
657 element, "subheaps", which contains the number of sub-heaps.
659 * The libresolv function p_secstodate is no longer supported for new
662 * The tilepro-*-linux-gnu configuration is no longer supported.
664 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
665 and will be removed in a future release. Software that is still using
666 either header should be updated to use standard <stdio.h> interfaces
669 libio.h was originally the header for a set of supported GNU extensions,
670 but they have not been maintained as such in many years, they are now
671 standing in the way of improvements to stdio, and we don't think there are
672 any remaining external users. _G_config.h was never intended for public
673 use, but predates the bits convention.
675 Changes to build and runtime requirements:
677 * bison version 2.7 or later is required to generate code in the 'intl'
680 Security related changes:
682 CVE-2009-5064: The ldd script would sometimes run the program under
683 examination directly, without preventing code execution through the
684 dynamic linker. (The glibc project disputes that this is a security
685 vulnerability; only trusted binaries must be examined using the ldd
688 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
689 suffered from a one-byte overflow during ~ operator processing (either
690 on the stack or the heap, depending on the length of the user name).
691 Reported by Tim Rühsen.
693 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
694 would sometimes fail to free memory allocated during ~ operator
695 processing, leading to a memory leak and, potentially, to a denial
698 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
699 without GLOB_NOESCAPE, could write past the end of a buffer while
700 unescaping user names. Reported by Tim Rühsen.
702 CVE-2017-17426: The malloc function, when called with an object size near
703 the value SIZE_MAX, would return a pointer to a buffer which is too small,
704 instead of NULL. This was a regression introduced with the new malloc
705 thread cache in glibc 2.26. Reported by Iain Buclaw.
707 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
708 to the allocation of too much memory. (This is not a security bug per se,
709 it is mentioned here only because of the CVE assignment.) Reported by
712 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
713 of the number of search path components. (This is not a security
714 vulnerability per se because no trust boundary is crossed if the fix for
715 CVE-2017-1000366 has been applied, but it is mentioned here only because
716 of the CVE assignment.) Reported by Qualys.
718 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
719 for AT_SECURE or SUID binaries could be used to load libraries from the
722 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
723 succeeds without returning an absolute path due to unexpected behaviour
724 of the Linux kernel getcwd syscall. Reported by halfdog.
726 CVE-2018-6485: The posix_memalign and memalign functions, when called with
727 an object size near the value of SIZE_MAX, would return a pointer to a
728 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
730 CVE-2018-6551: The malloc function, when called with an object size near
731 the value of SIZE_MAX, would return a pointer to a buffer which is too
732 small, instead of NULL.
734 The following bugs are resolved with this release:
736 [866] glob: glob should match dangling symlinks
737 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
738 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
740 [5997] math: Very slow execution of sinf function
741 [10580] localedata: hr_HR: updated locale
742 [10871] locale: 'mon' array should contain both nominative and genitive
744 [12349] localedata: eu_ES: incorrect thousands separator
745 [13605] localedata: shn_MM: new Shan locale
746 [13805] localedata: ru_RU: currency should use ',' as radix point
747 [13953] localedata: km_KH: locale update
748 [13994] localedata: mjw_IN: new locale
749 [14121] build: make writes .mo files in po directory
750 [14333] libc: Fix the race between atexit() and exit()
751 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
752 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
754 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
755 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
756 width Latin characters
757 [15332] localedata: es_CU: locale update
758 [15436] stdio: Don't close or flush stdio streams on abort
759 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
761 [16148] localedata: ca_ES: incorrect thousands separator
762 [16750] dynamic-link: ldd should not try to execute the binaries
764 [16777] localedata: pl_PL: incorrect thousands separator in locale
765 [16905] localedata: hanzi: new collation
766 [17563] localedata: cmn_TW: add hanzi collation
767 [17750] localedata: wrong collation order of diacritics in most locales
768 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
770 [17956] build: Build fails on missing definitions from header file
771 nss/nss.h when Mozilla NSS is used for cryptography
772 [18203] libc: realpath() does not handle unreachable paths correctly
773 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
774 [18812] localedata: kab_DZ: new Kabyle Algeria locale
775 [18822] libc: Internal functions are called via PLT
776 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
777 [19170] libc: __gmon_start__ defined in hppa in crtn.S
778 [19574] libc: glibc should support building static PIE binaries
779 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
781 [19971] glob: glob: Do not skip entries with zero d_ino values
782 [19982] localedata: fr.po: spelling mistake for error code EXDEV
783 [20008] localedata: km_KH: convert to translit_neutral
784 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
785 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
786 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
787 [20482] localedata: de_CH: abbreviated weekdays should be two letters
788 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
790 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
791 they're set, wrong errors returned by gaih_inet when lookup functions
793 [20756] localedata: [PATCH] Use Unicode wise thousands separator
794 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
796 [20952] localedata: yuw_PG: new locale
797 [21084] localedata: charmaps/IBM858: new codepage
798 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
799 [21242] libc: assert gives pedantic warning in old gcc versions
800 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
801 __regcall calling convention
802 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
803 [21326] libc: C99 functions are not declared for C++11 and later if
804 _GNU_SOURCE is not predefined by g++
805 [21457] libc: sys/ucontext.h namespace
806 [21530] libc: tmpfile() should be implemented using O_TMPFILE
807 [21660] math: GCC fails to compile a formula with tgmath.h
808 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
809 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
810 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
811 [21684] math: tgmath.h handling of complex integers
812 [21685] math: tgmath.h handling of bit-fields
813 [21686] math: tgmath.h handling of __int128
814 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
816 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
818 [21750] localedata: column width of characters incompatible with classical
820 [21754] malloc: malloc: Perform as little work as possible after heap
821 consistency check failures
822 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
823 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
824 [21791] string: Unused XXX_chk_XXX functions in libc.a
825 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
827 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
829 [21845] localedata: Added new Locale bho_NP
830 [21853] localedata: Fix abday Which looks same as day in zh_SG
831 [21854] localedata: Added New Locale en_SC
832 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
833 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
834 _dl_runtime_resolve_avx_slow
835 [21885] network: getaddrinfo: gethosts does not release resolver context
836 on memory allocation failure
837 [21899] libc: XPG4.2 sigaction namespace
838 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
840 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
842 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
843 errno value, causing getaddrinfo to fail
844 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
845 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
846 instead of EAI_NODATA
847 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
848 constant PTRACE_SEIZE_DEVEL
849 [21930] math: C-only gcc builtins used in <math.h> isinf
850 [21932] network: Unpaired __resolv_context_get in generic get*_r
852 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
853 [21944] libc: sigval namespace
854 [21951] localedata: Update hanzi collation by stroke
855 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
856 sysdeps/x86_64/fpu/e_expf.S
857 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
858 allocation in syscall loops)
859 [21959] localedata: Fix Country name for xh_ZA
860 [21960] localedata: Fix abmon for bem_ZM
861 [21966] math: AVX2 mathvec functions use FMA without checking
862 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
863 [21971] localedata: Added New Locale for mfe_MU
864 [21972] libc: assert macro requires operator== (int) for its argument type
865 [21973] math: [sparc] libm missing sqrtl compat symbol
866 [21974] libc: Remove __bb_init_func and __bb_exit_func
867 [21982] string: stratcliff.c: error: assuming signed overflow does not
869 [21986] stdio: __guess_grouping is called incorrectly
870 [21987] math: [sparc32] wrong bits/long-double.h installed
871 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
873 [22022] localedata: Missing country_name for mni_IN
874 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
876 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
878 [22026] locale: iconv_open: heap overflow on gconv_init failure
879 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
880 [22035] math: [m68k] bits/math-inline.h macro namespace
881 [22038] localedata: Fix abbreviated weeks and months for Somali
882 [22044] localedata: Remove redundant data for Limburgish Language
883 [22050] malloc: Linking with -lmcheck does not hook
884 __malloc_initialize_hook correctly
885 [22051] libc: zero terminator in the middle of glibc's .eh_frame
886 [22052] malloc: malloc failed to compile with GCC 7 and -O3
887 [22070] localedata: charmaps/UTF-8: wcwidth for
888 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
889 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
890 Jungseong and Jongseong) should be 0
891 [22078] nss: nss_files performance issue in multi mode
892 [22082] math: bits/math-finite.h exp10 condition
893 [22086] libc: pcprofiledump incorrect cross-endian condition
894 [22093] dynamic-link: ld.so no longer searches in .../x86_64
895 [22095] network: Name server address allocation memory leak in resolv.conf
897 [22096] network: __resolv_conf_attach can incorrectly free passed conf
899 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
900 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
902 [22111] malloc: malloc: per thread cache is not returned when thread exits
903 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
904 [22134] libc: [linux] implement fexecve with execveat
905 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
907 [22145] libc: ttyname() gives up too early in the face of namespaces
908 [22146] math: C++ build issue with float128 on x86_64
909 [22153] nptl: nptl: save error code before process termination
910 [22156] libc: [hppa,ia64,microblaze] Executable stack default
911 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
912 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
913 [22165] libc: [hppa] Text relocations in libc.so
914 [22180] libc: destructor registered via __cxa_atexit is called twice
915 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
916 symlinks") cause "make" segfaults
917 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
919 [22207] libc: FAIL: stdlib/test-atexit-race
920 [22225] math: nearbyint arithmetic moved before feholdexcept
921 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
922 [22235] math: iscanonical in C++ and float128
923 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
924 the svid compat wrapper
925 [22244] math: ynf and yn are wrong without the svid compat wrapper
926 [22273] libc: Improper assert in Linux posix_spawn implementation
927 [22284] libc: -pg -pie doesn't work
928 [22292] locale: localedef exits with error 4 when it should be error 1
929 [22294] locale: Allow "" for int_currency_symbol definition in locales.
930 [22295] locale: Don't warn on non-symbolic characters in locale sources in
932 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
933 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
935 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
936 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
937 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
938 [22322] libc: [mips64] wrong bits/long-double.h installed
939 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
940 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
942 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
943 [22343] malloc: Integer overflow in posix_memalign
944 [22347] libc: getrandom() returns the number of bytes that were copied to
945 the buffer even though the comments say "Return 0 on success and -1 on
947 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
948 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
950 [22370] dynamic-link: Incorrect note padding check
951 [22375] libc: malloc returns pointer from tcache_get when should return
952 NULL (CVE-2017-17426)
953 [22377] math: iseqsig, float128 and C++
954 [22382] localedata: Error in tpi_PG locale
955 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
956 inside the ASCII printable range
957 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
959 [22403] localedata: Slash needs escaping in some locales
960 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
961 [22409] network: res_hnok does not accept some host names used on the
963 [22412] network: res_dnok, res_hnok should perform syntax checks
964 [22413] network: ns_name_pton ignores syntactically invalid trailing
966 [22415] stdio: setvbuf can lead to invalid free/segfault
967 [22432] build: Non-deterministic build
968 [22439] malloc: malloc_info should compute summary statistics for all sub-
970 [22442] network: if_nametoindex could report index for the wrong
972 [22446] build: aliasing violation calling readlink in handle_request
973 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
974 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
975 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
976 __nscd_hash/__nss_hash
977 [22463] network: p_secstodate overflow handling
978 [22469] localedata: pl_PL LC_COLLATE does not use i18n
979 [22478] libc: sigwait can fail with EINTR
980 [22505] libc: ldconfig processes include directive in locale-specific
982 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
983 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
984 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
985 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
986 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
987 [22534] localedata: Collation rules for Serbian and Bosnian should be the
989 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
990 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
991 [22577] libc: missing newline after "cannot allocate TLS data structures
993 [22588] manual: manual/conf.texi: missing underscore in front of
995 [22593] math: nextafter and nexttoward are declared with const attribute
996 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
997 [22603] string: ia64 memchr overflows internal pointer check
998 [22605] libc: SH clone does not set the exit code correctly
999 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1001 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1002 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1003 modified in case of success
1004 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1005 [22615] manual: manual: ambiguous wording about errno value in case of
1007 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1009 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1010 binaries or if /proc is not mounted (CVE-2017-16997)
1011 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1012 [22630] build: $(no-pie-ldflag) is no longer effective
1013 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1014 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1015 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1016 [22637] nptl: guard size is subtracted from thread stack size instead of
1018 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1019 [22657] localedata: hu_HU: Avoid double space in date
1020 [22660] math: fmax, fmin sNaN handling on alpha
1021 [22664] libc: New warning of GCC8
1022 [22665] math: alpha: ceil and floor raise inexact exceptions
1023 [22666] math: alpha: trunc raise inexact exceptions
1024 [22667] libc: makecontext lacks stack alignment on i386
1025 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1026 [22679] libc: getcwd(3) can succeed without returning an absolute path
1028 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1029 elision and tunables
1030 [22687] math: [powerpc-nofpu] complex long double functions spurious
1032 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1033 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1034 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1036 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1037 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1039 [22701] nis: Incomplete removal of libnsl
1040 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1041 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1042 [22715] dynamic-link: FAIL: elf/tst-audit10
1043 [22719] libc: Backtrace tests fail on hppa
1044 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1045 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1046 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1047 before the first call to crypt_r () in crypt/badsalttest.c
1054 * A per-thread cache has been added to malloc. Access to the cache requires
1055 no locks and therefore significantly accelerates the fast path to allocate
1056 and free small amounts of memory. Refilling an empty cache requires locking
1057 the underlying arena. Performance measurements show significant gains in a
1058 wide variety of user workloads. Workloads were captured using a special
1059 instrumented malloc and analyzed with a malloc simulator. Contributed by
1060 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1062 * Unicode 10.0.0 Support: Character encoding, character type info, and
1063 transliteration tables are all updated to Unicode 10.0.0, using
1064 generator scripts contributed by Mike FABIAN (Red Hat).
1065 These updates cause user visible changes, especially the changes in
1066 wcwidth for many emoji characters cause problems when emoji sequences
1067 are rendered with pango, see for example:
1068 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1070 * Collation of Hungarian has been overhauled and is now consistent with "The
1071 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1074 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1076 - The GNU C Library will now detect when /etc/resolv.conf has been
1077 modified and reload the changed configuration. The new resolver option
1078 “no-reload” (RES_NORELOAD) disables this behavior.
1080 - The GNU C Library now supports an arbitrary number of search domains
1081 (configured using the “search” directive in /etc/resolv.conf);
1082 previously, there was a hard limit of six domains. For backward
1083 compatibility, applications that directly modify the ‘_res’ global
1084 object are still limited to six search domains.
1086 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1087 Library will now randomly pick a name server from the configuration as a
1088 starting point. (Previously, the second name server was always used.)
1090 * The tunables feature is now enabled by default. This allows users to tweak
1091 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1093 * New function reallocarray, which resizes an allocated block (like realloc)
1094 to the product of two sizes, with a guaranteed clean failure upon integer
1095 overflow in the multiplication. Originally from OpenBSD, contributed by
1096 Dennis Wölfing and Rüdiger Sonderfeld.
1098 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1099 These are extended versions of preadv and pwritev, respectively, taking an
1100 additional flags argument. The set of supported flags depends on the
1101 running kernel; full support currently requires kernel 4.7 or later.
1103 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1104 create a new session ID for the spawned process. This feature is
1105 scheduled to be added to the next major revision of POSIX; for the time
1106 being, it is available under _GNU_SOURCE.
1108 * errno.h is now safe to use from C-preprocessed assembly language on all
1109 supported operating systems. In this context, it will only define the
1110 Exxxx constants, as preprocessor macros expanding to integer literals.
1112 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1113 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1114 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1115 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1117 To compile programs that use this feature, the compiler must support
1118 128-bit floating point with the type name _Float128 (as defined by TS
1119 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1120 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1121 must be defined to make the new interfaces visible.
1123 The new functions and macros correspond to those present for other
1124 floating-point types (except for a few obsolescent interfaces not
1125 supported for the new type), with F128 or f128 suffixes; for example,
1126 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1127 printf or scanf formats for the new type; the strfromf128 and strtof128
1128 interfaces should be used instead.
1130 Deprecated and removed features, and other changes affecting compatibility:
1132 * The synchronization that pthread_spin_unlock performs has been changed to
1133 now be equivalent to a C11 atomic store with release memory order to the
1134 spin lock's memory location. Previously, several (but not all)
1135 architectures used stronger synchronization (e.g., containing what is
1136 often called a full barrier). This change can improve performance, but
1137 may affect odd fringe uses of spin locks that depend on the previous
1138 behavior (e.g., using spin locks as atomic variables to try to implement
1139 Dekker's mutual exclusion algorithm).
1141 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1144 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1145 will only be built and installed when the GNU C Library is configured with
1146 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1147 as TIRPC or rpcsvc-proto, to be used.
1149 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1150 libnss_compat, are deprecated, and will not be built or installed by
1153 The NIS(+) support library, libnsl, is also deprecated. By default, a
1154 compatibility shared library will be built and installed, but not headers
1155 or development libraries. Only a few NIS-related programs require this
1156 library. (In particular, the GNU C Library has never required programs
1157 that use 'gethostbyname' to be linked with libnsl.)
1159 Replacement implementations based on TIRPC, which additionally support
1160 IPv6, are available from <https://github.com/thkukuk/>. The configure
1161 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1162 name service modules, to be built and installed.
1164 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1165 support is enabled, the configured recursive resolver must support EDNS.
1166 (Responding to EDNS-enabled queries with responses which are not
1167 EDNS-enabled is fine, but FORMERR responses are not.)
1169 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1170 servers have not supported this opcode for a long time.
1172 * The _res_opcodes variable has been removed from libresolv. It had been
1173 exported by accident.
1175 * <string.h> no longer includes inline versions of any string functions,
1176 as this kind of optimization is better done by the compiler. The macros
1177 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1179 * The nonstandard header <xlocale.h> has been removed. Most programs should
1180 use <locale.h> instead. If you have a specific need for the definition of
1181 locale_t with no other declarations, please contact
1182 libc-alpha@sourceware.org and explain.
1184 * The obsolete header <sys/ultrasound.h> has been removed.
1186 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1188 * The obsolete function cfree has been removed. Applications should use
1191 * The stack_t type no longer has the name struct sigaltstack. This changes
1192 the C++ name mangling for interfaces involving this type.
1194 * The ucontext_t type no longer has the name struct ucontext. This changes
1195 the C++ name mangling for interfaces involving this type.
1197 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1198 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1199 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1200 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1201 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1202 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1203 fpregset_t type no longer has the name struct fpu. This changes the C++
1204 name mangling for interfaces involving those types.
1206 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1207 synced with the kernel:
1209 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1210 are not supported on this architecture and have been removed.
1212 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1213 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1214 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1216 Programs that assume the GET/SETREGS ptrace requests are universally
1217 available will now fail to build, instead of malfunctioning at runtime.
1219 Changes to build and runtime requirements:
1221 * Linux kernel 3.2 or later is required at runtime, on all architectures
1222 supported by that kernel. (This is a change from version 2.25 only for
1225 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1227 * On most architectures, GCC 4.9 or later is required to build the GNU C
1228 Library. On powerpc64le, GCC 6.2 or later is required.
1230 Older GCC versions and non-GNU compilers are still supported when
1231 compiling programs that use the GNU C Library. (We do not know exactly
1232 how old, and some GNU extensions to C may be _de facto_ required. If you
1233 are interested in helping us make this statement less vague, please
1234 contact libc-alpha@sourceware.org.)
1236 Security related changes:
1238 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1239 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1241 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1242 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1244 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1245 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1248 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1249 fixed (CVE-2017-12133).
1251 The following bugs are resolved with this release:
1253 [984] network: Respond to changed resolv.conf in gethostbyname
1254 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1255 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1257 [12189] libc: __stack_chk_fail should not attempt a backtrace
1259 [14096] time: Race condition on timezone/tst-timezone.out
1260 [14172] localedata: az_IR: new locale
1261 [14995] build: glibc fails to build if gold is the default linker, even if
1263 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1264 [16637] network: inet_pton function is accepting IPv6 with bad format
1265 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1266 x86 and other generic code
1267 [16875] localedata: ko_KR: fix lang_name
1268 [17225] localedata: ar_SY: localized month names for May and June are
1270 [17297] localedata: da_DK: wrong date_fmt string
1271 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1272 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1273 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1274 [19066] localedata: ar_SA abbreviated day and month names are in English
1275 [19569] network: resolv: Support an arbitrary number of search domains
1276 [19570] network: Implement random DNS server selection in the stub
1278 [19838] locale: localedef fails on PA-RISC
1279 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1280 order of 0D36 and 0D37
1281 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1283 [20098] libc: FAIL: debug/backtrace-tst on hppa
1284 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1286 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1288 [20313] localedata: Update locale data to Unicode 9.0
1289 [20424] manual: Document how to provide a malloc replacement
1290 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1292 [20686] locale: Add el_GR@euro to SUPPORTED.
1293 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1294 failures consistently
1295 [21015] dynamic-link: Document and fix --enable-bind-now
1296 [21016] nptl: pthread_cond support is broken on hppa
1297 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1298 frame-pointer on i386
1299 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1301 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1302 [21088] libc: Build fails with --enable-static-nss
1303 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1304 generic c code is used
1305 [21109] libc: Tunables broken on big-endian
1306 [21112] math: powf has large ulp errors with base close to 1 and exponent
1308 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1310 [21120] malloc: glibc malloc is incompatible with GCC 7
1311 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1313 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1314 (-0.0) when linking with -lieee
1315 [21171] math: log10, log2 and lgamma return incorrect results
1316 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1317 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1318 [21207] localedata: ce_RU: update weekdays from CLDR
1319 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1320 [21217] localedata: Update months from CLDR-31
1321 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1322 [21243] libc: support_delete_temp_file should issue warning for failed
1324 [21244] libc: support resolv_test_start() socket fd close should be
1326 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1327 new posix_spawn implementation
1328 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1329 leads to lower CPU frequency
1330 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1331 [21261] libc: [sparc64] bits/setjmp.h namespace
1332 [21267] network: [mips] bits/socket.h IOC* namespace
1333 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1334 [21270] libc: mmap64 silently truncates large offset values
1335 [21275] libc: posix_spawn always crashes on ia64 now
1336 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1337 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1338 [21289] libc: Incorrect declaration for 32-bit platforms with
1339 _FILE_OFFSET_BITS=64 causes build error
1340 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1341 not support gethostbyname4_r
1342 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1344 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1346 [21340] libc: Support POSIX_SPAWN_SETSID
1347 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1348 [21359] network: ns_name_pack needs additional byte in destination buffer
1349 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1350 against fragmentation attacks (CVE-2017-12132)
1351 [21369] network: resolv: Remove EDNS fallback
1352 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1354 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1355 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1356 [21393] stdio: Missing dup3 error check in freopen, freopen64
1357 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1358 [21399] localedata: Bad description for U00EC in
1359 localedata/charmaps/CP1254
1360 [21411] malloc: realloc documentation error
1361 [21426] network: sys/socket.h uio.h namespace
1362 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1363 [21445] libc: signal.h bsd_signal namespace
1364 [21455] network: Network headers stdint.h namespace
1365 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1367 [21475] network: resolv: Overlong search path is truncated mid-label
1368 [21511] libc: sigstack namespace
1369 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1370 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1371 bits/syscall.h: No such file or directory
1372 [21517] libc: struct sigaltstack namespace
1373 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1374 [21533] localedata: Update locale data to Unicode 10.0
1376 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1377 junk at end of line, first unrecognized character is `@'
1378 [21538] libc: SIG_HOLD missing for XPG4
1379 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1380 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1381 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1382 [21543] libc: sigevent namespace
1383 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1384 size for general purpose registers in mcontext_t structure
1385 [21550] libc: sigwait namespace
1386 [21552] libc: XPG4 bsd_signal namespace
1387 [21554] libc: sigpause namespace
1388 [21560] libc: sys/wait.h signal.h namespace
1389 [21561] libc: waitid namespace
1390 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1391 '/usr/include/stdlib.h': Permission denied
1392 [21575] libc: sys/wait.h missing struct rusage definition
1393 [21584] libc: sigaltstack etc namespace
1394 [21597] libc: siginterrupt namespace
1395 [21607] math: hppa: FAIL: math/test-tgmath
1396 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1398 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1399 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1400 stack and heap (CVE-2017-1000366)
1401 [21625] libc: wait3 namespace
1402 [21654] nss: Incorrect pointer alignment in NSS group merge result
1404 [21657] network: Parse interface zone id for node-local multicast
1405 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1407 [21666] libc: .symver is used on common symbol
1408 [21668] network: resolv: res_init cross-thread broadcast introduces race
1410 [21687] math: tgmath.h totalorder, totalordermag return type
1411 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1413 [21696] libc: Incorrect assumption of of __cpu_mask in
1414 posix/sched_cpucount.c
1415 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1416 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1418 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1419 leading to relocation crash
1420 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1421 `resolv_conf_matches (resp, conf)' failed.
1422 [21710] localedata: Added Samoan language locale for Samoa
1423 [21711] localedata: Pashto yesstr/nostr locale are missing
1424 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1425 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1426 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1427 [21723] localedata: yesstr/nostr missing for Chinese language locale
1428 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1430 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1432 [21728] localedata: New Locale for Tongan language
1433 [21729] localedata: incorrect LC_NAME fields for hi_IN
1434 [21733] localedata: yesstr and nostr are missing for zh_HK
1435 [21734] localedata: Missing yesstr and nostr are for kw_GB
1436 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1437 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1438 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1439 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1441 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1443 [21749] localedata: Wrong abbreviated day name (“abday”) for
1445 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1446 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1447 [21759] localedata: missing yesstr and nostr for Tigrinya
1448 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1449 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1450 [21767] localedata: Missing Bislama locales
1451 [21768] localedata: Missing yesstr and nostr for aa_ET
1452 [21770] localedata: Missing Field in li_NL
1453 [21778] nptl: Robust mutex may deadlock
1454 [21779] libc: MicroBlaze segfaults when loading libpthread
1455 [21783] localedata: Fix int_select international_call_prefixes
1456 [21784] localedata: Inconsistency in country_isbn
1457 [21788] localedata: Missing Country Postal Abbreviations
1458 [21794] localedata: Added-country_isbn-for-Italy
1459 [21795] localedata: Add/Fix country_isbn for France
1460 [21796] localedata: Added country_isbn for Republic of Korea
1461 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1462 [21799] localedata: Added int_select international_call_prefixes
1463 [21801] localedata: Added int_select international_call_prefixes
1464 [21804] nptl: Double semicolon in thread-shared-types.h
1465 [21807] localedata: LC_ADDRESS fix for pap_CW
1466 [21808] localedata: Fix LC_ADDRESS for pap_AW
1467 [21821] localedata: Added country_name in mai_IN
1468 [21822] localedata: Fix LC_TIME for mai_IN
1469 [21823] localedata: missing yesstr, nostr for sa_IN
1470 [21825] localedata: Fix name_mrs for mag_IN
1471 [21828] localedata: 2.26 changelog should mention user visible changes
1473 [21835] localedata: Added Maithili language locale for Nepal
1474 [21838] localedata: Removed redundant data for the_NP
1475 [21839] localedata: Fix LC_MONETARY for ta_LK
1476 [21844] localedata: Fix Latin characters and Months Sequence.
1477 [21848] localedata: Fix mai_NP Title Name
1482 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1483 24731-2:2010, is supported to enable declarations of functions from that
1484 TR. Note that not all functions from that TR are supported by the GNU C
1487 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1488 18661-1:2014, is supported to enable declarations of functions and macros
1489 from that TS. Note that not all features from that TS are supported by
1492 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1493 18661-4:2015, is supported to enable declarations of functions and macros
1494 from that TS. Note that most features from that TS are not supported by
1497 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1498 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1499 Since the GNU C Library defaults to a much newer revision of POSIX, this
1500 will only affect programs that specifically request an old conformance
1501 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1502 see a change in the visible declarations, but a program compiled with
1503 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1506 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1507 defined by all multithreaded code, but glibc has not required this for
1510 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1511 means that in a future release, the macros “major”, “minor”, and “makedev”
1512 will only be available from <sys/sysmacros.h>.
1514 These macros are not part of POSIX nor XSI, and their names frequently
1515 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1516 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1517 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1520 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1521 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1522 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1524 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1525 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1526 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1527 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1528 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1529 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1530 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1531 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1532 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1533 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1534 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1537 * New <math.h> features are added from TS 18661-1:2014:
1539 - Signaling NaN macros: SNANF, SNAN, SNANL.
1541 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1542 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1543 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1545 - llogb functions: the llogb, llogbf and llogbl functions, and the
1546 FP_LLOGB0 and FP_LLOGBNAN macros.
1548 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1551 - Comparison macros: iseqsig.
1553 - Classification macros: iscanonical, issubnormal, iszero.
1555 - Total order functions: totalorder, totalorderf, totalorderl,
1556 totalordermag, totalordermagf, totalordermagl.
1558 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1560 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1561 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1563 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1564 are added to libc. They convert a floating-point number into string.
1566 * Most of glibc can now be built with the stack smashing protector enabled.
1567 It is recommended to build glibc with --enable-stack-protector=strong.
1568 Implemented by Nick Alcock (Oracle).
1570 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1571 intended to be used instead of memset() to erase sensitive data after use;
1572 the compiler will not optimize out calls to explicit_bzero even if they
1573 are "unnecessary" (in the sense that no _correct_ program can observe the
1574 effects of the memory clear).
1576 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1577 to float instead of double. This does not affect the ABI of any libraries
1578 that are part of the GNU C Library, but may affect the ABI of other
1579 libraries that use this type in their interfaces.
1581 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1582 float_t and double_t types are now defined to long double instead of float
1583 and double. These options are not the default, and this does not affect
1584 the ABI of any libraries that are part of the GNU C Library, but it may
1585 affect the ABI of other libraries that use this type in their interfaces,
1586 if they are compiled or used with those options.
1588 * The getentropy and getrandom functions, and the <sys/random.h> header file
1591 * The buffer size for byte-oriented stdio streams is now limited to 8192
1592 bytes by default. Previously, on Linux, the default buffer size on most
1593 file systems was 4096 bytes (and thus remains unchanged), except on
1594 network file systems, where the buffer size was unpredictable and could be
1595 as large as several megabytes.
1597 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1598 for the Linux quota interface which predates kernel version 2.4.22 has
1601 * The malloc_get_state and malloc_set_state functions have been removed.
1602 Already-existing binaries that dynamically link to these functions will
1603 get a hidden implementation in which malloc_get_state is a stub. As far
1604 as we know, these functions are used only by GNU Emacs and this change
1605 will not adversely affect already-built Emacs executables. Any undumped
1606 Emacs executables, which normally exist only during an Emacs build, should
1607 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1609 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1610 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1611 “no-ip6-dotint” had already been the default, and support for the
1612 “ip6-dotint” option was removed from the Internet in 2006.
1614 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1615 flag from <resolv.h> have been removed. The option relied on a
1616 backwards-incompatible DNS extension which was never deployed on the
1619 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1620 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1621 They were already unimplemented.
1623 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1624 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1625 removed again from the IETF name lookup interface specification in RFC
1626 2553. Applications should use getaddrinfo instead.
1628 * DNSSEC-related declarations and definitions have been removed from the
1629 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1630 decode the data part of DNSSEC record types. Previous versions of glibc
1631 only implemented minimal support for the previous version of DNSSEC, which
1632 is incompatible with the currently deployed version.
1634 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1635 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1636 <arpa/nameser.h> header file because the distinction between RR types and
1637 meta-RR types is not officially standardized, subject to revision, and
1638 thus not suitable for encoding in a macro.
1640 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1641 and rhook members of the res_state type in <resolv.h> have been removed.
1642 The glibc stub resolver did not support these hooks, but the header file
1643 did not reflect that.
1645 * For multi-arch support it is recommended to use a GCC which has
1646 been built with support for GNU indirect functions. This ensures
1647 that correct debugging information is generated for functions
1648 selected by IFUNC resolvers. This support can either be enabled by
1649 configuring GCC with '--enable-gnu-indirect-function', or by
1650 enabling it by default by setting 'default_gnu_indirect_function'
1651 variable for a particular architecture in the GCC source file
1654 * GDB pretty printers have been added for mutex and condition variable
1655 structures in POSIX Threads. When installed and loaded in gdb these pretty
1656 printers show various pthread variables in human-readable form when read
1657 using the 'print' or 'display' commands in gdb.
1659 * Tunables feature added to allow tweaking of the runtime for an application
1660 program. This feature can be enabled with the '--enable-tunables' configure
1661 flag. The GNU C Library manual has details on usage and README.tunables has
1662 instructions on adding new tunables to the library.
1664 * A new version of condition variables functions have been implemented in
1665 the NPTL implementation of POSIX Threads to provide stronger ordering
1668 * A new version of pthread_rwlock functions have been implemented to use a more
1669 scalable algorithm primarily through not using a critical section anymore to
1672 Security related changes:
1674 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1675 have been created with makecontext could fail to terminate due to a
1676 missing .cantunwind annotation. This has been observed to lead to a hang
1677 (denial of service) in some Go applications compiled with gccgo. Reported
1678 by Andreas Schwab. (CVE-2016-6323)
1680 * The DNS stub resolver functions would crash due to a NULL pointer
1681 dereference when processing a query with a valid DNS question type which
1682 was used internally in the implementation. The stub resolver now uses a
1683 question type which is outside the range of valid question type values.
1686 The following bugs are resolved with this release:
1688 [4099] stdio: Overly agressive caching by stream i/o functions.
1689 [7065] build: Support building glibc with -fstack-protector or -fstack-
1691 [9842] localedata: en_CA: incorrect date format
1692 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1693 before it started waiting
1694 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1695 [15765] nptl: sem_open is wrongly a cancellation point
1696 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1697 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1699 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1700 library linked with pthread
1701 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1702 "FIXME: Ingo" issue)
1703 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1704 [17252] libc: getrandom and getentropy syscall
1705 [17730] malloc: thread-local storage is sometimes improperly free()'d
1706 after being __libc_memalign()'d
1707 [18241] stdio: failed fseek on memstream does not set errno and can fail
1709 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1711 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1712 PTHREAD_PRIO_INHERIT mutex on ARM
1713 [18784] network: res_query and related function crash for special record
1714 type queries (CVE-2015-5180)
1715 [19380] math: strtod does not raise "inexact"
1716 [19387] string: Integer overflow in memchr
1717 [19390] string: Integer overflow in strncat
1718 [19398] build: linknamespace tests fail with massively parallel build
1719 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1721 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1722 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1723 [19514] libc: [PATCH] Fix spelling errors in spelling
1724 "implement"/"implementation" in several places
1725 [19582] network: Deprecate RES_USE_INET6
1726 [19673] manual: clog10 docs appear to be erroneous
1727 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1728 causes a segmentation fault
1729 [19826] libc: invalid pointer returned from __tls_get_addr with static
1731 [20016] network: resolv: Remove hooks support from the API
1732 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1734 [20033] math: [x86_64] vectorized math function don't call the __finite
1736 [20116] nptl: use after free in pthread_create
1737 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1738 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1740 [20311] nptl: please install proc_service.h
1741 [20366] libc: Compilation errors in installed headers in strict-compliance
1743 [20370] malloc: malloc: Arena free list management is still racy
1744 (incorrect fix in bug 19243)
1745 [20386] libc: assert (X = 0) does not result in GCC warning
1746 [20432] malloc: malloc: Minimize interface required for interposition
1747 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1748 in _Unwind_Backtrace (CVE-2016-6323)
1749 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1750 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1752 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1753 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1754 [20477] network: resolv: incorrect double-checked locking related to
1756 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1758 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1760 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1761 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1762 cause transition penalty
1763 [20517] math: sparcv9 missing fdiml compat symbol
1764 [20524] manual: strverscmp is inconsistent
1765 [20525] libc: <sys/quota.h> should be based on kernel headers
1766 [20539] math: GCC 7 -static -lm fails to link at -O3
1767 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1769 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1771 [20591] network: Remove obsolete DNSSEC support
1772 [20592] network: DNS resource record type classification macros in
1773 <arpa/nameser.h> are incorrect
1774 [20593] network: Update DNS RR type definitions
1775 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1776 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1778 [20629] network: libresolv: Remove support for bitstring labels
1780 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1781 assertion: "maxidx >= 2"
1782 [20660] math: [arm] Use VSQRT
1783 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1784 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1785 [20689] libc: Test for FMA should also check for AVX.
1786 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1787 [20715] math: iszero macro breaks existing code
1788 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1789 [20728] libc: powerpc: Missing TOC stub in clone
1790 [20729] build: glibc-2.24 fails to build for i486 with -Os
1791 [20750] build: Build fails with default PIE enabled gcc-6.x
1792 [20768] math: [alpha] sqrt fegetenv namespace
1793 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1795 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1796 __FLT_EVAL_METHOD__ is 2
1797 [20790] network: rpcgen buffer overrun in get_prog_declaration
1798 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1799 [20829] libc: crypt snprintf namespace
1800 [20847] libc: tst-vfork3 failure
1801 [20855] math: Default bits/mathdef.h has inappropriate float_t
1802 [20859] math: [sh4] FP_ILOGB0 invalid
1803 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1804 U+20AC), not same as GBK
1805 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1806 [20916] math: pow handling of sNaN arguments
1807 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1808 [20919] math: Bad pow (qNaN, 0) result with -lieee
1809 [20940] math: hypot sNaN handling
1810 [20947] math: fmax, fmin sNaN handling
1811 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1812 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1814 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1815 [20973] nptl: robust mutexes: Lost wake-ups
1816 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1817 [20978] nis: strlen on null pointer in nss_nisplus
1818 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1820 [21014] string: i686 memchr overflows internal pointer check
1821 [21019] libc: [mips] n32 lseek incorrect on overflow
1822 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1823 [21026] libc: [MIPS] readahead syscall is broken on n64
1824 [21028] math: Fallback fesetexceptflag should always succeed
1825 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1826 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1828 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1829 [21061] librt: [microblaze] librt lost clock_* exports
1830 [21073] libc: tunables: insecure environment variables passed to
1831 subprocesses with AT_SECURE
1832 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1836 * The minimum Linux kernel version that this version of the GNU C Library
1837 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1838 kernel version 2.6.32 or later suffices (on architectures that already
1839 required kernel versions more recent than 3.2, those requirements remain
1840 unchanged). Linux 3.2 or later kernel headers are required on all
1843 * The pap_AN locale has been deleted. This has been deprecated for a long
1844 time. It has been replaced by pap_AW & pap_CW, both of which have long
1845 been included in previous releases.
1847 * The readdir_r and readdir64_r functions have been deprecated. It is
1848 recommended to use readdir and readdir64 instead.
1850 * The type “union wait” has been removed. It was deprecated in the early
1851 1990s and never part of POSIX. Application code should use the int type
1852 instead of “union wait”.
1854 * A new NSS action is added to facilitate large distributed system
1855 administration. The action, MERGE, allows remote user stores like LDAP
1856 to be merged into local user stores like /etc/groups in order to provide
1857 easy to use, updated, and managed sets of merged credentials. The new
1858 action can be used by configuring it in /etc/nsswitch.conf:
1859 group: files [SUCCESS=merge] nis
1860 Implemented by Stephen Gallagher (Red Hat).
1862 * The deprecated __malloc_initialize_hook variable has been removed from the
1865 * The long unused localedef --old-style option has been removed. It hasn't
1866 done anything in over 16 years. Scripts using this option can safely
1869 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1870 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1871 return the next representable value in the direction of positive infinity
1872 and the nextdown functions return the next representable value in the
1873 direction of negative infinity. These are currently enabled as GNU
1876 Security related changes:
1878 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1879 could result in a stack overflow when getnetbyname was called with an
1880 overly long name. (CVE-2016-3075)
1882 * Previously, getaddrinfo copied large amounts of address data to the stack,
1883 even after the fix for CVE-2013-4458 has been applied, potentially
1884 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1885 instead. Reported by Michael Petlan. (CVE-2016-3706)
1887 * The glob function suffered from a stack-based buffer overflow when it was
1888 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1889 Reported by Alexander Cherepanov. (CVE-2016-1234)
1891 * The Sun RPC UDP client could exhaust all available stack space when
1892 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1893 alloca plugin for GCC. (CVE-2016-4429)
1895 * The IPv6 name server management code in libresolv could result in a memory
1896 leak for each thread which is created, performs a failing naming lookup,
1897 and exits. Over time, this could result in a denial of service due to
1898 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1900 The following bugs are resolved with this release:
1902 [1170] localedata: ne_NP: update Nepali locale definition file
1903 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1905 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1906 [6796] math: fdim() does not set errno on overflow
1907 [10354] libc: posix_spawn should use vfork() in more cases than presently
1908 [11213] localedata: localedata: add copyright disclaimer to locale files
1909 [12143] localedata: chr_US: new Cherokee locale
1910 [12450] localedata: sgs_LT: new locale
1911 [12676] localedata: ln_CD: new locale
1912 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1914 [13304] math: fma, fmaf, fmal produce wrong results
1915 [14259] build: --localedir arg to configure is ignored
1916 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1917 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1918 [14934] localedata: es_CL: wrong first weekday chilean locale
1919 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1921 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1923 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1924 [15368] nptl: raise() is not async-signal-safe
1925 [15479] math: ceil, floor, round and trunc raise inexact exception
1926 [15578] localedata: kk_KZ: various updates
1927 [16003] localedata: pap_AN: punt old locale
1928 [16137] localedata: iw_IL: punt old locale
1929 [16190] localedata: eo: new esperanto locale
1930 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1931 [16742] malloc: race condition: pthread_atfork() called before first
1932 malloc() results in unexpected locking behaviour/deadlocks
1933 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1935 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1936 [17565] localedata: pt_PT: wrong (work-)week start
1937 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1938 [17950] build: Build fails with -msse
1939 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1940 [18433] libc: posix_spawn does not return correctly upon failure to
1942 [18453] localedata: charmaps/IBM875: incorrect codes
1943 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1945 [18896] localedata: he_IL: improvements for currency
1946 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1947 Romanian locale data
1948 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1950 [19056] libc: Deprecate readdir_r
1951 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1953 [19198] localedata: nl_NL: small improvements for Dutch locales
1954 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1955 nameservers (CVE-2016-5417)
1956 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1958 [19400] locale: Language missing in "iso-639.def", trivial fix in
1960 [19431] malloc: Deadlock between fflush, getdelim, and fork
1961 [19505] libc: Incorrect file descriptor validity checks in
1962 posix_spawn_file_actions_add{open,close,dup2}
1963 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1964 when using RTLD_NEXT
1965 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1966 `intl/{gettextP,loadinfo}.h'
1967 [19534] libc: execle, execlp may use malloc
1968 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1969 [19573] network: res_nclose and __res_maybe_init disagree about name
1970 server initialization, breaking Hesiod
1971 [19575] localedata: Status of GB18030 tables
1972 [19581] localedata: sr_* date_fmt string contains additional newline
1973 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1975 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1976 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1977 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1978 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1980 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1982 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1984 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1985 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1986 [19633] locale: strfmon_l applies global locale to number formatting
1987 [19642] network: Memory leak in getnameinfo
1988 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1989 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1991 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1992 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1994 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1995 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1997 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1999 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2001 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2002 pointers and lengths in error-case.
2003 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2004 report errors on UTF-16 surrogates values.
2005 [19755] nscd: nscd assertion failure in gc
2006 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2007 [19759] libc: mempcpy shouldn't be inlined
2008 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2009 [19765] libc: s390 needs an optimized mempcpy
2010 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2011 NAME_MAX limit assumption (CVE-2016-1234)
2012 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2013 [19787] network: Missing and incorrect truncation checks in getnameinfo
2014 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2016 [19791] network: Assertion failure in res_query.c with un-connectable name
2018 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2019 [19822] math: libm.so install clobbers old version
2020 [19825] network: resolv: send_vc can return uninitialized data in second
2021 response to getaddrinfo
2022 [19830] network: nss_dns: should check RDATA length against buffer length
2023 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2024 confronted with A/AAAA records of invalid size
2025 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2026 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2027 [19853] stdio: Printing IBM long double in decimal with high precision is
2029 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2030 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2031 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2033 [19865] network: Assertion failure or memory leak in
2034 _nss_dns_getcanonname_r
2035 [19868] network: nss_dns: netent code does not skip over non-PTR records
2036 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2038 [19881] string: Improve x86-64 memset
2039 [19907] string: Incorrect memcpy tests
2040 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2042 [19925] libc: termios.h XCASE namespace
2043 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2044 [19929] libc: limits.h NL_NMAX namespace
2045 [19931] stdio: Memory leak in vfprintf
2046 [19957] libc: clone(CLONE_VM) access invalid parent memory
2047 [19963] localedata: en_IL: New locale
2048 [19989] stdio: stdio.h cuserid namespace
2049 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2051 [19996] locale: langinfo.h nl_langinfo_l namespace
2052 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2053 [20010] network: getaddrinfo: Stack overflow in hostent translation
2055 [20012] stdio: libio: fmemopen append mode failure
2056 [20014] stdio: stdio.h namespace for pre-threads POSIX
2057 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2058 [20023] libc: fcntl.h timespec namespace
2059 [20024] math: [x86_64] vectorized sincos trashes the stack
2060 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2061 [20041] time: sys/time.h timespec namespace
2062 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2063 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2064 [20051] libc: ttyslot in wrong header under wrong conditions
2065 [20054] libc: gethostname not declared for XPG4
2066 [20055] libc: termios.h missing tcgetsid for XPG4
2067 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2069 [20073] libc: sys/stat.h fchmod namespace
2070 [20074] libc: stdlib.h rand_r namespace
2071 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2072 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2074 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2075 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2077 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2078 [20119] libc: Wrong mask for processors level type from CPUID
2079 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2080 AS not supporting AVX512
2081 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2083 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2084 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2086 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2087 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2088 [20178] libc: posix_spawn{p} should not call exit
2089 [20191] stdio: libio: vtables hardening
2090 [20195] string: FMA4 detection requires CPUID execution with register
2092 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2093 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2095 [20212] math: acos (sNaN) returns sNaN
2096 [20213] math: asin (sNaN) returns sNaN
2097 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2098 [20218] math: [i386] asinhl (sNaN) returns sNaN
2099 [20219] math: [i386] atanhl (sNaN) returns sNaN
2100 [20222] stdio: fopencookie: Mangle function pointers
2101 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2102 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2103 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2105 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2106 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2107 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2108 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2109 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2110 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2111 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2112 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2113 [20240] math: modf (sNaN) returns sNaN
2114 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2115 [20250] math: frexp (sNaN) returns sNaN
2116 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2117 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2119 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2120 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2121 subscript is below array bounds [-Werror=array-bounds]
2122 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2123 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2125 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2126 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2128 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2129 "invalid" exceptions
2130 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2131 [20316] localedata: id_ID: Februari instead of Pebruari
2132 [20327] string: POWER8 strcasecmp returns incorrect result
2133 [20347] math: Failure: Test: j0_downward (0xap+0)
2134 [20348] libc: FAIL: misc/tst-preadvwritev64
2135 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2136 [20350] libc: There is no test for p{read,write}64
2137 [20357] math: Incorrect cos result for 1.5174239687223976
2138 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2142 * Unicode 8.0.0 Support: Character encoding, character type info, and
2143 transliteration tables are all updated to Unicode 8.0.0, using new
2144 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2145 These updates cause user visible changes, such as the fixes for bugs
2146 89, 16061, and 18568.
2148 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2149 kernel-internal CPU set size. This means that requests that change the
2150 CPU affinity which failed before (for example, an all-ones CPU mask) will
2151 now succeed. Applications that need to determine the effective CPU
2152 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2153 setting it because the kernel can adjust it (and the previous size check
2154 would not detect this in the majority of cases).
2156 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2157 the following new symbols are used: fts64_children, fts64_close,
2158 fts64_open, fts64_read and fts64_set.
2160 * getaddrinfo now detects certain invalid responses on an internal netlink
2161 socket. If such responses are received, an affected process will
2162 terminate with an error message of "Unexpected error <number> on netlink
2163 descriptor <number>" or "Unexpected netlink response of size <number> on
2164 descriptor <number>". The most likely cause for these errors is a
2165 multi-threaded application which erroneously closes and reuses the netlink
2166 file descriptor while it is used by getaddrinfo.
2168 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2169 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2170 unnecessary serialization of memory allocation requests across threads.
2171 The defect is now corrected. Users should see a substantial increase in
2172 the concurent throughput of allocation requests for applications which
2173 trigger this bug. Affected applications typically create and
2174 destroy threads frequently. (Bug 19048 was reported and analyzed by
2177 * There is now a --disable-timezone-tools configure option for disabling the
2178 building and installing of the timezone related utilities (zic, zdump, and
2179 tzselect). This is useful for people who build the timezone data and code
2180 independent of the GNU C Library.
2182 * The obsolete header <regexp.h> has been removed. Programs that require
2183 this header must be updated to use <regex.h> instead.
2185 * The obsolete functions bdflush, create_module, get_kernel_syms,
2186 query_module and uselib are no longer available to newly linked binaries;
2187 the header <sys/kdaemon.h> has been removed. These functions and header
2188 were specific to systems using the Linux kernel and could not usefully be
2189 used with the GNU C Library on systems with version 2.6 or later of the
2192 * Optimized string, wcsmbs and memory functions for IBM z13.
2193 Implemented by Stefan Liebler.
2195 * Newly linked programs that define a variable called signgam will no longer
2196 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2197 require signgam to be set by those functions must ensure that they use the
2198 variable provided by the GNU C Library and declared in <math.h>, without
2199 defining their own copy.
2201 * The minimum GCC version that can be used to build this version of the GNU
2202 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2203 still be used to compile programs using the GNU C Library.
2205 Security related changes:
2207 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2208 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2210 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2211 the pointer guard feature. It is always enabled. Previously,
2212 LD_POINTER_GUARD could be used to disable security hardening in binaries
2213 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2216 * An integer overflow in hcreate and hcreate_r could lead to an
2217 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2219 * The catopen function no longer has unbounded stack usage. Reported by
2220 Max. (CVE-2015-8779)
2222 * The nan, nanf and nanl functions no longer have unbounded stack usage
2223 depending on the length of the string passed as an argument to the
2224 functions. Reported by Joseph Myers. (CVE-2014-9761)
2226 * A stack-based buffer overflow was found in libresolv when invoked from
2227 libnss_dns, allowing specially crafted DNS responses to seize control
2228 of execution flow in the DNS client. The buffer overflow occurs in
2229 the functions send_dg (send datagram) and send_vc (send TCP) for the
2230 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2231 family. The use of AF_UNSPEC triggers the low-level resolver code to
2232 send out two parallel queries for A and AAAA. A mismanagement of the
2233 buffers used for those queries could result in the response of a query
2234 writing beyond the alloca allocated buffer created by
2235 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2236 the overflow. Thanks to the Google Security Team and Red Hat for
2237 reporting the security impact of this issue, and Robert Holiday of
2238 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2240 The following bugs are resolved with this release:
2242 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2243 [887] math: Math library function "logb" and "nextafter" inconsistent
2244 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2245 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2246 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2247 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2249 [4404] localedata: German translation of "Alarm clock" is misleading
2250 [6799] math: nextafter() and nexttoward() doen't set errno on
2251 overflow/underflow errors
2252 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2254 [10432] nis: _nss_nis_setnetgrent assertion failure
2255 [11460] libc: fts has no LFS support
2256 [12926] network: getaddrinfo()/make_request() may spin forever
2257 [13065] nptl: Race condition in pthread barriers
2258 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2259 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2261 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2262 [14912] libc: Rename non-installed bits/*.h headers
2263 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2264 [15367] math: Let gcc use __builtin_isinf
2265 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2266 [15421] math: lgamma wrongly sets signgam for ISO C
2267 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2268 FE_INVALID with argument out of range
2269 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2271 [15786] dynamic-link: ifunc resolver functions can smash function
2273 [15918] math: Unnecessary check for equality in hypotf()
2274 [16061] localedata: Review / update transliteration data
2275 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2277 [16141] time: strptime %z offset restriction
2278 [16171] math: drem should be alias of remainder
2279 [16296] math: fegetround is pure?
2280 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2281 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2283 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2284 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2286 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2287 "Exception "Inexact" set" on ppc32
2288 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2289 [16517] math: Missing underflow exception from tanf/tan/tanl
2290 [16519] math: Missing underflow exception from sinhf
2291 [16520] math: Missing underflow exception from tanhf
2292 [16521] math: Missing underflow exception from exp2
2293 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2295 [16734] stdio: fopen calls mmap to allocate its buffer
2296 [16961] math: nan function incorrect handling of bad sequences
2297 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2298 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2299 [16985] locale: localedef: confusing error message when opening output
2301 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2302 [17197] locale: Redundant shift character in iconv conversion output at
2304 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2305 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2306 [17250] dynamic-link: static linking breaks nss loading
2307 (getaddrinfo/getpwnam/etc...)
2308 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2310 [17441] math: isnan() should use __builtin_isnan() in GCC
2311 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2312 (related to lock elision)
2313 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2314 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2315 [17887] time: strptime should be able to parse "+01:00" style timezones
2316 [17905] libc: catopen() Multiple unbounded stack allocations
2318 [18084] libc: backtrace (..., 0) dumps core on x86
2319 [18086] libc: nice() sets errno to 0 on success
2320 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2321 is too large (CVE-2015-8778)
2322 [18251] dynamic-link: SONAME missing when audit modules provides path
2323 [18265] libc: add attributes for wchar string and memory functions
2324 [18370] math: csqrt missing underflows
2325 [18421] libc: [hppa] read-only segment has dynamic relocations
2326 [18472] libc: Obsolete syscall wrappers should be compat symbols
2327 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2328 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2329 [18525] localedata: Remove locale timezone information
2330 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2331 [18568] localedata: Update locale data to Unicode 8.0
2332 [18589] locale: sort-test.sh fails at random
2333 [18595] math: ctan, ctanh missing underflows
2334 [18604] libc: assert macro-expands its argument
2335 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2336 contains a vector instruction exception.
2337 [18611] math: j1, jn missing errno setting on underflow
2338 [18618] localedata: sync Chechen locale definitions with other *_RU
2340 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2341 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2342 [18665] network: In send_dg, the recvfrom function is NOT always using the
2343 buffer size of a newly created buffer (CVE-2015-7547)
2344 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2346 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2348 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2349 [18699] math: tilegx cproj() for various complex infinities does not yield
2351 [18724] libc: Harden put*ent functions against data injection
2352 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2353 [18755] build: build errors with -DNDEBUG
2354 [18757] stdio: fmemopen fails to set errno on failure
2355 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2357 [18781] libc: openat64 lacks O_LARGEFILE
2358 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2359 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2360 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2361 [18790] math: [ldbl-128ibm] tanhl inaccurate
2362 [18795] libc: stpncpy fortification misses buffer lengths that are
2363 statically too large
2364 [18796] build: build fails for --disable-mathvec
2365 [18803] math: hypot missing underflows
2366 [18820] stdio: fmemopen may leak memory on failure
2367 [18823] math: csqrt spurious underflows
2368 [18824] math: fma spurious underflows
2369 [18825] math: pow missing underflows
2370 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2371 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2372 [18870] build: sem_open.c fails to compile with missing symbol
2374 [18872] stdio: Fix memory leak in printf_positional
2375 [18873] libc: posix_fallocate overflow check ineffective
2376 [18875] math: Excess precision leads incorrect libm
2377 [18877] libc: arm: mmap offset regression
2378 [18887] libc: memory corruption when using getmntent on blank lines
2379 [18918] localedata: hu_HU: change time to HH:MM:SS format
2380 [18921] libc: Regression: extraneous stat() and fstat() performed by
2382 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2383 binaries (CVE-2015-8777)
2384 [18951] math: tgamma missing underflows
2385 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2387 [18953] localedata: lt_LT: change currency symbol to the euro
2388 [18956] math: powf inaccuracy
2389 [18961] math: [i386] exp missing underflows
2390 [18966] math: [i386] exp10 missing underflows
2391 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2392 [18969] build: multiple string test failures due to missing locale
2394 [18970] libc: Reference of pthread_setcancelstate in libc.a
2395 [18977] math: float / long double Bessel functions not in XSI POSIX
2396 [18980] math: i386 libm functions return with excess range and precision
2397 [18981] math: i386 scalb*, ldexp return with excess range and precision
2398 [18982] stdio: va_list and vprintf
2399 [18985] time: Passing out of range data to strftime() causes a segfault
2401 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2402 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2403 [19012] locale: iconv_open leaks memory on error path
2404 [19016] math: clog, clog10 inaccuracy
2405 [19018] nptl: Mangle function pointers in tls_dtor_list
2406 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2407 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2408 [19048] malloc: malloc: arena free list can become cyclic, increasing
2410 [19049] math: [powerpc] erfc incorrect zero sign
2411 [19050] math: [powerpc] log* incorrect zero sign
2412 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2413 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2414 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2415 [19074] network: Data race in _res_hconf_reorder_addrs
2416 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2417 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2418 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2420 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2422 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2423 [19086] manual: posix_fallocate64 documented argument order is wrong.
2424 [19088] math: lround, llround missing exceptions close to overflow
2426 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2427 [19095] math: dbl-64 lrint incorrect for 64-bit long
2428 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2429 [19124] dynamic-link: ld.so failed to build with older assmebler
2430 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2431 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2432 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2433 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2434 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2435 pthread_setaffinity_np
2436 [19156] math: [ldbl-128] j0l spurious underflows
2437 [19164] nptl: tst-getcpu fails with many possible CPUs
2438 [19168] math: math/test-ildoubl and math/test-ldouble failure
2439 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2440 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2442 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2443 (FE_NOMASK_ENV) do not clear SSE exceptions
2444 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2445 _int_new_arena/reused_arena
2446 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2447 [19201] math: dbl-64 remainder incorrect sign of zero result
2448 [19205] math: bits/math-finite.h conditions do not match math.h and
2450 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2451 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2452 for C99-based standards
2453 [19212] libc: features.h not -Wundef clean
2454 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2456 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2457 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2458 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2460 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2461 "inexact" exceptions
2462 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2464 [19242] libc: strtol incorrect in Turkish locales
2465 [19243] malloc: reused_arena can pick an arena on the free list, leading
2466 to an assertion failure and reference count corruption
2467 [19253] time: tzset() ineffective when temporary TZ did not include DST
2469 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2470 [19270] math: [hppa] Shared libm missing __isnanl
2471 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2472 MAP_HUGETLB and MAP_STACK defines
2473 [19313] nptl: Wrong __cpu_mask for x32
2474 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2475 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2476 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2477 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2478 [19363] time: x32: times() return value wrongly truncates/sign extends
2480 [19367] dynamic-link: Improve branch prediction on Silvermont
2481 [19369] network: Default domain name not reset by res_ninit when "search"
2482 / "domain" entry is removed from resolv.conf
2483 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2485 [19385] localedata: bg_BG: time separator should be colon, not comma
2486 [19408] libc: linux personality syscall wrapper may erroneously return an
2487 error on 32-bit architectures
2488 [19415] libc: dladdr returns wrong names on hppa
2489 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2490 IBM905, IBM907, and IBM909
2491 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2492 [19443] build: build failures with -DDEBUG
2493 [19451] build: Make check fails on test-double-vlen2
2494 [19462] libc: Glibc failed to build with -Os
2495 [19465] math: Wrong code with -Os
2496 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2498 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2500 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2501 [19486] math: S390: Math tests fail with "Exception Inexact set".
2502 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2503 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2505 [19590] math: Fail to build shared objects that use libmvec.so functions.
2509 * The following bugs are resolved with this release:
2511 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2512 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2513 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2514 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2515 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2516 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2517 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2518 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2519 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2520 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2521 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2522 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2523 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2524 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2525 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2526 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2527 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2528 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2529 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2530 18657, 18676, 18694, 18696, 18887.
2532 * Cache information can be queried via sysconf() function on s390 e.g. with
2533 _SC_LEVEL1_ICACHE_SIZE as argument.
2535 * A buffer overflow in gethostbyname_r and related functions performing DNS
2536 requests has been fixed. If the NSS functions were called with a
2537 misaligned buffer, the buffer length change due to pointer alignment was
2538 not taken into account. This could result in application crashes or,
2539 potentially arbitrary code execution, using crafted, but syntactically
2540 valid DNS responses. (CVE-2015-1781)
2542 * The time zone file parser has been made more robust against crafted time
2543 zone files, avoiding heap buffer overflows related to the processing of
2544 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2545 large time zone data files. Overly long time zone specifiers in the TZ
2546 variable no longer result in stack overflows and crashes.
2548 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2549 for LD and GD on x86 and x86-64, has been implemented. You will need
2550 binutils-2.24 or later to enable this optimization.
2552 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2553 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2554 Hat). These updates cause user visible changes, such as the fix for bug
2557 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2558 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2559 condition in some applications.
2561 * Added vector math library named libmvec with the following vectorized x86_64
2562 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2564 The library can be disabled with --disable-mathvec. Use of the functions is
2565 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2566 Shared library libmvec.so is linked in as needed when using -lm (no need to
2567 specify -lmvec explicitly for not static builds).
2568 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2570 * A new fmemopen implementation has been added with the goal of POSIX
2571 compliance. The new implementation fixes the following long-standing
2572 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2573 old implementation is still present for use be by existing binaries.
2575 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2576 releases. It has been fixed to match 2.19 and older, but binaries built
2577 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2579 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2580 Contributed by Roland McGrath (Google).
2582 * The header <regexp.h> is deprecated, and will be removed in a future
2583 release. Use of this header will trigger a deprecation warning.
2584 Application developers should update their code to use <regex.h> instead.
2586 This header was formerly part of SUS, but was deprecated in 1994 and
2587 removed from the standard in 2001. Also, the glibc implementation
2588 leaks memory. See BZ#18681 for more details.
2592 * The following bugs are resolved with this release:
2594 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2595 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2596 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2597 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2598 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2599 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2600 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2601 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2602 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2603 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2606 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2607 too little memory for the to-be-scanned arguments and overflow the
2608 allocated buffer. The implementation now correctly computes the required
2609 buffer size when using malloc, and switches to malloc from alloca as
2612 * A new semaphore algorithm has been implemented in generic C code for all
2613 machines. Previous custom assembly implementations of semaphore were
2614 difficult to reason about or ensure that they were safe. The new version
2615 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2616 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2617 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2619 * Port to Altera Nios II has been contributed by Mentor Graphics.
2621 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2622 implementations for powerpc64/powerpc64le.
2623 Implemented by Adhemerval Zanella (IBM).
2625 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2626 and powerpc64le. This may improve lock scaling of existing programs on
2627 HTM capable systems. The lock elision code is only enabled with
2628 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2629 powerpc will issue a transaction abort on every syscall to avoid side
2630 effects being visible outside transactions.
2632 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2633 AArch64. Contributed by ARM Ltd.
2635 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2637 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2638 under certain input conditions resulting in the execution of a shell for
2639 command substitution when the application did not request it. The
2640 implementation now checks WRDE_NOCMD immediately before executing the
2641 shell and returns the error WRDE_CMDSUB as expected.
2643 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2644 processing format strings with a large number of format specifiers.
2646 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2647 infinite loop if the DNS response contained a PTR record of an unexpected
2650 * The minimum GCC version that can be used to build this version of the GNU
2651 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2652 still be used to compile programs using the GNU C Library.
2654 * The GNU C Library is now built with -Werror by default. This can be
2655 disabled by configuring with --disable-werror.
2657 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2659 * The obsolete sigvec function has been removed. This was the original
2660 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2661 programs have been using instead for about 25 years. Of course, ABI
2662 compatibility for old binaries using sigvec remains intact.
2664 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2665 with newer versions of bison.
2667 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2668 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2669 registers overlay two consecutive single-precision registers. MIPS32R2
2670 introduced a new FPU mode (FR=1) where double-precision registers extend the
2671 corresponding single-precision registers which is incompatible with the
2672 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2673 require the use of FR=1 making a transition necessary. New o32 ABI
2674 extensions enable users to migrate over time from the original o32 ABI
2675 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2676 tracks the ABI of any loaded object and verifies that new objects are
2677 compatible. Mode transitions will also be requested as required and
2678 unsupportable objects will be rejected. The ABI checks include both soft and
2679 hard float ABIs for o32, n32 and n64.
2681 GCC 5 with GNU binutils 2.25 onwards:
2682 It is strongly recommended that all o32 system libraries are built using the
2683 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2684 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2685 compiler using --with-fp-32=xx to set this by default.
2689 * The following bugs are resolved with this release:
2691 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2692 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2693 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2694 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2695 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2696 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2697 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2698 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2699 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2700 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2701 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2702 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2703 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2704 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2705 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2707 * Reverted change of ABI data structures for s390 and s390x:
2708 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2709 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2710 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2711 are preserved pointing straight to the same implementation as the old ones.
2712 Given that, new callers will simply provide a too-big buffer to these
2713 functions. Any applications/libraries out there that embed jmp_buf or
2714 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2715 against 2.19 headers will have to rebuilt again. This is necessary in any
2716 case to revert the breakage in their ABI caused by the glibc change.
2718 * Support for file description locks is added to systems running the
2719 Linux kernel. The standard file locking interfaces are extended to
2720 operate on file descriptions, not file descriptors, via the use of
2721 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2722 are associated with an open file instead of a process.
2724 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2726 * The minimum Linux kernel version that this version of the GNU C Library
2727 can be used with is 2.6.32.
2729 * Running the testsuite no longer terminates as soon as a test fails.
2730 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2731 with PASS or FAIL lines for individual tests. A summary of the results is
2732 printed, including a list of failing lists, and "make check" exits with
2733 error status if there were any unexpected failures. "make check
2734 stop-on-test-failure=y" may be used to keep the old behavior.
2736 * The am33 port, which had not worked for several years, has been removed
2739 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2740 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2741 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2742 conflicted with POSIX (support for which was removed in 2.19), the
2743 interfaces those macros enabled remain available when compiling with
2744 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2745 test macros defined.
2747 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2749 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2750 This may improve lock scaling of existing programs on TX capable systems.
2751 The lock elision code is only built with --enable-lock-elision=yes and
2752 then requires a GCC version supporting the TX builtins. With lock elision
2753 default mutexes are elided via __builtin_tbegin, if the cpu supports
2754 transactions. By default lock elision is not enabled and the elision code
2757 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2758 copy the path argument. This allowed programs to cause posix_spawn to
2759 deference a dangling pointer, or use an unexpected pathname argument if
2760 the string was modified after the posix_spawn_file_actions_addopen
2763 * All supported architectures now use the main glibc sysdeps directory
2764 instead of some being in a separate "ports" directory (which was
2765 distributed separately before glibc 2.17).
2767 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2768 On configurations that support it (all Linux configurations), it's now
2769 used regardless of the --enable-add-ons switch to configure. It is no
2770 longer possible to build such configurations without pthreads support.
2772 * Locale names, including those obtained from environment variables (LANG
2773 and the LC_* variables), are more tightly checked for proper syntax.
2774 setlocale will now fail (with EINVAL) for locale names that are overly
2775 long, contain slashes without starting with a slash, or contain ".." path
2776 components. (CVE-2014-0475) Previously, some valid locale names were
2777 silently replaced with the "C" locale when running in AT_SECURE mode
2778 (e.g., in a SUID program). This is no longer necessary because of the
2781 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2782 with application code using Intel MPX instructions. (With all previous
2783 versions, the MPX register state could be clobbered when making calls
2784 into or out of a shared library.) Note that while the new dynamic
2785 linker is compatible with all known x86 hardware whether or not it
2786 supports Intel MPX, some x86 instruction-set emulators might fail to
2787 handle the new instruction encodings. This is known to affect Valgrind
2788 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2789 release), and might affect other tools that do instruction emulation.
2791 * Support for loadable gconv transliteration modules has been removed.
2792 The support for transliteration modules has been non-functional for
2793 over a decade, and the removal is prompted by security defects. The
2794 normal gconv conversion modules are still supported. Transliteration
2795 with //TRANSLIT is still possible, and the //IGNORE specifier
2796 continues to be supported. (CVE-2014-5119)
2798 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2799 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2800 resulting a denial-of-service security vulnerability in applications which
2801 use functions related to iconv. (CVE-2014-6040)
2805 * The following bugs are resolved with this release:
2807 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2808 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2809 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2810 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2811 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2812 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2813 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2814 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2815 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2816 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2817 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2818 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2819 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2820 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2821 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2822 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2823 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2824 16431, 16453, 16474, 16506, 16510, 16529
2826 * Slovenian translations for glibc messages have been contributed by the
2827 Translation Project's Slovenian team of translators.
2829 * The public headers no longer use __unused nor __block. This change is to
2830 support compiling programs that are derived from BSD sources and use
2831 __unused internally, and to support compiling with Clang's -fblock
2832 extension which uses __block.
2834 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2835 large collation sequences to optimize multiple passes. This cache
2836 computation may overflow for large collation sequences and may cause a
2837 stack or buffer overflow. This is now fixed to use a slower algorithm
2838 which does not use a cache if there is an integer overflow.
2840 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2841 rules for large collation sequences to optimize multiple passes and falls
2842 back to alloca if malloc fails, resulting in a possible stack overflow.
2843 The implementation now falls back to an uncached collation sequence lookup
2846 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2847 initialized for static applications resulting in the security feature
2848 being disabled. The pointer guard is now correctly initialized to a
2849 random value for static applications. Existing static applications need
2850 to be recompiled to take advantage of the fix (bug 15754).
2852 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2853 to the d_name member of struct dirent, or omit the terminating NUL
2854 character. (Bugzilla #14699).
2856 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2857 aligned_alloc functions could allocate too few bytes or corrupt the
2858 heap when passed very large allocation size values (Bugzilla #15855,
2861 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2862 for AF_INET6 has been fixed (Bugzilla #16072).
2864 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2865 pap_CW, quz_PE, the_NP.
2867 * Substantially revised locales: gd_GB, ht_HT
2869 * The LC_ADDRESS field was updated to support country_car for almost all
2872 * ISO 1427 definitions were updated.
2874 * ISO 3166 definitions were updated.
2876 * The localedef utility now supports --big-endian and --little-endian
2877 command-line options to generate locales for a different system from that
2878 for which the C library was built.
2880 * Binary locale files now only depend on the endianness of the system for
2881 which they are generated and not on other properties of that system. As a
2882 consequence, binary files generated with new localedef may be incompatible
2883 with old versions of the GNU C Library, and binary files generated with
2884 old localedef may be incompatible with this version of the GNU C Library,
2885 in the following circumstances:
2887 + Locale files may be incompatible on m68k systems.
2889 + Locale archive files (but not separate files for individual locales) may
2890 be incompatible on systems where plain "char" is signed.
2892 * The configure option --disable-versioning has been removed. Builds with
2893 --disable-versioning had not worked for several years.
2895 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2896 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2898 * SystemTap probes for malloc have been introduced.
2900 * SystemTap probes for slow multiple precision fallback paths of
2901 transcendental functions have been introduced.
2903 * Support for powerpc64le has been added.
2905 * The soft-float powerpc port now supports e500 processors.
2907 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2909 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2910 set of header declarations that are enabled by default, even when other
2911 feature test macros or compiler options such as -std=c99 would otherwise
2912 disable some of those declarations.
2914 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2915 conflict with POSIX. The libbsd-compat library (which was a dummy library
2916 that did nothing) has also been removed.
2918 * Preliminary documentation about Multi-Thread, Async-Signal and
2919 Async-Cancel Safety has been added.
2921 * Change of ABI data structures for s390 and s390x:
2922 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2923 allow for future hardware extensions. All library functions that accept or
2924 return these structures were versioned in order to provide backwards
2925 compatibility with existing code. However, not all cases can be handled
2926 with interface versioning. If an application embeds either structure into
2927 another structure and that structure is passed to another compilation unit
2928 which expects the newer larger-sized object then any access to the new
2929 fields will result in undefined behaviour. Similarly any access to fields
2930 that were moved by the enlarging of the structures will cause undefined
2931 behaviour. To fix the undefined behaviour all objects that are part of the
2932 final in-memory image must agree on the size of structures, and this may
2933 require recompilation.
2937 * The following bugs are resolved with this release:
2939 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2940 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2941 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2942 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2943 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2944 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2945 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2946 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2947 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2948 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2949 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2950 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2951 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2954 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2955 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2956 Distributions can re-enable building and using pt_chown via the new configure
2957 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2958 considerable security risks and should only be used if the distribution
2959 understands and accepts the risks.
2961 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2964 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2965 fixed (Bugzilla #15330).
2967 * Add support for calling C++11 thread_local object destructors on thread
2968 and program exit. This needs compiler support for offloading C++11
2969 destructor calls to glibc.
2971 * Improved worst case performance of libm functions with double inputs and
2974 * Support for priority inherited mutexes in pthread condition variables on
2975 non-x86 architectures.
2977 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2979 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2981 * Optimized string functions for ARM. Implemented by Will Newton and
2984 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2986 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2989 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2990 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2992 * Added a benchmark framework to track performance of functions in glibc.
2994 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2995 It is based on draft TS 18661 and currently enabled as a GNU extension.
2997 * On Linux, the clock function now uses the clock_gettime system call
2998 for improved precision, rather than old times system call.
3000 * Added support for version-3 tz format files. This is needed when using
3001 the tz database release 2013e or later, and affects a few unusual cases --
3002 currently only TZ='America/Godthab' for time stamps after 2037.
3004 * Added new API functions pthread_getattr_default_np and
3005 pthread_setattr_default_np to get and set the default pthread
3006 attributes of a process.
3008 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3009 This may improve lock scaling of existing programs on TSX capable systems.
3010 When the --enable-lock-elision=yes parameter is specified at configure
3011 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3014 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3015 availability). Implemented by Ryan S. Arnold.
3017 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3021 * The following bugs are resolved with this release:
3023 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3024 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3025 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3026 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3027 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3028 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3029 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3030 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3031 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3032 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3033 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3034 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3035 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3036 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3037 14879, 14889, 14893, 14898, 14914.
3039 * Optimization of memcpy for MIPS.
3041 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3042 EMFILE has been fixed (Bugzilla #14889).
3044 * The add-on ports collection is now distributed in the "ports" subdirectory
3045 of the main GNU C Library distribution, rather than separately.
3047 * Port to ARM AArch64 contributed by Linaro.
3049 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3050 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3052 Implemented by Andreas Krebbel.
3054 * The new function secure_getenv allows secure access to the environment,
3055 returning NULL if running in a SUID/SGID process. This function replaces
3056 the internal function __secure_getenv.
3058 * SystemTap static probes have been added into the dynamic linker.
3059 Implemented by Gary Benson.
3061 * Optimizations of string functions strstr, strcasestr and memmem.
3062 Implemented by Maxim Kuvyrkov.
3064 * The minimum Linux kernel version that this version of the GNU C Library
3065 can be used with is 2.6.16.
3067 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3068 powerpc POWER7. Implemented by Will Schmidt.
3070 * New configure option --disable-nscd builds the C library such that it
3071 never attempts to contact the Name Service Caching Daemon (nscd).
3072 New configure option --disable-build-nscd avoids building nscd itself;
3073 this is the default if --disable-nscd is used.
3075 * Improved support for cross-compilation, including cross-testing and
3076 bootstrap builds without a previously built glibc.
3078 * Several testsuite tests are now able to test multiple IFUNC variants of an
3079 interface, rather than just testing the one that would be chooen by
3082 * New configure options --with-bugurl and --with-pkgversion, for
3083 distributors to use to embed their bug-reporting and package version
3084 information in --help and --version output.
3086 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3087 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3088 allows creation of chroots without the procfs mounted on /proc.
3090 * The `crypt' function now fails if passed salt bytes that violate the
3091 specification for those values. On Linux, the `crypt' function will
3092 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3093 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3094 when the mode is enabled.
3096 * The `clock_*' suite of functions (declared in <time.h>) is now available
3097 directly in the main C library. Previously it was necessary to link with
3098 -lrt to use these functions. This change has the effect that a
3099 single-threaded program that uses a function such as `clock_gettime' (and
3100 is not linked with -lrt) will no longer implicitly load the pthreads
3101 library at runtime and so will not suffer the overheads associated with
3102 multi-thread support in other code such as the C++ runtime library.
3104 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3109 * The following bugs are resolved with this release:
3111 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3112 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3113 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3114 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3115 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3116 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3117 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3118 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3119 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3120 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3121 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3122 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3123 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3124 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3125 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3126 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3127 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3128 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3129 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3130 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3131 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3132 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3135 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3136 configuring glibc with:
3137 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3138 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3139 Implemented by H.J. Lu.
3143 + define static_assert
3145 + do not declare gets
3147 + declare at_quick_exit and quick_exit also for ISO C11
3149 + aligned_alloc. NB: The code is deliberately allows the size parameter
3150 to not be a multiple of the alignment. This is a moronic requirement
3151 in the standard but it is only a requirement on the caller, not the
3154 + timespec_get added
3156 + uchar.h support added
3158 + CMPLX, CMPLXF, CMPLXL added
3160 Implemented by Ulrich Drepper.
3162 * Support for the IA-64 has been moved to ports.
3164 * Remove support for anything but ELF binary format
3166 * Checking versions of poll, ppoll added.
3167 Implemented by Ulrich Drepper.
3169 * More generic and 64-bit performance optimizations to math functions.
3170 Implemented by Ulrich Drepper.
3172 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3173 headers and functions available at compile time as they were before
3174 version 2.14. This option will be removed at some time in the future
3175 after the TI-RPC library becomes fully sufficient for the needs of
3176 existing applications.
3178 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3179 Note that glibc is not expected to work with any Linux kernel version
3182 * New header <sys/auxv.h> and function getauxval allowing easy access to
3183 the AT_* key-value pairs passed from the Linux kernel. The header also
3184 defines the HWCAP_* bits associated with the AT_HWCAP key.
3186 * New locales: mag_IN
3188 * New configure option --enable-systemtap builds SystemTap static probes
3189 into libc for setjmp and longjmp and into libpthread for various operations.
3190 So far the setjmp/longjmp probes and some of the libpthread probes are
3191 provided only for i*86 and x86_64.
3192 Implemented by Roland McGrath and Rayson Ho.
3194 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3196 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3199 * More optimized functions for SPARC. Implemented by David S. Miller.
3201 * Improved support for cross-compilation, especially bootstrap builds
3202 without a previously built glibc.
3204 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3205 Chris Metcalf from Tilera.
3207 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3208 now supported for ARM processors.
3210 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3211 as the name of the dynamic linker, to distinguish it from the
3212 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3214 * Support for CRIS has been removed from ports.
3216 * A new class of installed header has been documented for low-level
3217 platform-specific functionality. PowerPC added the first instance with a
3218 function to provide time base register access. Contributed by Tulio
3219 Magno Quites Machado Filho.
3221 * ix86 configurations now install header files that are consistent with
3222 what x86-64 configurations install. These same header files can be used
3223 for -m32, -m64, or -mx32 builds.
3224 Contributed by H.J. Lu.
3226 * Math library bug fixes. A thorough audit of all open math library bugs was
3227 conducted by Joseph Myers. Significant progress was made on many math
3228 library bugs resulting in more accurate exceptions and function results.
3229 Many thanks to all those that contributed including Andreas Jaeger for his
3230 patch review and work on the x87 trigonometric instruction issues.
3232 * Timezone data is no longer installed. Timezone-related binaries and scripts
3233 will continue to be installed. Users should obtain their timezone data from
3234 their distribution provider or from the tzdata package at
3235 <ftp://munnari.oz.au/pub/>.
3239 * The following bugs are resolved with this release:
3241 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3242 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3243 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3244 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3245 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3246 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3247 13484, 13506, 13515, 13523, 13524, 13538, 13540
3249 * New program pldd to list loaded object of a process
3250 Implemented by Ulrich Drepper.
3252 * Add nss_db support back to glibc. No more dependency on Berkeley db
3253 and support for initgroups lookups.
3254 Implemented by Ulrich Drepper.
3256 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3257 Contributed by HJ Lu.
3259 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3260 Contributed by HJ Lu.
3262 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3263 on x86-32 and x86-64.
3264 Contributed by Liubov Dmitrieva.
3266 * Optimized strchr and strrchr for SSE on x86-32.
3267 Contributed by Liubov Dmitrieva.
3269 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3270 for x86-64 and x86-32.
3271 Contributed by Liubov Dmitrieva.
3273 * New interfaces: scandirat, scandirat64
3274 Implemented by Ulrich Drepper.
3276 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3277 Implemented by Ulrich Drepper.
3279 * nscd now also caches the netgroup database.
3280 Implemented by Ulrich Drepper.
3282 * Integrate libm with gcc's -ffinite-math-only option.
3283 Implemented by Ulrich Drepper.
3285 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3286 to math functions. Implemented by Ulrich Drepper.
3288 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3289 Implemented by Ulrich Drepper.
3291 * New Linux interfaces: process_vm_readv, process_vm_writev
3293 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3294 Implemented by Ulrich Drepper.
3296 * Optimized nearbyint and strcasecmp for PPC.
3297 Implemented by Adhemerval Zanella.
3299 * New locales: bho_IN, unm_US, es_CU, ta_LK
3303 * The following bugs are resolved with this release:
3305 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3306 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3307 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3308 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3309 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3310 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3311 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3312 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3313 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3314 12795, 12811, 12813, 12814, 12841
3316 * The RPC implementation in libc is obsoleted. Old programs keep working
3317 but new programs cannot be linked with the routines in libc anymore.
3318 Programs in need of RPC functionality must be linked against TI-RPC.
3319 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3321 Visible changes of this change include (obviously) the inability to link
3322 programs using RPC functions without referencing the TI-RPC library and the
3323 removal of the RPC headers from the glibc headers.
3324 Implemented by Ulrich Drepper.
3326 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3327 syncfs, setns, sendmmsg
3329 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3330 yue_HK, lij_IT, mhr_RU
3332 * New encodings: CP770, CP771, CP772, CP773, CP774
3334 * New program sotruss to trace calls through PLTs
3335 Implemented by Ulrich Drepper.
3337 * The malloc hook implementation is marked deprecated and will be removed
3338 from the default implementation in the next version. The design never
3339 worked ever since the introduction of threads. Even programs which do
3340 not create threads themselves can use multiple threads created internally.
3344 * The following bugs are resolved with this release:
3346 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3347 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3348 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3349 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3352 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3354 * POWER7 optimizations: memset, memcmp, strncmp
3356 * New optimized string functions for x86-64: strnlen (SSE2),
3357 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3358 Implemented by Ulrich Drepper.
3362 * The following bugs are resolved with this release:
3364 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3365 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3366 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3367 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3368 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3369 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3370 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3371 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3373 * New interfaces: pthread_getname_np, pthread_setname_np
3375 * New Linux interface: recvmmsg
3377 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3379 * The dynamic linker now recognizes supported ABI versions from the
3380 EI_ABIVERSION field in the ELF header.
3381 Implemented by Ulrich Drepper.
3383 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3384 will not be used to fill in password fields in the passwd.byname replies.
3385 Instead it is used to synthesize the shadow.byname table, should it be
3386 missing. This is a useful mode in some installations involving Solaris.
3387 Implemented by Ulrich Drepper.
3389 * New locales: kok_IN, sq_MK, cv_RU
3393 * The following bugs are resolved with this release:
3395 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3396 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3397 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3398 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3399 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3400 10789, 10817, 10823, 10840, 10847
3402 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3404 Implemented by Ulrich Drepper.
3406 * Checking version of longjmp added that fails if an uninitialized stack
3407 frame would be created. Implemented by Ulrich Drepper.
3409 * STT_GNU_IFUNC is now supported in static executables.
3410 Implemented by H.J. Lu.
3412 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3414 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3415 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3416 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3417 Contributed by H.J. Lu.
3419 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3420 Implemented by Ulrich Drepper.
3422 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3424 Contributed by H.J. Lu.
3426 * Support for fma instruction in AVX on x86-64.
3427 Implemented by H.J. Lu and Ulrich Drepper.
3429 * AVX support in x86-64 auditing support in ld.so.
3430 Implemented by H.J. Lu.
3432 * STB_GNU_UNIQUE support added.
3433 Implemented by Ulrich Drepper.
3435 * Implement second fallback mode for optimized DNS lookup for even more
3436 broken environments. If two requests from the same port are not
3437 handled correctly close the socket and open a new one before sending
3438 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3439 can be used to select this mode right away, instead of rediscovering the
3440 necessity is every process again.
3441 Implemented by Ulrich Drepper.
3443 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3444 Implemented by Adam Tkac.
3446 * Optimized iconv conversions for S390x.
3447 Implemented by Andreas Krebbel.
3449 * Using condvars with PI mutexes is now more efficient due to kernel
3450 support for requeueing to PI futexes. NPTL support added for x86-64.
3451 Implemented by Ulrich Drepper.
3453 * New locales: ps_AF, my_MM
3457 * The following bugs are resolved with this release:
3459 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3460 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3461 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3462 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3463 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3465 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3466 Implemented by Ulrich Drepper.
3468 * New Linux interfaces: accept4, fallocate, fallocate64.
3469 Implemented by Ulrich Drepper.
3471 * Correct declarations of string function when used in C++ code. This
3472 could lead to compile errors for invalid C++ code.
3474 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3477 * New POSIX 2008 interface: psiginfo
3478 Implemented by Ulrich Drepper.
3480 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3481 Implemented by Ulrich Drepper.
3483 * Support for selecting between multiple function definitions at runtime
3484 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3486 * The libcrypt library can now use the hash function implementations in
3487 NSS. Implemented by Ulrich Drepper.
3489 * The malloc implementation can be compiled to be less memory efficient
3490 but higher performing in multi-threaded programs.
3491 Implemented by Ulrich Drepper.
3493 * New locales: nan_TW@latin, ks_IN
3495 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3496 Implemented by Ulrich Drepper.
3498 * Extended printf hook support. It is possible to use user-defined types
3499 and extend existing format specifiers.
3500 Implemented by Ulrich Drepper.
3502 * Handling for group shadow files has been added.
3503 Implemented by Ulrich Drepper.
3505 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3506 servers (the case, e.g., for some people using the built-in DNS
3507 server in ADSL modems/routers). There is a once-per-process timeout
3508 in case of a broken server. To avoid it, users can run nscd or put
3509 'options single-request' in /etc/resolv.conf.
3510 Implemented by Ulrich Drepper.
3514 * The following bugs are resolved with this release:
3516 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3517 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3518 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3519 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3520 6968, 6974, 6980, 6995, 7008, 7009, 7029
3522 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3523 up at the same time. Implemented by Ulrich Drepper.
3525 * TLS descriptors for LD and GD on x86 and x86-64.
3526 Implemented by Alexandre Oliva.
3528 * getaddrinfo now handles DCCP and UDPlite.
3529 Implemented by Ulrich Drepper.
3531 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3532 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3533 Implemented by Ulrich Drepper.
3535 * New implementation of memmem, strstr, and strcasestr which is O(n).
3536 Implemented by Eric Blake.
3538 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3540 * Implement "e" option for popen to open file descriptor with the
3541 close-on-exec flag set. Implemented by Ulrich Drepper.
3543 * Implement "b" mode for fmemopen. In this mode writes writes don't
3544 implicitly add a NUL byte and seeks from the end of the buffer really
3545 use the buffer end, not the string length as the basis.
3546 Implemented by Ulrich Drepper.
3548 * Many functions, exported and internal, now atomically set the close-on-exec
3549 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3551 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3553 Implemented by Pravin Satpute.
3555 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3559 * The following bugs are resolved with this release:
3561 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3562 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3563 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3564 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3565 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3566 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3567 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3569 * New locales: bo_CN, bo_IN, shs_CA.
3571 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3573 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3574 Implemented by Pravin Satpute.
3576 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3578 * nscd caches now all timeouts for DNS entries
3579 Implemented by Ulrich Drepper.
3581 * nscd is more efficient and wakes up less often.
3582 Implemented by Ulrich Drepper.
3584 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3585 vdprintf, and obstack_vprintf.
3586 Implemented by Jakub Jelinek.
3588 * Faster memset for x86-64.
3589 Implemented by Harsha Jagasia and H.J. Lu.
3591 * Faster memcpy on x86.
3592 Implemented by Ulrich Drepper.
3594 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3595 Implemented by Ulrich Drepper.
3597 * Faster sqrt and sqrtf implemention for some PPC variants.
3598 Implemented by Steven Munroe.
3602 * The following bugs are resolved with this release:
3604 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3605 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3606 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3607 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3608 5071, 5103, 5104, 5112, 5113, 5184, 5186
3610 * More checking functions: fread, fread_unlocked, open*, mq_open.
3611 Implemented by Jakub Jelinek and Ulrich Drepper.
3613 * Extend fortification to C++. Implemented by Jakub Jelinek.
3615 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3616 by not recognizing 'a' as a modifier when those specs are requested.
3617 Implemented by Jakub Jelinek.
3619 * PPC optimizations to math and string functions.
3620 Implemented by Steven Munroe.
3622 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3623 options to be passed. Implemented by Ulrich Drepper.
3625 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3627 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3628 Implemented by Ulrich Drepper.
3630 * Handle private futexes in the NPTL implementation.
3631 Implemented by Jakub Jelinek and Ulrich Drepper.
3633 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3634 Implemented by Roland McGrath and Ulrich Drepper.
3636 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3638 * SHA-256 and SHA-512 based password encryption.
3639 Implemented by Ulrich Drepper.
3641 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3642 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3645 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3646 Implemented by Ulrich Drepper.
3650 * The following bugs are resolved with this release:
3652 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3653 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3654 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3655 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3656 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3657 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3658 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3661 * New Linux interfaces: epoll_pwait, sched_getcpu.
3663 * New generic interfaces: strerror_l.
3665 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3669 * The following bugs are resolved with this release:
3671 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3672 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3673 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3674 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3675 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3676 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3677 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3678 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3679 3225, 3251, 3252, 3253, 3273
3681 * For Linux, the sorting of addresses returned by getaddrinfo now also
3682 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3683 Implemented by Ulrich Drepper.
3685 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3686 Implemented by Ulrich Drepper.
3688 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3690 * New iconv module for MIK. Contributed by Alexander Shopov.
3692 * For sites with broken group and/or passwd database, the auto-propagate
3693 option of nscd can prevent creating ID lookup entries from the results
3694 of a name lookup and vice versa. This usually is no problem but some
3695 site might have problems with the default behavior.
3696 Implemented by Ulrich Drepper.
3698 * Iterating over entire database in NIS can be slow. With the
3699 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3700 to trade time for memory. The entire database will be read at once.
3701 Implemented by Ulrich Drepper.
3703 * The interfaces introduced in RFC 3542 have been implemented by
3706 * Support for the new ELF hash table format was added by Ulrich Drepper.
3708 * Support for priority inheritance mutexes added by Jakub Jelinek and
3711 * Support for priority protected mutexes added by Jakub Jelinek.
3713 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3714 or_IN, csb_PL, fy_NL, sr_ME.
3718 * More overflow detection functions.
3720 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3721 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3723 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3724 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3725 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3726 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3727 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3728 by Masahide Washizawa.
3730 * It is now possible to install audit libraries for the dynamic linker using
3731 LD_AUDIT. Implemented by Ulrich Drepper.
3733 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3734 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3735 For a libc and libpthread that works well on Linux 2.4 kernels, we
3736 recommend using the stable 2.3 branch.
3738 * The new function open_wmemstream acts like open_memstream,
3739 but uses a wchar_t wide character buffer.
3741 * The new function ppoll is an improved version of poll, similar to pselect.
3743 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3744 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3745 renameat, unlinkat, linkat, symlinkat, readlinkat.
3747 * New Linux kernel system calls: unshare,
3748 inotify_init, inotify_add_watch, inotify_rm_watch.
3750 * The euidaccess function is now also known by the alias eaccess,
3751 for compatibility with some other systems.
3753 * Timezone data updated to 2006b version.
3757 * The following bugs are resolved with this release:
3759 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3760 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3761 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3762 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3763 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3764 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3766 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3768 * As of this release, GCC 4 can be used to compile the C Library.
3770 * Timezone data updated to 2005m version.
3774 * The following bugs are resolved with this release:
3776 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3777 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3778 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3779 777, 787, 821, 822, 823, 825
3781 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3785 * Support for RFC 3678. Real implementations exist only for Linux so far.
3786 Implemented by Ulrich Drepper.
3788 * nscd can now cache entries persistently. Expiring entries are reloaded.
3789 For speedups the cache can be shared in memory with client processes.
3790 Implemented by Ulrich Drepper.
3792 * nscd can now perform SELinux checks.
3793 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3795 * getaddrinfo queries are now cached. Canonical name lookup is performed
3797 Implemented by Ulrich Drepper.
3799 * The nothrow function attribute is used when headers are used by gcc when
3800 compiling C code. This can avoid significant amounts of exception
3803 * The malloc functions perform more error checking and are stricter when
3804 it comes to reacting on errors. The default action is to terminate
3805 the process after showing an error message. Implemented by Ulrich Drepper.
3807 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3808 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3810 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3811 namespaces using the new function dlmopen(). This feature is of course,
3812 like most other dynamic loading functionality, not available in statically
3813 linked applications. Implemented by Ulrich Drepper.
3815 * Low-overhead boundary checking variants of string and some stdio functions
3816 were added. These are to be used in conjunction with a gcc patch by
3817 Jakub Jelinek which adds calls to these functions if possible.
3818 Implemented by Jakub Jelinek and Ulrich Drepper.
3820 * Old code for several operating systems and machine architectures that
3821 have not been in working condition in a long time have been removed from
3822 the main source tree maintained by the GNU C Library's maintainers.
3823 These files are now reside in the separate `ports' source module
3824 that is usable as an add-on when building the library.
3828 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3829 interrogate the dynamic linker, compatible with the Solaris interface.
3831 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3832 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3834 * getifaddrs now uses the netlink interface on Linux to get its information.
3835 Implemented by Thorsten Kukuk.
3837 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3838 Implemented by Ulrich Drepper.
3840 * support for non-executable stacks on x86 has been added. Changes mostly
3843 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3846 * getaddrinfo now performs destination address selection according to
3851 * Thread-safe interfaces for many functions that access locale data
3852 were added in version 2.3, but these features were omitted from NEWS.
3853 Many functions have variants with an `_l' suffix that take a `locale_t'
3854 object as a parameter rather than consulting the current locale.
3855 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3856 create and maintain `locale_t' objects. Additionally, the new function
3857 `uselocale' sets "the current locale" (as used by functions not so
3858 parameterized) set for an individual thread. These features were added
3859 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3861 * The functions getresuid, getresgid, setresuid, and setresgid, which
3862 have long been available on Linux, are now declared in <unistd.h>
3863 and are now also available on the Hurd.
3865 * ELF thread-local storage support (TLS) now works on x86-64.
3867 * The new dynamic string token $LIB is expanded in shared library names.
3868 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3870 * Aldy Hernandez contributed complete software floating point support for
3871 PowerPC machines with no FPU.
3873 * fexecve is implemented on Linux.
3875 * The `btowc' function should work at least twice as fast due to
3876 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3878 * With approriate thread add-ons cancelable functions are now implemented
3879 in libc.so as well. No need to call the function in libpthread. This
3880 change allowed to finally disable the incorrect and expensive handling
3881 of weak definition in ld.so.
3883 * Yet more PLT entries in libc.so have been removed. We finally arrived
3884 at the bare minimum. Startup times improved appropriately.
3886 * Support for the new Linux/x86 system call interface was added. The
3887 AT_SYSINFO auxiliary vector entry is recognized and handled.
3891 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3894 * iconv (the program and the interface) now accepts empty names (excluding
3895 options like //TRANSLIT) to mean "use charset of current locale".
3897 * localedef can now transliterate characters in strings which are not in
3898 the provided charmap. The information from the input locale is used.
3900 * Prelinking support was added for ELF targets. This requires additional
3901 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3903 * Read-only stdio streams now use mmap to speed up operation by eliminating
3904 copying and buffer underflows. To use add 'm' to the mode string of
3905 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3907 * The malloc functions were completely rewritten by Wolfram Gloger based
3908 on Doug Lea's malloc-2.7.0.c.
3910 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3911 implementation of regex.
3913 * Bruno Haible upgraded the iconv and locale implementation to support
3916 * Contents of the LC_* and LANG environment variables in the CEN style are
3917 not recognized anymore. It never was used. Change by Ulrich Drepper.
3919 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3920 thread-local storage (TLS) ABI on some platforms.
3921 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3923 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3924 EUC-JISX0213, and TSCII.
3926 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3927 BSD-compatible interface for getting all network interface addresses.
3928 Implementation for IPv4 by Roland McGrath.
3930 * Loading of locale data is faster due to the introduction of a locale
3931 archive. Implemented by Roland McGrath and Ulrich Drepper.
3933 * Startup times are significantly reduced by not using exported functions
3934 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3937 * Steven Munroe contributed a port to PowerPC64/Linux.
3941 * The Hurd now uses the GNU libio implementation of stdio rather than the
3942 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3944 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3945 with the same behavior as the Linux system call of the same name.
3949 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3950 128-bit long double format.
3952 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3953 IBM1161, and IBM1162 charsets.
3955 * Andreas Jaeger contributed a port to x86-64/Linux.
3957 * Peter Bruin contributed a port to PowerPC/Hurd.
3959 * libc functions using I/O streams now can handle wide-oriented streams
3962 * optimizations in the dynamic linker. Binaries created by recent binutils
3963 versions start up quicker due to reduced time spent on relocations.
3965 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3969 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3970 asinh, atanh, j0 for the 128-bit long double format.
3972 * Bruno Haible updated all the code handling Unicode in some form to
3973 support Unicode 3.1.
3975 * Speed of regex for single-byte locales is back to previous levels.
3976 Patch by Isamu Hasegawa.
3978 * Alpha, SPARC, and IA-64 now also using floating stacks.
3980 * Startup time of internationalized applications greatly improved through
3981 iconv cache. Use iconvconfig to generate the cache file.
3982 Contributed by Ulrich Drepper.
3984 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3985 severe performance problems. Patch by David Mosberger.
3987 * The Hurd port got a lot more functionality like AIO, various stdio
3988 extensions, etc. Mainly done by Roland McGrath.
3990 * mtrace can now lookup symbols in shared libraries.
3994 * Intel's IA-64 math library is largely integrated. It provides fast and
3995 accurate implementatations for most basic and standard math functions
3996 in float, double, and long double format.
3998 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3999 and asin for the 96-bit long double format and asin, log, tan for the
4000 128-bit long double format.
4002 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4003 The basic double functions exist today. Contributed by Abraham Ziv
4004 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4005 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4007 * An asynchronous name lookup library was added. The interface is designed
4008 after POSIX AIO. The proposal was circulated beforehand to get comments.
4009 No negative ones came in. Implemented by Ulrich Drepper.
4011 * Port to S390/64bit contributed by Martin Schwidefsky
4012 <schwidefsky@de.ibm.com>.
4014 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4015 of functions for Linux/IA-64.
4017 * The RPC code is now thread safe. Threads can now use the same service
4018 of different services at the same time. Patch by Eric Norum
4019 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4021 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4022 family of functions for Linux/S390.
4024 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4025 of functions for Linux/x86.
4027 * Port to Linux/CRIS contributed by Axis Communications.
4031 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4032 we can now check for namespace violations and missing declarations. The
4033 result is that almost all headers are now Unix-compliant (as defined in
4034 the upcoming XPG6). The negative side is that some programs might need
4035 corrections, too, if they depend on the incorrect form of the headers in
4036 previous versions which defined too many symbols and included too many
4039 * regex now handles multibyte character sets correctly.
4040 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4042 * iconv (the program) does now conform to the upcoming XPG6 and handles
4043 charmaps. Instead of the charset names the path of charmaps can be
4044 provided and the conversion happens based on this data.
4045 Contributed by Ulrich Drepper.
4047 * The locale program now provides detailed information about the installed
4052 only lists the names of the supported locales
4056 provides details such as country, language, and codeset name.
4057 Contributed by Ulrich Drepper.
4061 * The gencat program now parses the input file according to the charset
4062 selected by the LC_CTYPE category. This is important for stateful
4063 character sets. To make generating catalogs easier there is a way
4064 to overwrite the charset selected by the locale: before the first
4065 message or $ quote line the catalog can contain a line like
4067 $ codeset=ISO-8859-2
4069 to select the charset (ISO-8859-2 in this case).
4071 Implemented by Shinya Hanataka and Ulrich Drepper.
4073 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4074 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4075 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4077 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4078 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4080 * The LANGUAGE environment variable is now ignored unless the locale is
4081 changed from the default "C" locale.
4083 * The usual bug fixes.
4087 * Greg McGary added runtime support for bounds checking using gcc's
4088 new -fbounded-pointers option. ix86 target is complete. PowerPC
4091 * Thorsten Kukuk added secure mode support to nscd.
4093 * The Berkeley db libraries have been removed.
4095 Related, the nss_db module is now in a separate package since it
4096 obviously requires a database library being available.
4098 * Wide character I/O streams implemented by Ulrich Drepper.
4100 * Functions from the extended socket API added by Ulrich Drepper.
4102 * Functions feenableexcept and fedisableexcept to control the
4103 behaviour of individual exceptions have been added by Andreas Jaeger.
4105 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4107 * The resolver code has been updated from bind 8.2.3-T5B which supports
4108 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4111 This change could in some situations effect backward compatibility. Since
4112 now `_res' is a thread-local instead of a global variable, modifying it
4113 in one thread does not have any effect in other threads.
4115 The resolver library was also extended to allow IPv6 as the transport
4116 protocol for the requests. This work was done by Stig Venaas.
4118 * Compatibility code for K&R C compilers has been removed from the
4119 header files. A ISO C compiler is needed to use the library
4120 (conforming to either C89 or C99 standard).
4122 * Complete rewrite of the localedef program to support multibyte character
4123 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4124 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4125 Implemented by Ulrich Drepper.
4127 Bruno Haible significantly improved the generation and use of the data
4128 structures for the wide character tables.
4130 * Plural handling in gettext implemented by Ulrich Drepper.
4132 * The utmp daemon has been removed.
4134 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4136 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4139 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4141 * POSIX barriers implemented by Kaz Kylheku.
4143 * POSIX spawn function family implemented by Ulrich Drepper.
4145 * POSIX spinlocks are now available.
4147 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4149 * the configure option --enable-kernel=X.Y.Z allows to strip out
4150 compatibility for kernel versions before X.Y.Z. This is currently only
4151 implemented for Linux.
4153 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4154 field sin6_scope_id. This means that all programs using IPv6 should be
4155 recompiled. Don't expect binary compatibility with previous glibc
4158 * various conversion modules for IBM character sets contributed by
4161 * IA-64 port by Jes Sorensen and HJ Lu.
4175 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4177 * New xdr functions are added; some rpc functions are now 64bit clean.
4179 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4181 * Fixed known incompatibilities with glibc 2.0.
4183 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4185 * Optimized a number of functions (especially the ELF dynamic loader).
4187 * Update timezone data files.
4189 * lots of charmaps corrections
4191 * some new locale definitions and charmaps
4196 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4197 conform to POSIX member type requirements. Symbol versions have been
4198 adjusted as needed within the library, and for direct use by applications,
4199 but there is potential for problems if third-party libraries use
4200 struct timeval as part of their interface. This does not present
4201 a problem for X and other "essential" system libraries.
4203 * An additional locale model to support C++ Standard Library locale
4204 model and probably more was implemented by Ulrich Drepper.
4206 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4209 * Miles Bader provided the `argp' function family to support hierachical
4210 command line argument parsing, layered on top of getopt.
4212 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4214 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4217 * scanf recognizes the %a and %A format for scanning floating point numbers.
4219 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4220 information and interfaces for the available integer types.
4222 * about 130 new math functions were added to implement the ISO C9x math
4225 * the new header <complex.h> contains definitions of the complex math
4226 functions from ISO C 9X.
4228 * the new header <tgmath.h> defines generic macros to use complex or
4229 real valued functions.
4231 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4233 * Andreas Jaeger provided a test suite for the math library.
4235 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4237 * Ulrich Drepper added character set conversion functions (iconv).
4239 * Optimized string functions have been added.
4241 * The localedata addon is now part of glibc.
4243 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4245 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4246 daemon for NSS (nscd).
4248 Missing a better place here are some numbers on improvements. Under
4249 Linux 2.1.125 un-tar-ing the kernel sources takes
4253 using local files 12.19s 6.88s 22.91s
4255 using NIS 13.92s 8.91s 26.34s
4257 using NIS & nscd 10.37s 7.34s 25.30s
4259 using NIS+ 27.57s 30.37s 640.46s
4261 using NIS+ & nscd 10.25s 7.83s 26.51s
4263 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4265 Keep in mind that non-namelookup related operations dominate above times.
4266 It was just a common complain that using NIS+ unpacking the kernel is
4269 [1] The old nscd implementation is not available anymore since it was
4270 distributed with glibc up to version 2.0.98 and thus is now replaced.
4272 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4274 * Mark Kettenis provided a Hesiod NSS module.
4276 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4277 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4279 * rcmd can now handle netgroups (Dick Streefland).
4281 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4284 * Support for the IPv6 protocol has been added to the socket API, as per the
4285 latest draft standards.
4287 * Support for Linux 2.2 has been added.
4289 * Interface changes relative to the latest 2.0.x release:
4290 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4291 addseverity NEW: Unix98
4292 alphasort64 NEW: LFS
4293 argp_err_exit_status NEW: argp, GNU ext
4294 argp_error NEW: argp, GNU ext
4295 argp_failure NEW: argp, GNU ext
4296 argp_help NEW: argp, GNU ext
4297 argp_parse NEW: argp, GNU ext
4298 argp_program_bug_address NEW: argp, GNU ext
4299 argp_program_version NEW: argp, GNU ext
4300 argp_program_version_hook NEW: argp, GNU ext
4301 argp_state_help NEW: argp, GNU ext
4302 argp_usage NEW: argp, GNU ext
4303 authdes_create NEW: Secure RPC
4304 authdes_getucred NEW: Secure RPC
4305 authdes_pk_create NEW: Secure RPC
4306 backtrace NEW: GNU ext.
4307 backtrace_symbols NEW: GNU ext.
4308 backtrace_symbols_fd NEW: GNU ext.
4310 cacosf NEW: ISO C 9x
4311 cacosh NEW: ISO C 9x
4312 cacoshf NEW: ISO C 9x
4313 cacoshl NEW: ISO C 9x
4314 cacosl NEW: ISO C 9x
4321 casinf NEW: ISO C 9x
4322 casinh NEW: ISO C 9x
4323 casinhf NEW: ISO C 9x
4324 casinhl NEW: ISO C 9x
4325 casinl NEW: ISO C 9x
4327 catanf NEW: ISO C 9x
4328 catanh NEW: ISO C 9x
4329 catanhf NEW: ISO C 9x
4330 catanhl NEW: ISO C 9x
4331 catanl NEW: ISO C 9x
4332 cbc_crypt NEW: Secure RPC
4336 ccoshf NEW: ISO C 9x
4337 ccoshl NEW: ISO C 9x
4343 cimagf NEW: ISO C 9x
4344 cimagl NEW: ISO C 9x
4345 clearerr_locked REMOVED
4346 clntunix_create NEW: sunrpc ext
4348 clog10 NEW: ISO C 9x
4349 clog10f NEW: ISO C 9x
4350 clog10l NEW: ISO C 9x
4360 cprojf NEW: ISO C 9x
4361 cprojl NEW: ISO C 9x
4363 crealf NEW: ISO C 9x
4364 creall NEW: ISO C 9x
4369 csinhf NEW: ISO C 9x
4370 csinhl NEW: ISO C 9x
4373 csqrtf NEW: ISO C 9x
4374 csqrtl NEW: ISO C 9x
4378 ctanhf NEW: ISO C 9x
4379 ctanhl NEW: ISO C 9x
4381 des_setparity NEW: Secure RPC
4382 ecb_crypt NEW: Secure RPC
4383 endutxent NEW: Unix98
4385 exp10f NEW: ISO C 9x
4386 exp10l NEW: ISO C 9x
4390 fattach NEW: STREAMS
4391 fdetach NEW: STREAMS
4395 feclearexcept NEW: ISO C 9x
4396 fegetenv NEW: ISO C 9x
4397 fegetexceptflag NEW: ISO C 9x
4398 fegetround NEW: ISO C 9x
4399 feholdexcept NEW: ISO C 9x
4401 feraiseexcept NEW: ISO C 9x
4402 ferror_locked REMOVED
4403 fesetenv NEW: ISO C 9x
4404 fesetexceptflag NEW: ISO C 9x
4405 fesetround NEW: ISO C 9x
4406 fetestexcept NEW: ISO C 9x
4407 feupdateenv NEW: ISO C 9x
4408 fflush_locked REMOVED
4412 fgets_unlocked NEW: GNU ext.
4413 fileno_locked REMOVED
4425 fputc_locked REMOVED
4426 fputs_unlocked NEW: GNU ext.
4427 fread_unlocked NEW: GNU ext.
4432 fstatvfs NEW: Unix98
4436 ftruncate64 NEW: LFS
4438 fwrite_unlocked NEW: GNU ext.
4439 gai_strerror NEW: IPv6
4443 getchar_locked REMOVED
4445 getdate_err NEW: Unix98
4446 getdate_r NEW: GNU ext.
4448 getnameinfo NEW: IPv6
4449 getnetname NEW: Secure RPC
4450 getpmsg NEW: STREAMS
4451 getpt NEW: Unix98 PTY
4452 getrlimit64 NEW: LFS
4453 getutxent NEW: Unix98
4454 getutxid NEW: Unix98
4455 getutxline NEW: Unix98
4456 glob64 NEW: GNU ext.
4457 globfree64 NEW: GNU ext.
4458 gnu_get_libc_release NEW: GNU ext.
4459 gnu_get_libc_version NEW: GNU ext.
4460 grantpt NEW: Unix98 PTY
4461 host2netname NEW: Secure RPC
4463 iconv_close NEW: iconv
4464 iconv_open NEW: iconv
4465 if_freenameindex NEW: IPv6
4466 if_indextoname NEW: IPv6
4467 if_nameindex NEW: IPv6
4468 if_nametoindex NEW: IPv6
4469 in6addr_any NEW: IPv6
4470 in6addr_loopback NEW: IPv6
4471 inet6_isipv4mapped NEW: IPv6
4472 isastream NEW: STREAMS
4473 iswblank NEW: GNU ext.
4474 key_decryptsession NEW: Secure RPC
4475 key_decryptsession_pk NEW: Secure RPC
4476 key_encryptsession NEW: Secure RPC
4477 key_encryptsession_pk NEW: Secure RPC
4478 key_gendes NEW: Secure RPC
4479 key_get_conv NEW: Secure RPC
4480 key_secretkey_is_set NEW: Secure RPC
4481 key_setnet NEW: Secure RPC
4482 key_setsecret NEW: Secure RPC
4483 llrint NEW: ISO C 9x
4484 llrintf NEW: ISO C 9x
4485 llrintl NEW: ISO C 9x
4486 llround NEW: ISO C 9x
4487 llroundf NEW: ISO C 9x
4488 llroundl NEW: ISO C 9x
4493 lrintf NEW: ISO C 9x
4494 lrintl NEW: ISO C 9x
4495 lround NEW: ISO C 9x
4496 lroundf NEW: ISO C 9x
4497 lroundl NEW: ISO C 9x
4499 makecontext NEW: Unix98
4500 mempcpy NEW: GNU ext.
4503 modify_ldt NEW: kernel
4507 nearbyint NEW: ISO C 9x
4508 nearbyintf NEW: ISO C 9x
4509 nearbyintl NEW: ISO C 9x
4510 netname2host NEW: Secure RPC
4511 netname2user NEW: Secure RPC
4512 nexttoward NEW: ISO C 9x
4513 nexttowardf NEW: ISO C 9x
4514 nexttowardl NEW: ISO C 9x
4518 passwd2des NEW: Secure RPC
4520 pow10f NEW: GNU ext.
4521 pow10l NEW: GNU ext.
4524 printf_size NEW: GNU ext.
4525 printf_size_info NEW: GNU ext.
4526 profil_counter REMOVED
4527 pthread_mutexattr_getkind_np REPLACED
4528 pthread_mutexattr_setkind_np REPLACED
4529 ptsname NEW: Unix98 PTY
4530 ptsname_r NEW: Unix98 PTY
4532 putchar_locked REMOVED
4533 putgrent NEW: GNU ext.
4535 putpmsg NEW: STREAMS
4536 pututxline NEW: Unix98
4540 readdir64_r NEW: LFS
4541 remquo NEW: ISO C 9x
4542 remquof NEW: ISO C 9x
4543 remquol NEW: ISO C 9x
4545 roundf NEW: ISO C 9x
4546 roundl NEW: ISO C 9x
4548 scalbln NEW: ISO C 9x
4549 scalblnf NEW: ISO C 9x
4550 scalblnl NEW: ISO C 9x
4552 sendfile NEW: kernel
4553 setcontext NEW: Unix98
4554 setrlimit64 NEW: LFS
4555 setutxent NEW: Unix98
4557 sigignore NEW: Unix98
4558 sigqueue NEW: POSIX.1b
4559 sigrelse NEW: Unix98
4560 sigset NEW: POSIX.1b
4561 sigtimedwait NEW: POSIX.1b
4562 sigwaitinfo NEW: POSIX.1b
4563 sincos NEW: GNU ext.
4564 sincosf NEW: GNU ext.
4565 sincosl NEW: GNU ext.
4569 strcasestr NEW: GNU ext.
4570 strtoimax NEW: ISO C 9x
4571 strtoumax NEW: ISO C 9x
4572 strverscmp NEW: GNU ext.
4573 svcauthdes_stats NEW: Secure RPC
4574 svcunix_create NEW: sunrpc etx
4575 svcunixfd_create NEW: sunrpc ext
4576 swapcontext NEW: Unix98
4577 tcgetsid NEW: Unix98 PTY
4578 tdestroy NEW: GNU ext.
4579 tgamma NEW: ISO C 9x
4580 tgammaf NEW: ISO C 9x
4581 tgammal NEW: ISO C 9x
4585 truncf NEW: ISO C 9x
4586 truncl NEW: ISO C 9x
4588 unlockpt NEW: Unix98 PTY
4589 updwtmpx NEW: Unix98
4590 user2netname NEW: Secure RPC
4591 utmpxname NEW: Unix98
4592 versionsort NEW: GNU ext.
4593 versionsort64 NEW: GNU ext.
4595 wcscasecmp NEW: GNU ext.
4596 wcsncasecmp NEW: GNU ext.
4597 wcsnlen NEW: GNU ext.
4598 wcstoimax NEW: ISO C 9x
4599 wcstoll NEW: ISO C 9x
4600 wcstoull NEW: ISO C 9x
4601 wcstoumax NEW: ISO C 9x
4603 wordexp NEW: POSIX.2
4604 wordfree NEW: POSIX.2
4605 write_profiling REMOVED
4606 xdecrypt NEW: Secure RPC
4607 xdr_authdes_cred NEW: Secure RPC
4608 xdr_authdes_verf NEW: Secure RPC
4609 xdr_cryptkeyarg NEW: Secure RPC
4610 xdr_cryptkeyarg2 NEW: Secure RPC
4611 xdr_cryptkeyres NEW: Secure RPC
4612 xdr_getcredres NEW: Secure RPC
4613 xdr_key_netstarg NEW: Secure RPC
4614 xdr_key_netstres NEW: Secure RPC
4615 xdr_keybuf NEW: Secure RPC
4616 xdr_keystatus NEW: Secure RPC
4617 xdr_netnamestr NEW: Secure RPC
4618 xdr_sizeof NEW: Secure RPC
4619 xdr_unixcred NEW: sunrpc ext
4620 xencrypt NEW: Secure RPC
4621 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4632 * inet_ntoa is thread-safe
4634 * updwtmp is moved from libutil to libc
4636 * rewrite of cbrt function
4638 * update of timezone data
4652 * add atoll function
4654 * fix complex problems in Berkeley DB code
4656 * fix math functions
4660 * fixed lots of header problems (especially Linux/GNU specific)
4662 * dynamic loader preserves all registers
4664 * Roland McGrath provided support for handling of auxiliary objects in
4665 the ELF dynamic loader.
4667 * support for parallel builds is improved
4671 * GNU extensions are no longer declared by default. To enable them you
4672 must define the macro `_GNU_SOURCE' in your program or compile with
4675 * The library has changed from using GNU ld symbol aliases to using weak
4676 symbols where available. The ELF object file format supports weak
4677 symbols; GNU ld also supports weak symbols in the a.out format. (There
4678 is also now support for other GNU ld extensions in ELF. Use the
4679 `--with-elf' option to configure to indicate you have ELF, and
4680 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4681 of many files which contained only symbol aliases, reducing the size of
4682 the source and the compiled library; many other files were renamed to
4683 less cryptic names previously occupied by the symbol alias files.
4684 There is a new header file <elf.h> for programs which operate on
4685 files in the ELF format.
4687 * Converted to Autoconf version 2, so `configure' has more options.
4688 Run `configure --help' to see the details.
4690 * The library can now be configured to build profiling, highly-optimized
4691 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4692 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4693 `configure' enable building these extra libraries. The shared library is
4694 built by default when using both ELF and GNU ld. When shared libraries
4695 are enabled, the new library `-ldl' is available for arbitrary run-time
4696 loading of shared objects; its interface is defined in <dlfcn.h>. The
4697 new header file <link.h> gives access to the internals of the run-time
4698 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4699 application of same name on other systems and it provides information
4700 about dynamically linked binaries.
4702 * The C library now provides the run-time support code for profiling
4703 executables compiled with `-pg'. Programs can control the profiling code
4704 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4705 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4706 the support for this file format was contributed by David Mosberger-Tang.
4708 * The math code has been replaced with a math library based on fdlibm from
4709 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4710 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4711 functions. The math functions now reside in a separate library, so
4712 programs using them will need to use `-lm' their linking commands.
4714 * John C. Bowman contributed optimized ix87 assembler inline functions.
4716 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4717 mechanism similar to that found in Solaris 2. This is now used for the
4718 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4719 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4720 file controls what services are used for each individual database. This
4721 works by loading shared libraries with names specified in `nsswitch.conf',
4722 so service modules can be changed or added at any time without even
4723 relinking any program. Currently there are the file, db, and NIS based
4724 NSS services available.
4726 * The new functions `strtoq' and `strtouq' parse integer values from
4727 strings, like `strtol' and `strtoul', but they return `long long int' and
4728 `unsigned long long int' values, respectively (64-bit quantities).
4730 * The new functions `strtof' and `strtold' parse floating-point values from
4731 strings, like `strtod', but they return `float' and `long double' values,
4732 respectively (on some machines `double' and `long double' are the same).
4734 * Ulrich Drepper has contributed new implementations of the floating-point
4735 printing and reading code used in the `printf' family of functions and
4736 `strtod', `strtof', and `strtold'. These new functions are perfectly
4737 accurate, and much faster than the old ones.
4739 * The implementation of the POSIX locale model was completely rewritten by
4740 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4741 to compile the POSIX locale definition.
4743 * The former dummy implementations of the strcoll and strxfrm function are
4744 now replaced by fully functional code contributed by Ulrich Drepper. The
4745 collation information comes from the POSIX locale definitions.
4747 * The new header <langinfo.h> defines an interface for accessing
4748 various locale-dependent data (using the locale chosen with `setlocale').
4750 * Ulrich Drepper has contributed a new suite of functions for operation on
4751 wide-character and multibyte-character strings, in <wchar.h>;
4752 and classification and case conversion of wide characters, in <wctype.h>.
4753 These new functions are conforming to the ISO C, Amendement 1 specification.
4755 * There is now a second implementation of the standard I/O library available.
4756 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4757 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4759 * You can now use positional parameter specifications in format strings
4760 for the `printf' and `scanf' families of functions. For example,
4761 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4762 ``Number 6, Mr Jones''. This is mainly useful when providing different
4763 format strings for different languages, whose grammars may dictate
4764 different orderings of the values being printed. To support this
4765 feature, the interface for `register_printf_handler' has changed; see
4766 the header file <printf.h> for details.
4768 * The `printf' and `scanf' families of functions now understand a new
4769 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4770 to group numbers as indicated by the locale; for `scanf' and friends, this
4771 says to accept as valid only a number with all the proper grouping
4772 separators in the right places. In the default "C" locale, numbers are
4773 not grouped; but locales for specific countries will define the usual
4774 conventions (i.e. separate thousands with `,' in the US locale).
4776 * The pgrp functions have been regularized, slightly incompatibly but much
4777 less confusingly. The core functions are now `getpgid' and `setpgid',
4778 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4779 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4780 provided for compatibility. There is no longer an incompatible `getpgrp'
4781 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4783 * The new header file <fts.h> and suite of functions simplify programs that
4784 operate on directory trees. This code comes from 4.4 BSD.
4786 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4787 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4788 NSS scheme used in glibc.
4790 * There is a new malloc debugging hook `__memalign_hook'.
4792 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4793 `unsigned int' in <sys/types.h>. These are for compatibility only and
4794 their use is discouraged.
4796 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4797 done differently, so that it works even without GNU ld.
4799 * New function `euidaccess' checks allowed access to a file like `access',
4800 but using the effective IDs instead of the real IDs.
4802 * The time zone data files have been updated for the latest and greatest
4803 local time conventions of the countries of the world.
4805 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4808 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4809 for formatting floating-point numbers. They are provided only for
4810 compatibility; new programs should use `sprintf' instead. There are
4811 also equivalent function for the `long double' floating-point type and
4812 all functions also exist in a reentrant form.
4814 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4815 functions for maintaining the login-record files (primarily of use to
4816 system programs such as `login'), and convenient functions for
4817 allocating and initializing a pseudo-terminal (pty) device.
4819 * Ulrich Drepper has contributed new support for System V style
4820 shared memory and IPC on systems that support it.
4822 * Ulrich Drepper has contributed several miscellaneous new functions found
4823 in System V: The `hsearch' family of functions provide an effective
4824 implementation of hash tables; `a64l' and `l64a' provide a very simple
4825 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4828 * Ulrich Drepper has contributed new reentrant counterparts for the
4829 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4831 * Ulrich Drepper has contributed new, highly-optimized versions of several
4832 string functions for the i486/Pentium family of processors.
4834 * Ulrich Drepper has updated the Linux-specific code, based largely
4835 on work done in Hongjiu Lu's version of GNU libc for Linux.
4836 The GNU library now supports Linux versions 2.0.10 and later,
4837 using the ELF object file format (i[3456]86-*-linux).
4839 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4841 * David Mosberger-Tang and Richard Henderson have ported the C library
4842 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4843 dynamic linking support for ELF/Alpha.
4845 * Richard Henderson contributed several Alpha optimized assembler function
4846 for arithmetic and string handling.
4848 * Ulrich Drepper has contributed a new set of message catalog functions to
4849 support multiple languages using the <libintl.h> interface, for use with
4850 his new package GNU gettext. Translation volunteers have contributed
4851 catalogs of the library's messages in Spanish, German, and Korean.
4853 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4854 program and the `catgets' function for reading the catalog files it
4855 creates. (The <libintl.h> interface is preferred; we include the
4856 <nl_types.h> interface using `catgets' only for source compatibility with
4857 programs already written to use it.)
4859 * New header file <values.h> gives SVID-compatible names for <limits.h>
4862 * Various new macros, declarations, and small header files for compatibility
4865 * New function `group_member' is a convenient way to check if a process has
4866 a given effective group ID.
4868 * When using GCC 2.7 and later, the socket functions are now declared in a
4869 special way so that passing an argument of type `struct sockaddr_in *',
4870 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4871 `struct sockaddr *' type, does not generate a type-clash warning.
4873 * New function `error' declared in header file <error.h> is a convenient
4874 function for printing error messages and optionally exiting; this is the
4875 canonical function used in GNU programs. The new functions `err', `warn',
4876 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4877 doing the same thing.
4879 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4880 POSIX.2 `glob' function to do ~ and {...} expansion.
4882 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4883 `clearenv' which is used in POSIX.9 is also available.
4885 * New function `getsid' returns session ID number on systems that support it.
4887 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4888 files <db.h> and <mpool.h> provide a rich set of functions for several
4889 types of simple databases stored in memory and in files, and <ndbm.h> is
4890 an old `ndbm'-compatible interface using the `db' functions. Link with
4891 `-ldb' to get these functions.
4893 * New macro `strdupa' copies a string like `strdup', but uses local stack
4894 space from `alloca' instead of dynamic heap space from `malloc'.
4896 * New function `strnlen' is like `strlen' but searches only a given maximum
4897 number of characters for the null terminator. `stpncpy', `strndup' and
4898 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4901 * New function `statfs' in header <sys/statfs.h>.
4903 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4904 provide convenient functions for operating on blocks of null-terminated
4907 * A new suite of functions in <utmp.h> handle all the details of reading
4908 and writing the utmp file.
4910 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4913 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4914 completely POSIX compliant and also implemented the extended functionality
4915 to handle alternate digit representation and alternate era date formats.
4917 * Ulrich Drepper provided an implementation of the `strptime' function
4918 defined in XPG4.2 which transforms a string into a `struct tm' value.
4920 * Paul Eggert provided the tzselect shell script as part of the timezone
4921 code. The shell script makes it easy to select the correct timezone
4924 * The implementation of the malloc family of functions is completely replaced
4925 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4926 The implementation uses the mmap function (if available) and it is
4927 optimized for the use in multi threaded programs.
4929 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4930 functions. This new functionality is usable by specifying a special
4931 salt string and it is compatible with implementation on *BSD systems.
4933 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4934 `getsubopt' to handle second level command line options, `bsd_signal'
4935 to access BSD style `signal' functionality, the obsolete `regexp' style
4938 * the `lchown' function is available on system which support this
4941 * The implementation of the shadow password handling function was contributed
4944 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4946 * POSIX.1g support was added. The <sys/select.h> header is available,
4947 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4948 implementation of `getaddrinfo'.
4952 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4954 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4955 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4957 * The new function `daemon' from 4.4 BSD is useful for server programs that
4958 want to put themselves in the background.
4960 * Joel Sherrill has contributed support for several standalone boards that
4961 run without an operating system.
4963 * `printf', `scanf' and friends now accept a `q' type modifier for long
4964 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4966 * All of the code taken from BSD (notably most of the math and networking
4967 routines) has been updated from the BSD 4.4-Lite release.
4969 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4971 * The new functions `getdomainname' and `setdomainname' fetch or change the
4972 YP/NIS domain name. These are system calls which exist on systems which
4975 * The time zone data files have been updated for the latest international
4978 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4979 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4983 * The C library now includes support for Sun RPC, from Sun's free
4984 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4985 are included. (There is still no support for YP.)
4987 * Tom Quinn has contributed a port of the C library to SGI machines running
4988 Irix 4 (mips-sgi-irix4).
4990 * The new `lockf' function is a simplified interface to the locking
4991 facilities of `fcntl', included for compatibility.
4993 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4995 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4997 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5000 * The error message printed by `assert' for a failed assertion now includes
5001 the name of the program (if using GNU ld) and the name of the calling
5002 function (with versions of GCC that support this).
5004 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5006 * The library now includes the <sys/mman.h> header file and memory
5007 management functions `mmap', `munmap', `mprotect', `msync', and
5008 `madvise', on systems that support those facilities.
5010 * The interface for `mcheck' has changed slightly: the function called to
5011 abort the program when an allocation inconsistency is detected now takes
5012 an argument that indicates the type of failure. The new function
5013 `mprobe' lets you request a consistency check for a particular block at
5014 any time (checks are normally done only when you call `free' or `realloc'
5017 * It is now possible to easily cross-compile the C library, building on one
5018 system a library to run on another machine and/or operating system. All
5019 you need to do is set the variable `HOST_CC' in `configparms' to the
5020 native compiler for programs to run on the machine you are building on (a
5021 few generator programs are used on Unix systems); set `CC' to the
5024 * The new function `fexecve' (only implemented on the GNU system) executes
5025 a program file given a file descriptor already open on the file.
5029 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5030 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5032 * You can set the variable `libprefix' in `configparms' to specify a prefix
5033 to be prepended to installed library files; this makes it easy to install
5034 the GNU C library to be linked as `-lgnuc' or whatever.
5036 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5037 copies a limited number of characters from a string, and returns the
5038 address of the last character written.
5040 * You no longer need to check for whether the installed `stddef.h' is
5041 compatible with the GNU C library. configure now checks for you.
5043 * You can now define a per-stream `fileno' function to convert the
5044 stream's cookie into an integral file descriptor.
5046 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5047 allocates zero bytes of storage, and returns a unique pointer which
5048 you can pass to `realloc' or `free'. The behavior is undefined if
5049 you dereference this pointer.
5051 * The C library now runs on Sony NEWS m68k machines running either
5052 NewsOS 3 or NewsOS 4.
5054 * The new `syscall' function is a system-dependent primitive function
5055 for invoking system calls. It has the canonical behavior on Unix
5056 systems, including unreliable return values for some calls (such as
5057 `pipe', `fork' and `getppid').
5059 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5060 to `EAGAIN', which is the preferred name. On systems whose kernels
5061 use two distinct codes, the C library now translates EWOULDBLOCK to
5062 EAGAIN in every system call function.
5066 * The GNU C Library Reference Manual is now distributed with the library.
5067 `make dvi' will produce a DVI file of the printed manual.
5068 `make info' will produce Info files that you can read on line using C-h i
5069 in Emacs or the `info' program.
5070 Please send comments on the manual to bug-glibc-manual@gnu.org.
5072 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5074 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5076 * Jason Merrill has contributed a port to the Sequent Symmetry running
5077 Dynix version 3 (i386-sequent-dynix).
5079 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5080 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5082 * New function `memory_warnings' lets you arrange to get warnings when
5083 malloc is running out of memory to allocate, like Emacs gives you.
5085 * The C library now contains the relocating allocator used in Emacs 19 for
5086 its editing buffers. This allocator (ralloc) minimizes allocation
5087 overhead and fragmentation by moving allocated regions around whenever it
5088 needs to. You always refer to a ralloc'd region with a "handle" (a
5089 pointer to a pointer--an object of type `void **').
5091 * There is a new `printf' format: `%m' gives you the string corresponding
5092 to the error code in `errno'.
5094 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5095 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5096 pass, the `a' modifier says to fill in a `char **' you pass with a
5099 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5100 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5101 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5103 * `mkstemp' is a traditional Unix function to atomically create and open a
5104 uniquely-named temporary file.
5108 * The standard location for the file that says what the local timezone is
5109 has changed again. It is now `/usr/local/etc/localtime' (or more
5110 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5112 * The distribution no longer contains any files with names longer than 14
5115 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5116 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5118 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5122 * The configuration process has changed quite a bit. The `configure'
5123 script is now used just like the configuration scripts for other GNU
5124 packages. The `sysdeps' directory hierarchy is much rearranged.
5125 The file `INSTALL' explains the new scheme in detail.
5127 * The header files no longer need to be processed into ANSI C and
5128 traditional C versions. There is just one set of files to install, and
5129 it will work with ANSI or old C compilers (including `gcc -traditional').
5131 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5132 MIPS DECStation running Ultrix 4.
5134 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5135 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5136 made itself into a shared library.
5138 * Yet further improved support for the i386, running 4.3 BSD-like systems
5139 (such as Mach 3 with the Unix single-server), or System V.
5141 * New function `strncasecmp' to do case-insensitive string comparison
5142 with limited length.
5144 * New function `strsep' is a reentrant alternative to `strtok'.
5146 * New functions `scandir' and `alphasort' for searching directories.
5148 * New function `setenv' is a better interface to `putenv'.
5150 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5151 function for traversing a directory tree.
5153 * The GNU obstack package is now also part of the C library.
5154 The new function `open_obstack_stream' creates a stdio stream that
5155 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5156 formatted output directly to an obstack.
5158 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5159 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5161 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5163 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5164 things to your strings.
5166 * There are some new test programs: test-fseek, testmb, and testrand.
5168 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5169 These ports are not finished, but are a good starting place for really
5170 supporting those systems.
5172 * `/etc/localtime' is now the standard location for the file that says what
5173 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5174 This follows the general principle that `/etc' is the place for all local
5175 configuration files.
5177 * The C library header files now use `extern "C"' when used by the C++
5178 compiler, so the C library should now work with C++ code.
5180 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5181 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5184 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5185 function which uses the `merge sort' algorithm, and is said to be
5186 significantly faster than the old GNU `qsort' function. Merge sort is
5187 now the standard `qsort' function. The new algorithm can require a lot
5188 of temporary storage; so, the old sorting function is called when the
5189 required storage is not available.
5191 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5192 provides the Unix `crypt' function, plus some other entry points.
5194 * The code and header files taken from 4.4 BSD have been updated with the
5195 latest files released from Berkeley.
5197 ----------------------------------------------------------------------
5198 Copyright information:
5200 Copyright (C) 1992-2018 Free Software Foundation, Inc.
5202 Permission is granted to anyone to make or distribute verbatim copies
5203 of this document as received, in any medium, provided that the
5204 copyright notice and this permission notice are preserved,
5205 thus giving the recipient permission to redistribute in turn.
5207 Permission is granted to distribute modified versions
5208 of this document, or of portions of it,
5209 under the above conditions, provided also that they
5210 carry prominent notices stating who last changed them.
5213 version-control: never