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 * The localization data for ISO 14651 is updated to match the 2016
13 Edition 4 release of the standard, this matches data provided by
14 Unicode 9.0.0. This update introduces significant improvements to the
15 collation of Unicode characters. This release deviates slightly from
16 the standard in that the collation element ordering for lowercase and
17 uppercase LATIN script characters is adjusted to ensure that regular
18 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
19 is not matched by [a-z]. With the update many locales have been
20 updated to take advantage of the new collation information. The new
21 collation information has increased the size of the compiled locale
22 archive or binary locales.
24 * The GNU C Library can now be compiled with support for Intel CET, AKA
25 Intel Control-flow Enforcement Technology. When the library is built
26 with --enable-cet, the resulting glibc is protected with indirect
27 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
28 compatible with all existing executables and shared libraries. This
29 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
30 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
31 capable of multi-byte NOPs, like x86-64 processors as well as Intel
32 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
33 x86_64 and x32 on non-CET processors. --enable-cet has been tested
34 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
37 * The GNU C Library now has correct support for ABSOLUTE symbols
38 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
39 relocated incorrectly or in some cases discarded. The GNU linker can
40 make use of the newer semantics, but it must communicate it to the
41 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
42 field) to indicate such support is required.
44 * Unicode 11.0.0 Support: Character encoding, character type info, and
45 transliteration tables are all updated to Unicode 11.0.0, using
46 generator scripts contributed by Mike FABIAN (Red Hat).
48 * <math.h> functions that round their results to a narrower type are added
49 from TS 18661-1:2014 and TS 18661-3:2015:
51 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
54 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
57 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
60 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
63 * Two grammatical forms of month names are now supported for the following
64 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
65 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
66 support two grammatical forms in abbreviated month names: Catalan, Greek,
69 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
70 include the support for two grammatical forms of month names.
72 * Building and running on GNU/Hurd systems now works without out-of-tree
75 * The renameat2 function has been added, a variant of the renameat function
76 which has a flags argument. If the flags are zero, the renameat2 function
77 acts like renameat. If the flag is not zero and there is no kernel
78 support for renameat2, the function will fail with an errno value of
79 EINVAL. This is different from the existing gnulib function renameatu,
80 which performs a plain rename operation in case of a RENAME_NOREPLACE
81 flags and a non-existing destination (and therefore has a race condition
82 that can clobber the destination inadvertently).
84 * The statx function has been added, a variant of the fstatat64
85 function with an additional flags argument. If there is no direct
86 kernel support for statx, glibc provides basic stat support based on
87 the fstatat64 function.
89 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
90 library if installed. libidn2 version 2.0.5 or later is recommended. If
91 libidn2 is not available, internationalized domain names are not encoded
92 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
93 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
94 with an encoding error.) Flags which used to change the IDN encoding and
95 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
96 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
97 deprecated. They no longer have any effect.
99 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
100 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
101 range of ELF gABI expressions including such constructs as
102 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
103 have been further restricted, and where in the past a dynamic string
104 token sequence may have been interpreted as a literal string it will
105 now cause a load failure. These load failures were always considered
106 unspecified behaviour from the perspective of the dynamic loader, and
107 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
108 results in a load failure now.
110 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
111 implementation includes all the standard functions provided by
114 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
115 thrd_detach, thrd_exit, and thrd_join for thread management.
117 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
118 mtx_destroy for mutual exclusion.
120 - call_once for function call synchronization.
122 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
123 cnd_wait for conditional variables.
125 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
127 Application developers must link against libpthread to use ISO C threads.
129 Deprecated and removed features, and other changes affecting compatibility:
131 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
132 installed. Software that was using either header should be updated to
133 use standard <stdio.h> interfaces instead.
135 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
136 This was never required by the C standard, and the macros just expanded
137 to call alternative names for the same functions. If you hoped getc and
138 putc would provide performance improvements over fgetc and fputc, instead
139 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
140 necessary, flockfile and funlockfile.
142 * All stdio functions now treat end-of-file as a sticky condition. If you
143 read from a file until EOF, and then the file is enlarged by another
144 process, you must call clearerr or another function with the same effect
145 (e.g. fseek, rewind) before you can read the additional data. This
146 corrects a longstanding C99 conformance bug. It is most likely to affect
147 programs that use stdio to read interactive input from a terminal.
150 * The macros 'major', 'minor', and 'makedev' are now only available from
151 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
152 headers that happen to include <sys/types.h>. These macros are rarely
153 used, not part of POSIX nor XSI, and their names frequently collide with
154 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
157 <sys/sysmacros.h> is a GNU extension. Portable programs that require
158 these macros should first include <sys/types.h>, and then include
159 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
161 * The tilegx*-*-linux-gnu configurations are no longer supported.
163 * The obsolete function ustat is no longer available to newly linked
164 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
165 function has been deprecated in favor of fstatfs and statfs.
167 * The obsolete function nfsservctl is no longer available to newly linked
168 binaries. This function was specific to systems using the Linux kernel
169 and could not usefully be used with the GNU C Library on systems with
170 version 3.1 or later of the Linux kernel.
172 * The obsolete function name llseek is no longer available to newly linked
173 binaries. This function was specific to systems using the Linux kernel
174 and was not declared in a header. Programs should use the lseek64 name
175 for this function instead.
177 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
178 getaddrinfo and getnameinfo functions have been deprecated. The behavior
179 previously selected by them is now always enabled.
181 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
182 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
183 restriction (rejecting '_' in host names, among other things) has been
184 removed, for increased compatibility with non-IDN name resolution.
186 * The fcntl function now have a Long File Support variant named fcntl64. It
187 is added to fix some Linux Open File Description (OFD) locks usage on non
188 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
189 fcntl and LFS support is handled transparently. Also for Linux, the OFD
190 locks act as a cancellation entrypoint.
192 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
193 ecb_crypt, and des_setparity are no longer available to newly linked
194 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
195 longer installed. These functions encrypted and decrypted data with the
196 DES block cipher, which is no longer considered secure. Software that
197 still uses these functions should switch to a modern cryptography library,
200 * Reflecting the removal of the encrypt and setkey functions above, the
201 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
202 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
205 * The obsolete function fcrypt is no longer available to newly linked
206 binaries. It was just another name for the standard function crypt,
207 and it has not appeared in any header file in many years.
209 * We have tentative plans to hand off maintenance of the passphrase-hashing
210 library, libcrypt, to a separate development project that will, we hope,
211 keep up better with new passphrase-hashing algorithms. We will continue
212 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
213 'crypt_r' should not need to change at all; however, distributions will
214 need to install <crypt.h> and libcrypt from a separate project.
216 In this release, if the configure option --disable-crypt is used, glibc
217 will not install <crypt.h> or libcrypt, making room for the separate
218 project's versions of these files. The plan is to make this the default
219 behavior in a future release.
221 Changes to build and runtime requirements:
223 GNU make 4.0 or later is now required to build glibc.
225 Security related changes:
227 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
228 been fixed by removing the glibc-internal IDNA implementation and using
229 the system-provided libidn2 library instead. Originally reported by Hanno
230 Böck and Christian Weisgerber.
232 CVE-2017-18269: An SSE2-based memmove implementation for the i386
233 architecture could corrupt memory. Reported by Max Horn.
235 CVE-2018-11236: Very long pathname arguments to realpath function could
236 result in an integer overflow and buffer overflow. Reported by Alexey
239 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
240 architecture could write beyond the target buffer, resulting in a buffer
241 overflow. Reported by Andreas Schwab.
243 The following bugs are resolved with this release:
245 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
246 [6889] manual: 'PWD' mentioned but not specified
247 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
248 when __WORDSIZE != 64
249 [13762] regex: re_search etc. should return -2 on memory exhaustion
250 [13888] build: /tmp usage during testing
251 [13932] math: dbl-64 pow unexpectedly slow for some inputs
252 [14092] nptl: Support C11 threads
253 [14095] localedata: Review / update collation data from Unicode / ISO
255 [14508] libc: -Wformat warnings
256 [14553] libc: Namespace pollution loff_t in sys/types.h
257 [14890] libc: Make NT_PRFPREG canonical.
258 [15105] libc: Extra PLT references with -Os
259 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
261 [16335] manual: Feature test macro documentation incomplete and out of
263 [16552] libc: Unify umount implementations in terms of umount2
264 [17082] libc: htons et al.: statement-expressions prevent use on global
265 scope with -O1 and higher
266 [17343] libc: Signed integer overflow in /stdlib/random_r.c
267 [17438] localedata: pt_BR: wrong d_fmt delimiter
268 [17662] libc: please implement binding for the new renameat2 syscall
269 [17721] libc: __restrict defined as /* Ignore */ even in c11
270 [17979] libc: inconsistency between uchar.h and stdint.h
271 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
272 [18023] libc: extend_alloca is broken (questionable pointer comparison,
273 horrible machine code)
274 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
276 [18471] libc: llseek should be a compat symbol
277 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
279 [18991] nss: nss_files skips large entry in database
280 [19239] libc: Including stdlib.h ends up with macros major and minor being
282 [19463] libc: linknamespace failures when compiled with -Os
283 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
284 [19527] locale: Normalized charset name not recognized by setlocale
285 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
286 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
288 [19728] network: out of bounds stack read in libidn function
289 idna_to_ascii_4i (CVE-2016-6261)
290 [19729] network: out of bounds heap read on invalid utf-8 inputs in
291 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
292 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
294 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
295 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
296 [20419] dynamic-link: files with large allocated notes crash in
298 [20530] libc: bswap_16 should use __builtin_bswap16() when available
299 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
300 [20980] manual: CFLAGS environment variable replaces vital options
301 [21163] regex: Assertion failure in pop_fail_stack when executing a
302 malformed regexp (CVE-2015-8985)
303 [21234] manual: use of CFLAGS makes glibc detect no optimization
304 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
305 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
306 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
307 [21508] locale: intl/tst-gettext failure with latest msgfmt
308 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
309 [21812] network: getifaddrs() returns entries with ifa_name == NULL
310 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
312 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
314 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
315 [22247] network: Integer overflow in the decode_digit function in
316 puny_decode.c in libidn (CVE-2017-14062)
317 [22342] nscd: NSCD not properly caching netgroup
318 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
319 [22550] localedata: es_ES locale (and other es_* locales): collation
320 should treat ñ as a primary different character, sync the collation
321 for Spanish with CLDR
322 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
323 by gcc configured with --enable-default-pie
324 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
325 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
326 crossing 2GB threshold (CVE-2017-18269)
327 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
329 [22735] time: Misleading typo in time.h source comment regarding
331 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
332 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
333 `__assert_perror_fail ()' to `__assert_fail_base ()'
334 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
336 [22786] libc: Stack buffer overflow in realpath() if input size is close
337 to SSIZE_MAX (CVE-2018-11236)
338 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
339 through an absolute DT_NEEDED path
340 [22792] build: tcb-offsets.h dependency dropped
341 [22797] libc: pkey_get() uses non-reserved name of argument
342 [22807] libc: PTRACE_* constants missing for powerpc
343 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
344 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
345 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
346 [22848] localedata: ca_ES: update date definitions from CLDR
347 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
348 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
349 [22896] localedata: Update locale data for an_ES
350 [22902] math: float128 test failures with GCC 8
351 [22918] libc: multiple common of `__nss_shadow_database'
352 [22919] libc: sparc32: backtrace yields infinite backtrace with
354 [22926] libc: FTBFS on powerpcspe
355 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
357 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
358 [22947] libc: FAIL: misc/tst-preadvwritev2
359 [22963] localedata: cs_CZ: Add alternative month names
360 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
361 [22996] localedata: change LC_PAPER to en_US in es_BO locale
362 [22998] dynamic-link: execstack tests are disabled when SELinux is
364 [23005] network: Crash in __res_context_send after memory allocation
366 [23007] math: strtod cannot handle -nan
367 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
368 [23036] regex: regex equivalence class regression
369 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
370 [23069] libc: sigaction broken on riscv64-linux-gnu
371 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
372 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
373 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
374 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
375 and libc build with -Os)
376 [23140] localedata: More languages need two forms of month names
377 [23145] libc: _init/_fini aren't marked as hidden
378 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
379 [23171] math: C++ iseqsig for long double converts arguments to double
380 [23178] nscd: sudo will fail when it is run in concurrent with commands
381 that changes /etc/passwd
382 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
384 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
385 [23208] localedata: New locale - Lower Sorbian (dsb)
386 [23233] regex: Memory leak in build_charclass_op function in file
388 [23236] stdio: Harden function pointers in _IO_str_fields
389 [23250] nptl: Offset of __private_ss differs from GCC
390 [23253] math: tgamma test suite failures on i686 with -march=x86-64
391 -mtune=generic -mfpmath=sse
392 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
394 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
395 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
397 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
398 [23277] math: nan function should not have const attribute
399 [23279] math: scanf and strtod wrong for some hex floating-point
400 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
402 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
403 [23303] build: undefined reference to symbol
404 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
405 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
407 [23313] stdio: libio vtables validation and standard file object
409 [23329] libc: The __libc_freeres infrastructure is not properly run across
411 [23349] libc: Various glibc headers no longer compatible with
413 [23351] malloc: Remove unused code related to heap dumps and malloc
415 [23363] stdio: stdio-common/tst-printf.c has non-free license
416 [23396] regex: Regex equivalence regression in single-byte locales
417 [23422] localedata: oc_FR: More updates of locale data
418 [23442] build: New warning with GCC 8
419 [23448] libc: Out of bounds access in IBM-1390 converter
420 [23456] libc: Wrong index_cpu_LZCNT
421 [23458] build: tst-get-cpu-features-static isn't added to tests
422 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
424 [23467] dynamic-link: x86/CET: A property note parser bug
431 * The GNU C Library can now be compiled with support for building static
432 PIE executables (See --enable-static-pie in INSTALL). These static PIE
433 executables are like static executables but can be loaded at any address
434 and provide additional security hardening benefits at the cost of some
435 memory and performance. When the library is built with --enable-static-pie
436 the resulting libc.a is usable with GCC 8 and above to create static PIE
437 executables using the GCC option '-static-pie'. This feature is currently
438 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
439 aarch64 with binutils 2.30 or later.
441 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
442 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
445 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
447 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
449 * In order to support faster and safer process termination the malloc API
450 family of functions will no longer print a failure address and stack
451 backtrace after detecting heap corruption. The goal is to minimize the
452 amount of work done after corruption is detected and to avoid potential
453 security issues in continued process execution. Reducing shutdown time
454 leads to lower overall process restart latency, so there is benefit both
455 from a security and performance perspective.
457 * The abort function terminates the process immediately, without flushing
458 stdio streams. Previous glibc versions used to flush streams, resulting
459 in deadlocks and further data corruption. This change also affects
460 process aborts as the result of assertion failures.
462 * On platforms where long double has the IEEE binary128 format (aarch64,
463 alpha, mips64, riscv, s390 and sparc), the math library now implements
464 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
465 These are the same interfaces added in version 2.26 for some platforms where
466 this format is supported but is not the format of long double.
468 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
469 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
470 implements interfaces for that type, as defined by ISO/IEC TS
471 18661-3:2015. These are corresponding interfaces to those supported for
474 * The math library now implements interfaces for the _Float32, _Float64 and
475 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
476 corresponding interfaces to those supported for _Float128.
478 * glibc now implements the memfd_create and mlock2 functions on Linux.
480 * Support for memory protection keys was added. The <sys/mman.h> header now
481 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
484 * The copy_file_range function was added.
486 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
488 * The ldconfig utility now processes `include' directives using the C/POSIX
489 collation ordering. Previous glibc versions used locale-specific
490 ordering, the change might break systems that relied on that.
492 * Support for two grammatical forms of month names has been added.
493 In a call to strftime, the "%B" and "%b" format specifiers will now
494 produce the grammatical form required when the month is used as part
495 of a complete date. New "%OB" and "%Ob" specifiers produce the form
496 required when the month is named by itself. For instance, in Greek
497 and in many Slavic and Baltic languages, "%B" will produce the month
498 in genitive case, and "%OB" will produce the month in nominative case.
500 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
501 are all valid and will all accept any known form of month
502 name---standalone or complete, abbreviated or full. In a call to
503 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
504 the strings used by "%B" and "%b", respectively. New query
505 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
506 used by "%OB" and "%Ob", respectively.
508 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
509 define the strings for %OB and %Ob, respectively; these have the
510 same syntax as "mon" and "abmon". These arrays are optional; if they
511 are not provided then they have the same content as "mon" and "abmon",
514 These features are provided for locales which define "alt_mon" and/or
515 "ab_alt_mon" in their locale source data. This release includes such
516 alternative month name data for the following languages: Belarusian,
517 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
519 This feature is currently a GNU extension, but it is expected to
520 be added to the next revision of POSIX, and it is also already
521 available on some BSD-derived operating systems.
523 This feature will cause existing statically compiled applications
524 to fail to load locales and fall back to the builtin C/POSIX locales.
525 See notes below for other changes affecting compatibility.
527 * Support for the RISC-V ISA running on Linux has been added. This port
528 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
529 for the following ISA and ABI pairs:
535 Deprecated and removed features, and other changes affecting compatibility:
537 * Statically compiled applications attempting to load locales compiled for the
538 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
539 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
540 support for two grammatical forms of the month names, also extends the locale
541 data binary format. Static applications needing locale support must be
542 recompiled to match the runtime and data they are deployed with. In some
543 distributions there is an upgrade window where dynamically linked applications
544 may use a new library but the old locale data and also fall back to the
545 builtin C/POSIX locales; restarting the application process is sufficient to
548 * Support for statically linked applications which call dlopen is deprecated
549 and will be removed in a future version of glibc. Applications which call
550 dlopen need to be linked dynamically instead.
552 * Support for old programs which use internal stdio data structures and
553 functions is deprecated. This includes programs which use the C++ streams
554 provided by libstdc++ in GCC 2.95. Programs which use the internal
555 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
556 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
557 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
558 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
559 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
560 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
561 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
562 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
563 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
564 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
565 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
566 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
567 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
568 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
569 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
570 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
571 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
572 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
573 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
574 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
575 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
576 future version of glibc. Unlike other symbol removals, these old
577 applications will not be supported using compatibility symbols.
579 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
580 defined by <sys/ptrace.h>.
582 * libm no longer supports SVID error handling (calling a user-provided
583 matherr function on error) or the _LIB_VERSION variable to control error
584 handling. (SVID error handling and the _LIB_VERSION variable still work
585 for binaries linked against older versions of the GNU C Library.) The
586 libieee.a library is no longer provided. math.h no longer defines struct
587 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
588 TLOSS, PLOSS and HUGE.
590 * The libm functions pow10, pow10f and pow10l are no longer supported for
591 new programs. Programs should use the standard names exp10, exp10f and
592 exp10l for these functions instead.
594 * The mcontext_t type is no longer the same as struct sigcontext. On
595 platforms where it was previously the same, this changes the C++ name
596 mangling for interfaces involving this type.
598 * The add-ons mechanism for building additional packages at the same time as
599 glibc has been removed. The --enable-add-ons configure option is now
602 * The --without-fp configure option is now ignored. Whether hardware
603 floating-point instructions are used is now configured based on whether
604 the compiler used at configure time (without any options implied by a
605 --with-cpu= configure option) uses such instructions.
607 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
608 the specified string can be parsed as a domain name.
610 * In the malloc_info output, the <heap> element may contain another <aspace>
611 element, "subheaps", which contains the number of sub-heaps.
613 * The libresolv function p_secstodate is no longer supported for new
616 * The tilepro-*-linux-gnu configuration is no longer supported.
618 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
619 and will be removed in a future release. Software that is still using
620 either header should be updated to use standard <stdio.h> interfaces
623 libio.h was originally the header for a set of supported GNU extensions,
624 but they have not been maintained as such in many years, they are now
625 standing in the way of improvements to stdio, and we don't think there are
626 any remaining external users. _G_config.h was never intended for public
627 use, but predates the bits convention.
629 Changes to build and runtime requirements:
631 * bison version 2.7 or later is required to generate code in the 'intl'
634 Security related changes:
636 CVE-2009-5064: The ldd script would sometimes run the program under
637 examination directly, without preventing code execution through the
638 dynamic linker. (The glibc project disputes that this is a security
639 vulnerability; only trusted binaries must be examined using the ldd
642 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
643 suffered from a one-byte overflow during ~ operator processing (either
644 on the stack or the heap, depending on the length of the user name).
645 Reported by Tim Rühsen.
647 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
648 would sometimes fail to free memory allocated during ~ operator
649 processing, leading to a memory leak and, potentially, to a denial
652 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
653 without GLOB_NOESCAPE, could write past the end of a buffer while
654 unescaping user names. Reported by Tim Rühsen.
656 CVE-2017-17426: The malloc function, when called with an object size near
657 the value SIZE_MAX, would return a pointer to a buffer which is too small,
658 instead of NULL. This was a regression introduced with the new malloc
659 thread cache in glibc 2.26. Reported by Iain Buclaw.
661 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
662 to the allocation of too much memory. (This is not a security bug per se,
663 it is mentioned here only because of the CVE assignment.) Reported by
666 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
667 of the number of search path components. (This is not a security
668 vulnerability per se because no trust boundary is crossed if the fix for
669 CVE-2017-1000366 has been applied, but it is mentioned here only because
670 of the CVE assignment.) Reported by Qualys.
672 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
673 for AT_SECURE or SUID binaries could be used to load libraries from the
676 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
677 succeeds without returning an absolute path due to unexpected behaviour
678 of the Linux kernel getcwd syscall. Reported by halfdog.
680 CVE-2018-6485: The posix_memalign and memalign functions, when called with
681 an object size near the value of SIZE_MAX, would return a pointer to a
682 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
684 CVE-2018-6551: The malloc function, when called with an object size near
685 the value of SIZE_MAX, would return a pointer to a buffer which is too
686 small, instead of NULL.
688 The following bugs are resolved with this release:
690 [866] glob: glob should match dangling symlinks
691 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
692 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
694 [5997] math: Very slow execution of sinf function
695 [10580] localedata: hr_HR: updated locale
696 [10871] locale: 'mon' array should contain both nominative and genitive
698 [12349] localedata: eu_ES: incorrect thousands separator
699 [13605] localedata: shn_MM: new Shan locale
700 [13805] localedata: ru_RU: currency should use ',' as radix point
701 [13953] localedata: km_KH: locale update
702 [13994] localedata: mjw_IN: new locale
703 [14121] build: make writes .mo files in po directory
704 [14333] libc: Fix the race between atexit() and exit()
705 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
706 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
708 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
709 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
710 width Latin characters
711 [15332] localedata: es_CU: locale update
712 [15436] stdio: Don't close or flush stdio streams on abort
713 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
715 [16148] localedata: ca_ES: incorrect thousands separator
716 [16750] dynamic-link: ldd should not try to execute the binaries
718 [16777] localedata: pl_PL: incorrect thousands separator in locale
719 [16905] localedata: hanzi: new collation
720 [17563] localedata: cmn_TW: add hanzi collation
721 [17750] localedata: wrong collation order of diacritics in most locales
722 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
724 [17956] build: Build fails on missing definitions from header file
725 nss/nss.h when Mozilla NSS is used for cryptography
726 [18203] libc: realpath() does not handle unreachable paths correctly
727 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
728 [18812] localedata: kab_DZ: new Kabyle Algeria locale
729 [18822] libc: Internal functions are called via PLT
730 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
731 [19170] libc: __gmon_start__ defined in hppa in crtn.S
732 [19574] libc: glibc should support building static PIE binaries
733 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
735 [19971] glob: glob: Do not skip entries with zero d_ino values
736 [19982] localedata: fr.po: spelling mistake for error code EXDEV
737 [20008] localedata: km_KH: convert to translit_neutral
738 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
739 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
740 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
741 [20482] localedata: de_CH: abbreviated weekdays should be two letters
742 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
744 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
745 they're set, wrong errors returned by gaih_inet when lookup functions
747 [20756] localedata: [PATCH] Use Unicode wise thousands separator
748 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
750 [20952] localedata: yuw_PG: new locale
751 [21084] localedata: charmaps/IBM858: new codepage
752 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
753 [21242] libc: assert gives pedantic warning in old gcc versions
754 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
755 __regcall calling convention
756 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
757 [21326] libc: C99 functions are not declared for C++11 and later if
758 _GNU_SOURCE is not predefined by g++
759 [21457] libc: sys/ucontext.h namespace
760 [21530] libc: tmpfile() should be implemented using O_TMPFILE
761 [21660] math: GCC fails to compile a formula with tgmath.h
762 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
763 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
764 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
765 [21684] math: tgmath.h handling of complex integers
766 [21685] math: tgmath.h handling of bit-fields
767 [21686] math: tgmath.h handling of __int128
768 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
770 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
772 [21750] localedata: column width of characters incompatible with classical
774 [21754] malloc: malloc: Perform as little work as possible after heap
775 consistency check failures
776 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
777 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
778 [21791] string: Unused XXX_chk_XXX functions in libc.a
779 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
781 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
783 [21845] localedata: Added new Locale bho_NP
784 [21853] localedata: Fix abday Which looks same as day in zh_SG
785 [21854] localedata: Added New Locale en_SC
786 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
787 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
788 _dl_runtime_resolve_avx_slow
789 [21885] network: getaddrinfo: gethosts does not release resolver context
790 on memory allocation failure
791 [21899] libc: XPG4.2 sigaction namespace
792 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
794 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
796 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
797 errno value, causing getaddrinfo to fail
798 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
799 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
800 instead of EAI_NODATA
801 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
802 constant PTRACE_SEIZE_DEVEL
803 [21930] math: C-only gcc builtins used in <math.h> isinf
804 [21932] network: Unpaired __resolv_context_get in generic get*_r
806 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
807 [21944] libc: sigval namespace
808 [21951] localedata: Update hanzi collation by stroke
809 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
810 sysdeps/x86_64/fpu/e_expf.S
811 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
812 allocation in syscall loops)
813 [21959] localedata: Fix Country name for xh_ZA
814 [21960] localedata: Fix abmon for bem_ZM
815 [21966] math: AVX2 mathvec functions use FMA without checking
816 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
817 [21971] localedata: Added New Locale for mfe_MU
818 [21972] libc: assert macro requires operator== (int) for its argument type
819 [21973] math: [sparc] libm missing sqrtl compat symbol
820 [21974] libc: Remove __bb_init_func and __bb_exit_func
821 [21982] string: stratcliff.c: error: assuming signed overflow does not
823 [21986] stdio: __guess_grouping is called incorrectly
824 [21987] math: [sparc32] wrong bits/long-double.h installed
825 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
827 [22022] localedata: Missing country_name for mni_IN
828 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
830 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
832 [22026] locale: iconv_open: heap overflow on gconv_init failure
833 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
834 [22035] math: [m68k] bits/math-inline.h macro namespace
835 [22038] localedata: Fix abbreviated weeks and months for Somali
836 [22044] localedata: Remove redundant data for Limburgish Language
837 [22050] malloc: Linking with -lmcheck does not hook
838 __malloc_initialize_hook correctly
839 [22051] libc: zero terminator in the middle of glibc's .eh_frame
840 [22052] malloc: malloc failed to compile with GCC 7 and -O3
841 [22070] localedata: charmaps/UTF-8: wcwidth for
842 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
843 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
844 Jungseong and Jongseong) should be 0
845 [22078] nss: nss_files performance issue in multi mode
846 [22082] math: bits/math-finite.h exp10 condition
847 [22086] libc: pcprofiledump incorrect cross-endian condition
848 [22093] dynamic-link: ld.so no longer searches in .../x86_64
849 [22095] network: Name server address allocation memory leak in resolv.conf
851 [22096] network: __resolv_conf_attach can incorrectly free passed conf
853 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
854 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
856 [22111] malloc: malloc: per thread cache is not returned when thread exits
857 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
858 [22134] libc: [linux] implement fexecve with execveat
859 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
861 [22145] libc: ttyname() gives up too early in the face of namespaces
862 [22146] math: C++ build issue with float128 on x86_64
863 [22153] nptl: nptl: save error code before process termination
864 [22156] libc: [hppa,ia64,microblaze] Executable stack default
865 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
866 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
867 [22165] libc: [hppa] Text relocations in libc.so
868 [22180] libc: destructor registered via __cxa_atexit is called twice
869 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
870 symlinks") cause "make" segfaults
871 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
873 [22207] libc: FAIL: stdlib/test-atexit-race
874 [22225] math: nearbyint arithmetic moved before feholdexcept
875 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
876 [22235] math: iscanonical in C++ and float128
877 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
878 the svid compat wrapper
879 [22244] math: ynf and yn are wrong without the svid compat wrapper
880 [22273] libc: Improper assert in Linux posix_spawn implementation
881 [22284] libc: -pg -pie doesn't work
882 [22292] locale: localedef exits with error 4 when it should be error 1
883 [22294] locale: Allow "" for int_currency_symbol definition in locales.
884 [22295] locale: Don't warn on non-symbolic characters in locale sources in
886 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
887 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
889 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
890 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
891 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
892 [22322] libc: [mips64] wrong bits/long-double.h installed
893 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
894 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
896 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
897 [22343] malloc: Integer overflow in posix_memalign
898 [22347] libc: getrandom() returns the number of bytes that were copied to
899 the buffer even though the comments say "Return 0 on success and -1 on
901 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
902 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
904 [22370] dynamic-link: Incorrect note padding check
905 [22375] libc: malloc returns pointer from tcache_get when should return
906 NULL (CVE-2017-17426)
907 [22377] math: iseqsig, float128 and C++
908 [22382] localedata: Error in tpi_PG locale
909 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
910 inside the ASCII printable range
911 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
913 [22403] localedata: Slash needs escaping in some locales
914 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
915 [22409] network: res_hnok does not accept some host names used on the
917 [22412] network: res_dnok, res_hnok should perform syntax checks
918 [22413] network: ns_name_pton ignores syntactically invalid trailing
920 [22415] stdio: setvbuf can lead to invalid free/segfault
921 [22432] build: Non-deterministic build
922 [22439] malloc: malloc_info should compute summary statistics for all sub-
924 [22442] network: if_nametoindex could report index for the wrong
926 [22446] build: aliasing violation calling readlink in handle_request
927 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
928 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
929 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
930 __nscd_hash/__nss_hash
931 [22463] network: p_secstodate overflow handling
932 [22469] localedata: pl_PL LC_COLLATE does not use i18n
933 [22478] libc: sigwait can fail with EINTR
934 [22505] libc: ldconfig processes include directive in locale-specific
936 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
937 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
938 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
939 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
940 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
941 [22534] localedata: Collation rules for Serbian and Bosnian should be the
943 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
944 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
945 [22577] libc: missing newline after "cannot allocate TLS data structures
947 [22588] manual: manual/conf.texi: missing underscore in front of
949 [22593] math: nextafter and nexttoward are declared with const attribute
950 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
951 [22603] string: ia64 memchr overflows internal pointer check
952 [22605] libc: SH clone does not set the exit code correctly
953 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
955 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
956 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
957 modified in case of success
958 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
959 [22615] manual: manual: ambiguous wording about errno value in case of
961 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
963 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
964 binaries or if /proc is not mounted (CVE-2017-16997)
965 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
966 [22630] build: $(no-pie-ldflag) is no longer effective
967 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
968 [22635] nptl: pthread_self returns NULL before libpthread is loaded
969 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
970 [22637] nptl: guard size is subtracted from thread stack size instead of
972 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
973 [22657] localedata: hu_HU: Avoid double space in date
974 [22660] math: fmax, fmin sNaN handling on alpha
975 [22664] libc: New warning of GCC8
976 [22665] math: alpha: ceil and floor raise inexact exceptions
977 [22666] math: alpha: trunc raise inexact exceptions
978 [22667] libc: makecontext lacks stack alignment on i386
979 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
980 [22679] libc: getcwd(3) can succeed without returning an absolute path
982 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
984 [22687] math: [powerpc-nofpu] complex long double functions spurious
986 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
987 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
988 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
990 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
991 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
993 [22701] nis: Incomplete removal of libnsl
994 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
995 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
996 [22715] dynamic-link: FAIL: elf/tst-audit10
997 [22719] libc: Backtrace tests fail on hppa
998 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
999 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1000 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1001 before the first call to crypt_r () in crypt/badsalttest.c
1008 * A per-thread cache has been added to malloc. Access to the cache requires
1009 no locks and therefore significantly accelerates the fast path to allocate
1010 and free small amounts of memory. Refilling an empty cache requires locking
1011 the underlying arena. Performance measurements show significant gains in a
1012 wide variety of user workloads. Workloads were captured using a special
1013 instrumented malloc and analyzed with a malloc simulator. Contributed by
1014 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1016 * Unicode 10.0.0 Support: Character encoding, character type info, and
1017 transliteration tables are all updated to Unicode 10.0.0, using
1018 generator scripts contributed by Mike FABIAN (Red Hat).
1019 These updates cause user visible changes, especially the changes in
1020 wcwidth for many emoji characters cause problems when emoji sequences
1021 are rendered with pango, see for example:
1022 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1024 * Collation of Hungarian has been overhauled and is now consistent with "The
1025 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1028 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1030 - The GNU C Library will now detect when /etc/resolv.conf has been
1031 modified and reload the changed configuration. The new resolver option
1032 “no-reload” (RES_NORELOAD) disables this behavior.
1034 - The GNU C Library now supports an arbitrary number of search domains
1035 (configured using the “search” directive in /etc/resolv.conf);
1036 previously, there was a hard limit of six domains. For backward
1037 compatibility, applications that directly modify the ‘_res’ global
1038 object are still limited to six search domains.
1040 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1041 Library will now randomly pick a name server from the configuration as a
1042 starting point. (Previously, the second name server was always used.)
1044 * The tunables feature is now enabled by default. This allows users to tweak
1045 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1047 * New function reallocarray, which resizes an allocated block (like realloc)
1048 to the product of two sizes, with a guaranteed clean failure upon integer
1049 overflow in the multiplication. Originally from OpenBSD, contributed by
1050 Dennis Wölfing and Rüdiger Sonderfeld.
1052 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1053 These are extended versions of preadv and pwritev, respectively, taking an
1054 additional flags argument. The set of supported flags depends on the
1055 running kernel; full support currently requires kernel 4.7 or later.
1057 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1058 create a new session ID for the spawned process. This feature is
1059 scheduled to be added to the next major revision of POSIX; for the time
1060 being, it is available under _GNU_SOURCE.
1062 * errno.h is now safe to use from C-preprocessed assembly language on all
1063 supported operating systems. In this context, it will only define the
1064 Exxxx constants, as preprocessor macros expanding to integer literals.
1066 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1067 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1068 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1069 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1071 To compile programs that use this feature, the compiler must support
1072 128-bit floating point with the type name _Float128 (as defined by TS
1073 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1074 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1075 must be defined to make the new interfaces visible.
1077 The new functions and macros correspond to those present for other
1078 floating-point types (except for a few obsolescent interfaces not
1079 supported for the new type), with F128 or f128 suffixes; for example,
1080 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1081 printf or scanf formats for the new type; the strfromf128 and strtof128
1082 interfaces should be used instead.
1084 Deprecated and removed features, and other changes affecting compatibility:
1086 * The synchronization that pthread_spin_unlock performs has been changed to
1087 now be equivalent to a C11 atomic store with release memory order to the
1088 spin lock's memory location. Previously, several (but not all)
1089 architectures used stronger synchronization (e.g., containing what is
1090 often called a full barrier). This change can improve performance, but
1091 may affect odd fringe uses of spin locks that depend on the previous
1092 behavior (e.g., using spin locks as atomic variables to try to implement
1093 Dekker's mutual exclusion algorithm).
1095 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1098 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1099 will only be built and installed when the GNU C Library is configured with
1100 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1101 as TIRPC or rpcsvc-proto, to be used.
1103 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1104 libnss_compat, are deprecated, and will not be built or installed by
1107 The NIS(+) support library, libnsl, is also deprecated. By default, a
1108 compatibility shared library will be built and installed, but not headers
1109 or development libraries. Only a few NIS-related programs require this
1110 library. (In particular, the GNU C Library has never required programs
1111 that use 'gethostbyname' to be linked with libnsl.)
1113 Replacement implementations based on TIRPC, which additionally support
1114 IPv6, are available from <https://github.com/thkukuk/>. The configure
1115 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1116 name service modules, to be built and installed.
1118 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1119 support is enabled, the configured recursive resolver must support EDNS.
1120 (Responding to EDNS-enabled queries with responses which are not
1121 EDNS-enabled is fine, but FORMERR responses are not.)
1123 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1124 servers have not supported this opcode for a long time.
1126 * The _res_opcodes variable has been removed from libresolv. It had been
1127 exported by accident.
1129 * <string.h> no longer includes inline versions of any string functions,
1130 as this kind of optimization is better done by the compiler. The macros
1131 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1133 * The nonstandard header <xlocale.h> has been removed. Most programs should
1134 use <locale.h> instead. If you have a specific need for the definition of
1135 locale_t with no other declarations, please contact
1136 libc-alpha@sourceware.org and explain.
1138 * The obsolete header <sys/ultrasound.h> has been removed.
1140 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1142 * The obsolete function cfree has been removed. Applications should use
1145 * The stack_t type no longer has the name struct sigaltstack. This changes
1146 the C++ name mangling for interfaces involving this type.
1148 * The ucontext_t type no longer has the name struct ucontext. This changes
1149 the C++ name mangling for interfaces involving this type.
1151 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1152 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1153 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1154 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1155 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1156 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1157 fpregset_t type no longer has the name struct fpu. This changes the C++
1158 name mangling for interfaces involving those types.
1160 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1161 synced with the kernel:
1163 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1164 are not supported on this architecture and have been removed.
1166 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1167 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1168 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1170 Programs that assume the GET/SETREGS ptrace requests are universally
1171 available will now fail to build, instead of malfunctioning at runtime.
1173 Changes to build and runtime requirements:
1175 * Linux kernel 3.2 or later is required at runtime, on all architectures
1176 supported by that kernel. (This is a change from version 2.25 only for
1179 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1181 * On most architectures, GCC 4.9 or later is required to build the GNU C
1182 Library. On powerpc64le, GCC 6.2 or later is required.
1184 Older GCC versions and non-GNU compilers are still supported when
1185 compiling programs that use the GNU C Library. (We do not know exactly
1186 how old, and some GNU extensions to C may be _de facto_ required. If you
1187 are interested in helping us make this statement less vague, please
1188 contact libc-alpha@sourceware.org.)
1190 Security related changes:
1192 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1193 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1195 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1196 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1198 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1199 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1202 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1203 fixed (CVE-2017-12133).
1205 The following bugs are resolved with this release:
1207 [984] network: Respond to changed resolv.conf in gethostbyname
1208 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1209 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1211 [12189] libc: __stack_chk_fail should not attempt a backtrace
1213 [14096] time: Race condition on timezone/tst-timezone.out
1214 [14172] localedata: az_IR: new locale
1215 [14995] build: glibc fails to build if gold is the default linker, even if
1217 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1218 [16637] network: inet_pton function is accepting IPv6 with bad format
1219 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1220 x86 and other generic code
1221 [16875] localedata: ko_KR: fix lang_name
1222 [17225] localedata: ar_SY: localized month names for May and June are
1224 [17297] localedata: da_DK: wrong date_fmt string
1225 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1226 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1227 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1228 [19066] localedata: ar_SA abbreviated day and month names are in English
1229 [19569] network: resolv: Support an arbitrary number of search domains
1230 [19570] network: Implement random DNS server selection in the stub
1232 [19838] locale: localedef fails on PA-RISC
1233 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1234 order of 0D36 and 0D37
1235 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1237 [20098] libc: FAIL: debug/backtrace-tst on hppa
1238 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1240 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1242 [20313] localedata: Update locale data to Unicode 9.0
1243 [20424] manual: Document how to provide a malloc replacement
1244 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1246 [20686] locale: Add el_GR@euro to SUPPORTED.
1247 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1248 failures consistently
1249 [21015] dynamic-link: Document and fix --enable-bind-now
1250 [21016] nptl: pthread_cond support is broken on hppa
1251 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1252 frame-pointer on i386
1253 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1255 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1256 [21088] libc: Build fails with --enable-static-nss
1257 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1258 generic c code is used
1259 [21109] libc: Tunables broken on big-endian
1260 [21112] math: powf has large ulp errors with base close to 1 and exponent
1262 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1264 [21120] malloc: glibc malloc is incompatible with GCC 7
1265 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1267 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1268 (-0.0) when linking with -lieee
1269 [21171] math: log10, log2 and lgamma return incorrect results
1270 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1271 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1272 [21207] localedata: ce_RU: update weekdays from CLDR
1273 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1274 [21217] localedata: Update months from CLDR-31
1275 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1276 [21243] libc: support_delete_temp_file should issue warning for failed
1278 [21244] libc: support resolv_test_start() socket fd close should be
1280 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1281 new posix_spawn implementation
1282 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1283 leads to lower CPU frequency
1284 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1285 [21261] libc: [sparc64] bits/setjmp.h namespace
1286 [21267] network: [mips] bits/socket.h IOC* namespace
1287 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1288 [21270] libc: mmap64 silently truncates large offset values
1289 [21275] libc: posix_spawn always crashes on ia64 now
1290 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1291 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1292 [21289] libc: Incorrect declaration for 32-bit platforms with
1293 _FILE_OFFSET_BITS=64 causes build error
1294 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1295 not support gethostbyname4_r
1296 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1298 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1300 [21340] libc: Support POSIX_SPAWN_SETSID
1301 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1302 [21359] network: ns_name_pack needs additional byte in destination buffer
1303 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1304 against fragmentation attacks (CVE-2017-12132)
1305 [21369] network: resolv: Remove EDNS fallback
1306 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1308 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1309 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1310 [21393] stdio: Missing dup3 error check in freopen, freopen64
1311 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1312 [21399] localedata: Bad description for U00EC in
1313 localedata/charmaps/CP1254
1314 [21411] malloc: realloc documentation error
1315 [21426] network: sys/socket.h uio.h namespace
1316 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1317 [21445] libc: signal.h bsd_signal namespace
1318 [21455] network: Network headers stdint.h namespace
1319 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1321 [21475] network: resolv: Overlong search path is truncated mid-label
1322 [21511] libc: sigstack namespace
1323 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1324 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1325 bits/syscall.h: No such file or directory
1326 [21517] libc: struct sigaltstack namespace
1327 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1328 [21533] localedata: Update locale data to Unicode 10.0
1330 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1331 junk at end of line, first unrecognized character is `@'
1332 [21538] libc: SIG_HOLD missing for XPG4
1333 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1334 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1335 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1336 [21543] libc: sigevent namespace
1337 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1338 size for general purpose registers in mcontext_t structure
1339 [21550] libc: sigwait namespace
1340 [21552] libc: XPG4 bsd_signal namespace
1341 [21554] libc: sigpause namespace
1342 [21560] libc: sys/wait.h signal.h namespace
1343 [21561] libc: waitid namespace
1344 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1345 '/usr/include/stdlib.h': Permission denied
1346 [21575] libc: sys/wait.h missing struct rusage definition
1347 [21584] libc: sigaltstack etc namespace
1348 [21597] libc: siginterrupt namespace
1349 [21607] math: hppa: FAIL: math/test-tgmath
1350 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1352 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1353 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1354 stack and heap (CVE-2017-1000366)
1355 [21625] libc: wait3 namespace
1356 [21654] nss: Incorrect pointer alignment in NSS group merge result
1358 [21657] network: Parse interface zone id for node-local multicast
1359 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1361 [21666] libc: .symver is used on common symbol
1362 [21668] network: resolv: res_init cross-thread broadcast introduces race
1364 [21687] math: tgmath.h totalorder, totalordermag return type
1365 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1367 [21696] libc: Incorrect assumption of of __cpu_mask in
1368 posix/sched_cpucount.c
1369 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1370 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1372 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1373 leading to relocation crash
1374 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1375 `resolv_conf_matches (resp, conf)' failed.
1376 [21710] localedata: Added Samoan language locale for Samoa
1377 [21711] localedata: Pashto yesstr/nostr locale are missing
1378 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1379 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1380 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1381 [21723] localedata: yesstr/nostr missing for Chinese language locale
1382 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1384 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1386 [21728] localedata: New Locale for Tongan language
1387 [21729] localedata: incorrect LC_NAME fields for hi_IN
1388 [21733] localedata: yesstr and nostr are missing for zh_HK
1389 [21734] localedata: Missing yesstr and nostr are for kw_GB
1390 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1391 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1392 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1393 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1395 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1397 [21749] localedata: Wrong abbreviated day name (“abday”) for
1399 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1400 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1401 [21759] localedata: missing yesstr and nostr for Tigrinya
1402 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1403 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1404 [21767] localedata: Missing Bislama locales
1405 [21768] localedata: Missing yesstr and nostr for aa_ET
1406 [21770] localedata: Missing Field in li_NL
1407 [21778] nptl: Robust mutex may deadlock
1408 [21779] libc: MicroBlaze segfaults when loading libpthread
1409 [21783] localedata: Fix int_select international_call_prefixes
1410 [21784] localedata: Inconsistency in country_isbn
1411 [21788] localedata: Missing Country Postal Abbreviations
1412 [21794] localedata: Added-country_isbn-for-Italy
1413 [21795] localedata: Add/Fix country_isbn for France
1414 [21796] localedata: Added country_isbn for Republic of Korea
1415 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1416 [21799] localedata: Added int_select international_call_prefixes
1417 [21801] localedata: Added int_select international_call_prefixes
1418 [21804] nptl: Double semicolon in thread-shared-types.h
1419 [21807] localedata: LC_ADDRESS fix for pap_CW
1420 [21808] localedata: Fix LC_ADDRESS for pap_AW
1421 [21821] localedata: Added country_name in mai_IN
1422 [21822] localedata: Fix LC_TIME for mai_IN
1423 [21823] localedata: missing yesstr, nostr for sa_IN
1424 [21825] localedata: Fix name_mrs for mag_IN
1425 [21828] localedata: 2.26 changelog should mention user visible changes
1427 [21835] localedata: Added Maithili language locale for Nepal
1428 [21838] localedata: Removed redundant data for the_NP
1429 [21839] localedata: Fix LC_MONETARY for ta_LK
1430 [21844] localedata: Fix Latin characters and Months Sequence.
1431 [21848] localedata: Fix mai_NP Title Name
1436 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1437 24731-2:2010, is supported to enable declarations of functions from that
1438 TR. Note that not all functions from that TR are supported by the GNU C
1441 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1442 18661-1:2014, is supported to enable declarations of functions and macros
1443 from that TS. Note that not all features from that TS are supported by
1446 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1447 18661-4:2015, is supported to enable declarations of functions and macros
1448 from that TS. Note that most features from that TS are not supported by
1451 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1452 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1453 Since the GNU C Library defaults to a much newer revision of POSIX, this
1454 will only affect programs that specifically request an old conformance
1455 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1456 see a change in the visible declarations, but a program compiled with
1457 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1460 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1461 defined by all multithreaded code, but glibc has not required this for
1464 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1465 means that in a future release, the macros “major”, “minor”, and “makedev”
1466 will only be available from <sys/sysmacros.h>.
1468 These macros are not part of POSIX nor XSI, and their names frequently
1469 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1470 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1471 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1474 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1475 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1476 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1478 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1479 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1480 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1481 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1482 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1483 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1484 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1485 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1486 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1487 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1488 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1491 * New <math.h> features are added from TS 18661-1:2014:
1493 - Signaling NaN macros: SNANF, SNAN, SNANL.
1495 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1496 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1497 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1499 - llogb functions: the llogb, llogbf and llogbl functions, and the
1500 FP_LLOGB0 and FP_LLOGBNAN macros.
1502 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1505 - Comparison macros: iseqsig.
1507 - Classification macros: iscanonical, issubnormal, iszero.
1509 - Total order functions: totalorder, totalorderf, totalorderl,
1510 totalordermag, totalordermagf, totalordermagl.
1512 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1514 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1515 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1517 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1518 are added to libc. They convert a floating-point number into string.
1520 * Most of glibc can now be built with the stack smashing protector enabled.
1521 It is recommended to build glibc with --enable-stack-protector=strong.
1522 Implemented by Nick Alcock (Oracle).
1524 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1525 intended to be used instead of memset() to erase sensitive data after use;
1526 the compiler will not optimize out calls to explicit_bzero even if they
1527 are "unnecessary" (in the sense that no _correct_ program can observe the
1528 effects of the memory clear).
1530 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1531 to float instead of double. This does not affect the ABI of any libraries
1532 that are part of the GNU C Library, but may affect the ABI of other
1533 libraries that use this type in their interfaces.
1535 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1536 float_t and double_t types are now defined to long double instead of float
1537 and double. These options are not the default, and this does not affect
1538 the ABI of any libraries that are part of the GNU C Library, but it may
1539 affect the ABI of other libraries that use this type in their interfaces,
1540 if they are compiled or used with those options.
1542 * The getentropy and getrandom functions, and the <sys/random.h> header file
1545 * The buffer size for byte-oriented stdio streams is now limited to 8192
1546 bytes by default. Previously, on Linux, the default buffer size on most
1547 file systems was 4096 bytes (and thus remains unchanged), except on
1548 network file systems, where the buffer size was unpredictable and could be
1549 as large as several megabytes.
1551 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1552 for the Linux quota interface which predates kernel version 2.4.22 has
1555 * The malloc_get_state and malloc_set_state functions have been removed.
1556 Already-existing binaries that dynamically link to these functions will
1557 get a hidden implementation in which malloc_get_state is a stub. As far
1558 as we know, these functions are used only by GNU Emacs and this change
1559 will not adversely affect already-built Emacs executables. Any undumped
1560 Emacs executables, which normally exist only during an Emacs build, should
1561 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1563 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1564 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1565 “no-ip6-dotint” had already been the default, and support for the
1566 “ip6-dotint” option was removed from the Internet in 2006.
1568 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1569 flag from <resolv.h> have been removed. The option relied on a
1570 backwards-incompatible DNS extension which was never deployed on the
1573 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1574 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1575 They were already unimplemented.
1577 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1578 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1579 removed again from the IETF name lookup interface specification in RFC
1580 2553. Applications should use getaddrinfo instead.
1582 * DNSSEC-related declarations and definitions have been removed from the
1583 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1584 decode the data part of DNSSEC record types. Previous versions of glibc
1585 only implemented minimal support for the previous version of DNSSEC, which
1586 is incompatible with the currently deployed version.
1588 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1589 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1590 <arpa/nameser.h> header file because the distinction between RR types and
1591 meta-RR types is not officially standardized, subject to revision, and
1592 thus not suitable for encoding in a macro.
1594 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1595 and rhook members of the res_state type in <resolv.h> have been removed.
1596 The glibc stub resolver did not support these hooks, but the header file
1597 did not reflect that.
1599 * For multi-arch support it is recommended to use a GCC which has
1600 been built with support for GNU indirect functions. This ensures
1601 that correct debugging information is generated for functions
1602 selected by IFUNC resolvers. This support can either be enabled by
1603 configuring GCC with '--enable-gnu-indirect-function', or by
1604 enabling it by default by setting 'default_gnu_indirect_function'
1605 variable for a particular architecture in the GCC source file
1608 * GDB pretty printers have been added for mutex and condition variable
1609 structures in POSIX Threads. When installed and loaded in gdb these pretty
1610 printers show various pthread variables in human-readable form when read
1611 using the 'print' or 'display' commands in gdb.
1613 * Tunables feature added to allow tweaking of the runtime for an application
1614 program. This feature can be enabled with the '--enable-tunables' configure
1615 flag. The GNU C Library manual has details on usage and README.tunables has
1616 instructions on adding new tunables to the library.
1618 * A new version of condition variables functions have been implemented in
1619 the NPTL implementation of POSIX Threads to provide stronger ordering
1622 * A new version of pthread_rwlock functions have been implemented to use a more
1623 scalable algorithm primarily through not using a critical section anymore to
1626 Security related changes:
1628 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1629 have been created with makecontext could fail to terminate due to a
1630 missing .cantunwind annotation. This has been observed to lead to a hang
1631 (denial of service) in some Go applications compiled with gccgo. Reported
1632 by Andreas Schwab. (CVE-2016-6323)
1634 * The DNS stub resolver functions would crash due to a NULL pointer
1635 dereference when processing a query with a valid DNS question type which
1636 was used internally in the implementation. The stub resolver now uses a
1637 question type which is outside the range of valid question type values.
1640 The following bugs are resolved with this release:
1642 [4099] stdio: Overly agressive caching by stream i/o functions.
1643 [7065] build: Support building glibc with -fstack-protector or -fstack-
1645 [9842] localedata: en_CA: incorrect date format
1646 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1647 before it started waiting
1648 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1649 [15765] nptl: sem_open is wrongly a cancellation point
1650 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1651 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1653 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1654 library linked with pthread
1655 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1656 "FIXME: Ingo" issue)
1657 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1658 [17252] libc: getrandom and getentropy syscall
1659 [17730] malloc: thread-local storage is sometimes improperly free()'d
1660 after being __libc_memalign()'d
1661 [18241] stdio: failed fseek on memstream does not set errno and can fail
1663 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1665 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1666 PTHREAD_PRIO_INHERIT mutex on ARM
1667 [18784] network: res_query and related function crash for special record
1668 type queries (CVE-2015-5180)
1669 [19380] math: strtod does not raise "inexact"
1670 [19387] string: Integer overflow in memchr
1671 [19390] string: Integer overflow in strncat
1672 [19398] build: linknamespace tests fail with massively parallel build
1673 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1675 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1676 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1677 [19514] libc: [PATCH] Fix spelling errors in spelling
1678 "implement"/"implementation" in several places
1679 [19582] network: Deprecate RES_USE_INET6
1680 [19673] manual: clog10 docs appear to be erroneous
1681 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1682 causes a segmentation fault
1683 [19826] libc: invalid pointer returned from __tls_get_addr with static
1685 [20016] network: resolv: Remove hooks support from the API
1686 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1688 [20033] math: [x86_64] vectorized math function don't call the __finite
1690 [20116] nptl: use after free in pthread_create
1691 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1692 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1694 [20311] nptl: please install proc_service.h
1695 [20366] libc: Compilation errors in installed headers in strict-compliance
1697 [20370] malloc: malloc: Arena free list management is still racy
1698 (incorrect fix in bug 19243)
1699 [20386] libc: assert (X = 0) does not result in GCC warning
1700 [20432] malloc: malloc: Minimize interface required for interposition
1701 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1702 in _Unwind_Backtrace (CVE-2016-6323)
1703 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1704 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1706 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1707 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1708 [20477] network: resolv: incorrect double-checked locking related to
1710 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1712 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1714 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1715 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1716 cause transition penalty
1717 [20517] math: sparcv9 missing fdiml compat symbol
1718 [20524] manual: strverscmp is inconsistent
1719 [20525] libc: <sys/quota.h> should be based on kernel headers
1720 [20539] math: GCC 7 -static -lm fails to link at -O3
1721 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1723 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1725 [20591] network: Remove obsolete DNSSEC support
1726 [20592] network: DNS resource record type classification macros in
1727 <arpa/nameser.h> are incorrect
1728 [20593] network: Update DNS RR type definitions
1729 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1730 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1732 [20629] network: libresolv: Remove support for bitstring labels
1734 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1735 assertion: "maxidx >= 2"
1736 [20660] math: [arm] Use VSQRT
1737 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1738 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1739 [20689] libc: Test for FMA should also check for AVX.
1740 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1741 [20715] math: iszero macro breaks existing code
1742 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1743 [20728] libc: powerpc: Missing TOC stub in clone
1744 [20729] build: glibc-2.24 fails to build for i486 with -Os
1745 [20750] build: Build fails with default PIE enabled gcc-6.x
1746 [20768] math: [alpha] sqrt fegetenv namespace
1747 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1749 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1750 __FLT_EVAL_METHOD__ is 2
1751 [20790] network: rpcgen buffer overrun in get_prog_declaration
1752 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1753 [20829] libc: crypt snprintf namespace
1754 [20847] libc: tst-vfork3 failure
1755 [20855] math: Default bits/mathdef.h has inappropriate float_t
1756 [20859] math: [sh4] FP_ILOGB0 invalid
1757 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1758 U+20AC), not same as GBK
1759 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1760 [20916] math: pow handling of sNaN arguments
1761 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1762 [20919] math: Bad pow (qNaN, 0) result with -lieee
1763 [20940] math: hypot sNaN handling
1764 [20947] math: fmax, fmin sNaN handling
1765 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1766 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1768 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1769 [20973] nptl: robust mutexes: Lost wake-ups
1770 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1771 [20978] nis: strlen on null pointer in nss_nisplus
1772 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1774 [21014] string: i686 memchr overflows internal pointer check
1775 [21019] libc: [mips] n32 lseek incorrect on overflow
1776 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1777 [21026] libc: [MIPS] readahead syscall is broken on n64
1778 [21028] math: Fallback fesetexceptflag should always succeed
1779 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1780 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1782 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1783 [21061] librt: [microblaze] librt lost clock_* exports
1784 [21073] libc: tunables: insecure environment variables passed to
1785 subprocesses with AT_SECURE
1786 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1790 * The minimum Linux kernel version that this version of the GNU C Library
1791 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1792 kernel version 2.6.32 or later suffices (on architectures that already
1793 required kernel versions more recent than 3.2, those requirements remain
1794 unchanged). Linux 3.2 or later kernel headers are required on all
1797 * The pap_AN locale has been deleted. This has been deprecated for a long
1798 time. It has been replaced by pap_AW & pap_CW, both of which have long
1799 been included in previous releases.
1801 * The readdir_r and readdir64_r functions have been deprecated. It is
1802 recommended to use readdir and readdir64 instead.
1804 * The type “union wait” has been removed. It was deprecated in the early
1805 1990s and never part of POSIX. Application code should use the int type
1806 instead of “union wait”.
1808 * A new NSS action is added to facilitate large distributed system
1809 administration. The action, MERGE, allows remote user stores like LDAP
1810 to be merged into local user stores like /etc/groups in order to provide
1811 easy to use, updated, and managed sets of merged credentials. The new
1812 action can be used by configuring it in /etc/nsswitch.conf:
1813 group: files [SUCCESS=merge] nis
1814 Implemented by Stephen Gallagher (Red Hat).
1816 * The deprecated __malloc_initialize_hook variable has been removed from the
1819 * The long unused localedef --old-style option has been removed. It hasn't
1820 done anything in over 16 years. Scripts using this option can safely
1823 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1824 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1825 return the next representable value in the direction of positive infinity
1826 and the nextdown functions return the next representable value in the
1827 direction of negative infinity. These are currently enabled as GNU
1830 Security related changes:
1832 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1833 could result in a stack overflow when getnetbyname was called with an
1834 overly long name. (CVE-2016-3075)
1836 * Previously, getaddrinfo copied large amounts of address data to the stack,
1837 even after the fix for CVE-2013-4458 has been applied, potentially
1838 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1839 instead. Reported by Michael Petlan. (CVE-2016-3706)
1841 * The glob function suffered from a stack-based buffer overflow when it was
1842 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1843 Reported by Alexander Cherepanov. (CVE-2016-1234)
1845 * The Sun RPC UDP client could exhaust all available stack space when
1846 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1847 alloca plugin for GCC. (CVE-2016-4429)
1849 * The IPv6 name server management code in libresolv could result in a memory
1850 leak for each thread which is created, performs a failing naming lookup,
1851 and exits. Over time, this could result in a denial of service due to
1852 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1854 The following bugs are resolved with this release:
1856 [1170] localedata: ne_NP: update Nepali locale definition file
1857 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1859 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1860 [6796] math: fdim() does not set errno on overflow
1861 [10354] libc: posix_spawn should use vfork() in more cases than presently
1862 [11213] localedata: localedata: add copyright disclaimer to locale files
1863 [12143] localedata: chr_US: new Cherokee locale
1864 [12450] localedata: sgs_LT: new locale
1865 [12676] localedata: ln_CD: new locale
1866 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1868 [13304] math: fma, fmaf, fmal produce wrong results
1869 [14259] build: --localedir arg to configure is ignored
1870 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1871 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1872 [14934] localedata: es_CL: wrong first weekday chilean locale
1873 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1875 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1877 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1878 [15368] nptl: raise() is not async-signal-safe
1879 [15479] math: ceil, floor, round and trunc raise inexact exception
1880 [15578] localedata: kk_KZ: various updates
1881 [16003] localedata: pap_AN: punt old locale
1882 [16137] localedata: iw_IL: punt old locale
1883 [16190] localedata: eo: new esperanto locale
1884 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1885 [16742] malloc: race condition: pthread_atfork() called before first
1886 malloc() results in unexpected locking behaviour/deadlocks
1887 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1889 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1890 [17565] localedata: pt_PT: wrong (work-)week start
1891 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1892 [17950] build: Build fails with -msse
1893 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1894 [18433] libc: posix_spawn does not return correctly upon failure to
1896 [18453] localedata: charmaps/IBM875: incorrect codes
1897 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1899 [18896] localedata: he_IL: improvements for currency
1900 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1901 Romanian locale data
1902 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1904 [19056] libc: Deprecate readdir_r
1905 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1907 [19198] localedata: nl_NL: small improvements for Dutch locales
1908 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1909 nameservers (CVE-2016-5417)
1910 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1912 [19400] locale: Language missing in "iso-639.def", trivial fix in
1914 [19431] malloc: Deadlock between fflush, getdelim, and fork
1915 [19505] libc: Incorrect file descriptor validity checks in
1916 posix_spawn_file_actions_add{open,close,dup2}
1917 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1918 when using RTLD_NEXT
1919 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1920 `intl/{gettextP,loadinfo}.h'
1921 [19534] libc: execle, execlp may use malloc
1922 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1923 [19573] network: res_nclose and __res_maybe_init disagree about name
1924 server initialization, breaking Hesiod
1925 [19575] localedata: Status of GB18030 tables
1926 [19581] localedata: sr_* date_fmt string contains additional newline
1927 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1929 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1930 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1931 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1932 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1934 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1936 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1938 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1939 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1940 [19633] locale: strfmon_l applies global locale to number formatting
1941 [19642] network: Memory leak in getnameinfo
1942 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1943 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1945 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1946 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1948 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1949 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1951 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1953 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1955 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1956 pointers and lengths in error-case.
1957 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1958 report errors on UTF-16 surrogates values.
1959 [19755] nscd: nscd assertion failure in gc
1960 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1961 [19759] libc: mempcpy shouldn't be inlined
1962 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1963 [19765] libc: s390 needs an optimized mempcpy
1964 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1965 NAME_MAX limit assumption (CVE-2016-1234)
1966 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1967 [19787] network: Missing and incorrect truncation checks in getnameinfo
1968 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1970 [19791] network: Assertion failure in res_query.c with un-connectable name
1972 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1973 [19822] math: libm.so install clobbers old version
1974 [19825] network: resolv: send_vc can return uninitialized data in second
1975 response to getaddrinfo
1976 [19830] network: nss_dns: should check RDATA length against buffer length
1977 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1978 confronted with A/AAAA records of invalid size
1979 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1980 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1981 [19853] stdio: Printing IBM long double in decimal with high precision is
1983 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1984 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1985 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1987 [19865] network: Assertion failure or memory leak in
1988 _nss_dns_getcanonname_r
1989 [19868] network: nss_dns: netent code does not skip over non-PTR records
1990 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1992 [19881] string: Improve x86-64 memset
1993 [19907] string: Incorrect memcpy tests
1994 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1996 [19925] libc: termios.h XCASE namespace
1997 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1998 [19929] libc: limits.h NL_NMAX namespace
1999 [19931] stdio: Memory leak in vfprintf
2000 [19957] libc: clone(CLONE_VM) access invalid parent memory
2001 [19963] localedata: en_IL: New locale
2002 [19989] stdio: stdio.h cuserid namespace
2003 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2005 [19996] locale: langinfo.h nl_langinfo_l namespace
2006 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2007 [20010] network: getaddrinfo: Stack overflow in hostent translation
2009 [20012] stdio: libio: fmemopen append mode failure
2010 [20014] stdio: stdio.h namespace for pre-threads POSIX
2011 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2012 [20023] libc: fcntl.h timespec namespace
2013 [20024] math: [x86_64] vectorized sincos trashes the stack
2014 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2015 [20041] time: sys/time.h timespec namespace
2016 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2017 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2018 [20051] libc: ttyslot in wrong header under wrong conditions
2019 [20054] libc: gethostname not declared for XPG4
2020 [20055] libc: termios.h missing tcgetsid for XPG4
2021 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2023 [20073] libc: sys/stat.h fchmod namespace
2024 [20074] libc: stdlib.h rand_r namespace
2025 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2026 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2028 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2029 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2031 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2032 [20119] libc: Wrong mask for processors level type from CPUID
2033 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2034 AS not supporting AVX512
2035 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2037 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2038 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2040 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2041 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2042 [20178] libc: posix_spawn{p} should not call exit
2043 [20191] stdio: libio: vtables hardening
2044 [20195] string: FMA4 detection requires CPUID execution with register
2046 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2047 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2049 [20212] math: acos (sNaN) returns sNaN
2050 [20213] math: asin (sNaN) returns sNaN
2051 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2052 [20218] math: [i386] asinhl (sNaN) returns sNaN
2053 [20219] math: [i386] atanhl (sNaN) returns sNaN
2054 [20222] stdio: fopencookie: Mangle function pointers
2055 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2056 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2057 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2059 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2060 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2061 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2062 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2063 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2064 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2065 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2066 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2067 [20240] math: modf (sNaN) returns sNaN
2068 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2069 [20250] math: frexp (sNaN) returns sNaN
2070 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2071 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2073 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2074 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2075 subscript is below array bounds [-Werror=array-bounds]
2076 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2077 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2079 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2080 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2082 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2083 "invalid" exceptions
2084 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2085 [20316] localedata: id_ID: Februari instead of Pebruari
2086 [20327] string: POWER8 strcasecmp returns incorrect result
2087 [20347] math: Failure: Test: j0_downward (0xap+0)
2088 [20348] libc: FAIL: misc/tst-preadvwritev64
2089 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2090 [20350] libc: There is no test for p{read,write}64
2091 [20357] math: Incorrect cos result for 1.5174239687223976
2092 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2096 * Unicode 8.0.0 Support: Character encoding, character type info, and
2097 transliteration tables are all updated to Unicode 8.0.0, using new
2098 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2099 These updates cause user visible changes, such as the fixes for bugs
2100 89, 16061, and 18568.
2102 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2103 kernel-internal CPU set size. This means that requests that change the
2104 CPU affinity which failed before (for example, an all-ones CPU mask) will
2105 now succeed. Applications that need to determine the effective CPU
2106 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2107 setting it because the kernel can adjust it (and the previous size check
2108 would not detect this in the majority of cases).
2110 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2111 the following new symbols are used: fts64_children, fts64_close,
2112 fts64_open, fts64_read and fts64_set.
2114 * getaddrinfo now detects certain invalid responses on an internal netlink
2115 socket. If such responses are received, an affected process will
2116 terminate with an error message of "Unexpected error <number> on netlink
2117 descriptor <number>" or "Unexpected netlink response of size <number> on
2118 descriptor <number>". The most likely cause for these errors is a
2119 multi-threaded application which erroneously closes and reuses the netlink
2120 file descriptor while it is used by getaddrinfo.
2122 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2123 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2124 unnecessary serialization of memory allocation requests across threads.
2125 The defect is now corrected. Users should see a substantial increase in
2126 the concurent throughput of allocation requests for applications which
2127 trigger this bug. Affected applications typically create and
2128 destroy threads frequently. (Bug 19048 was reported and analyzed by
2131 * There is now a --disable-timezone-tools configure option for disabling the
2132 building and installing of the timezone related utilities (zic, zdump, and
2133 tzselect). This is useful for people who build the timezone data and code
2134 independent of the GNU C Library.
2136 * The obsolete header <regexp.h> has been removed. Programs that require
2137 this header must be updated to use <regex.h> instead.
2139 * The obsolete functions bdflush, create_module, get_kernel_syms,
2140 query_module and uselib are no longer available to newly linked binaries;
2141 the header <sys/kdaemon.h> has been removed. These functions and header
2142 were specific to systems using the Linux kernel and could not usefully be
2143 used with the GNU C Library on systems with version 2.6 or later of the
2146 * Optimized string, wcsmbs and memory functions for IBM z13.
2147 Implemented by Stefan Liebler.
2149 * Newly linked programs that define a variable called signgam will no longer
2150 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2151 require signgam to be set by those functions must ensure that they use the
2152 variable provided by the GNU C Library and declared in <math.h>, without
2153 defining their own copy.
2155 * The minimum GCC version that can be used to build this version of the GNU
2156 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2157 still be used to compile programs using the GNU C Library.
2159 Security related changes:
2161 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2162 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2164 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2165 the pointer guard feature. It is always enabled. Previously,
2166 LD_POINTER_GUARD could be used to disable security hardening in binaries
2167 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2170 * An integer overflow in hcreate and hcreate_r could lead to an
2171 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2173 * The catopen function no longer has unbounded stack usage. Reported by
2174 Max. (CVE-2015-8779)
2176 * The nan, nanf and nanl functions no longer have unbounded stack usage
2177 depending on the length of the string passed as an argument to the
2178 functions. Reported by Joseph Myers. (CVE-2014-9761)
2180 * A stack-based buffer overflow was found in libresolv when invoked from
2181 libnss_dns, allowing specially crafted DNS responses to seize control
2182 of execution flow in the DNS client. The buffer overflow occurs in
2183 the functions send_dg (send datagram) and send_vc (send TCP) for the
2184 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2185 family. The use of AF_UNSPEC triggers the low-level resolver code to
2186 send out two parallel queries for A and AAAA. A mismanagement of the
2187 buffers used for those queries could result in the response of a query
2188 writing beyond the alloca allocated buffer created by
2189 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2190 the overflow. Thanks to the Google Security Team and Red Hat for
2191 reporting the security impact of this issue, and Robert Holiday of
2192 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2194 The following bugs are resolved with this release:
2196 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2197 [887] math: Math library function "logb" and "nextafter" inconsistent
2198 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2199 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2200 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2201 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2203 [4404] localedata: German translation of "Alarm clock" is misleading
2204 [6799] math: nextafter() and nexttoward() doen't set errno on
2205 overflow/underflow errors
2206 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2208 [10432] nis: _nss_nis_setnetgrent assertion failure
2209 [11460] libc: fts has no LFS support
2210 [12926] network: getaddrinfo()/make_request() may spin forever
2211 [13065] nptl: Race condition in pthread barriers
2212 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2213 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2215 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2216 [14912] libc: Rename non-installed bits/*.h headers
2217 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2218 [15367] math: Let gcc use __builtin_isinf
2219 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2220 [15421] math: lgamma wrongly sets signgam for ISO C
2221 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2222 FE_INVALID with argument out of range
2223 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2225 [15786] dynamic-link: ifunc resolver functions can smash function
2227 [15918] math: Unnecessary check for equality in hypotf()
2228 [16061] localedata: Review / update transliteration data
2229 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2231 [16141] time: strptime %z offset restriction
2232 [16171] math: drem should be alias of remainder
2233 [16296] math: fegetround is pure?
2234 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2235 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2237 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2238 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2240 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2241 "Exception "Inexact" set" on ppc32
2242 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2243 [16517] math: Missing underflow exception from tanf/tan/tanl
2244 [16519] math: Missing underflow exception from sinhf
2245 [16520] math: Missing underflow exception from tanhf
2246 [16521] math: Missing underflow exception from exp2
2247 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2249 [16734] stdio: fopen calls mmap to allocate its buffer
2250 [16961] math: nan function incorrect handling of bad sequences
2251 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2252 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2253 [16985] locale: localedef: confusing error message when opening output
2255 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2256 [17197] locale: Redundant shift character in iconv conversion output at
2258 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2259 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2260 [17250] dynamic-link: static linking breaks nss loading
2261 (getaddrinfo/getpwnam/etc...)
2262 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2264 [17441] math: isnan() should use __builtin_isnan() in GCC
2265 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2266 (related to lock elision)
2267 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2268 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2269 [17887] time: strptime should be able to parse "+01:00" style timezones
2270 [17905] libc: catopen() Multiple unbounded stack allocations
2272 [18084] libc: backtrace (..., 0) dumps core on x86
2273 [18086] libc: nice() sets errno to 0 on success
2274 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2275 is too large (CVE-2015-8778)
2276 [18251] dynamic-link: SONAME missing when audit modules provides path
2277 [18265] libc: add attributes for wchar string and memory functions
2278 [18370] math: csqrt missing underflows
2279 [18421] libc: [hppa] read-only segment has dynamic relocations
2280 [18472] libc: Obsolete syscall wrappers should be compat symbols
2281 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2282 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2283 [18525] localedata: Remove locale timezone information
2284 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2285 [18568] localedata: Update locale data to Unicode 8.0
2286 [18589] locale: sort-test.sh fails at random
2287 [18595] math: ctan, ctanh missing underflows
2288 [18604] libc: assert macro-expands its argument
2289 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2290 contains a vector instruction exception.
2291 [18611] math: j1, jn missing errno setting on underflow
2292 [18618] localedata: sync Chechen locale definitions with other *_RU
2294 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2295 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2296 [18665] network: In send_dg, the recvfrom function is NOT always using the
2297 buffer size of a newly created buffer (CVE-2015-7547)
2298 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2300 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2302 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2303 [18699] math: tilegx cproj() for various complex infinities does not yield
2305 [18724] libc: Harden put*ent functions against data injection
2306 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2307 [18755] build: build errors with -DNDEBUG
2308 [18757] stdio: fmemopen fails to set errno on failure
2309 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2311 [18781] libc: openat64 lacks O_LARGEFILE
2312 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2313 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2314 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2315 [18790] math: [ldbl-128ibm] tanhl inaccurate
2316 [18795] libc: stpncpy fortification misses buffer lengths that are
2317 statically too large
2318 [18796] build: build fails for --disable-mathvec
2319 [18803] math: hypot missing underflows
2320 [18820] stdio: fmemopen may leak memory on failure
2321 [18823] math: csqrt spurious underflows
2322 [18824] math: fma spurious underflows
2323 [18825] math: pow missing underflows
2324 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2325 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2326 [18870] build: sem_open.c fails to compile with missing symbol
2328 [18872] stdio: Fix memory leak in printf_positional
2329 [18873] libc: posix_fallocate overflow check ineffective
2330 [18875] math: Excess precision leads incorrect libm
2331 [18877] libc: arm: mmap offset regression
2332 [18887] libc: memory corruption when using getmntent on blank lines
2333 [18918] localedata: hu_HU: change time to HH:MM:SS format
2334 [18921] libc: Regression: extraneous stat() and fstat() performed by
2336 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2337 binaries (CVE-2015-8777)
2338 [18951] math: tgamma missing underflows
2339 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2341 [18953] localedata: lt_LT: change currency symbol to the euro
2342 [18956] math: powf inaccuracy
2343 [18961] math: [i386] exp missing underflows
2344 [18966] math: [i386] exp10 missing underflows
2345 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2346 [18969] build: multiple string test failures due to missing locale
2348 [18970] libc: Reference of pthread_setcancelstate in libc.a
2349 [18977] math: float / long double Bessel functions not in XSI POSIX
2350 [18980] math: i386 libm functions return with excess range and precision
2351 [18981] math: i386 scalb*, ldexp return with excess range and precision
2352 [18982] stdio: va_list and vprintf
2353 [18985] time: Passing out of range data to strftime() causes a segfault
2355 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2356 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2357 [19012] locale: iconv_open leaks memory on error path
2358 [19016] math: clog, clog10 inaccuracy
2359 [19018] nptl: Mangle function pointers in tls_dtor_list
2360 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2361 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2362 [19048] malloc: malloc: arena free list can become cyclic, increasing
2364 [19049] math: [powerpc] erfc incorrect zero sign
2365 [19050] math: [powerpc] log* incorrect zero sign
2366 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2367 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2368 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2369 [19074] network: Data race in _res_hconf_reorder_addrs
2370 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2371 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2372 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2374 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2376 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2377 [19086] manual: posix_fallocate64 documented argument order is wrong.
2378 [19088] math: lround, llround missing exceptions close to overflow
2380 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2381 [19095] math: dbl-64 lrint incorrect for 64-bit long
2382 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2383 [19124] dynamic-link: ld.so failed to build with older assmebler
2384 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2385 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2386 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2387 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2388 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2389 pthread_setaffinity_np
2390 [19156] math: [ldbl-128] j0l spurious underflows
2391 [19164] nptl: tst-getcpu fails with many possible CPUs
2392 [19168] math: math/test-ildoubl and math/test-ldouble failure
2393 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2394 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2396 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2397 (FE_NOMASK_ENV) do not clear SSE exceptions
2398 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2399 _int_new_arena/reused_arena
2400 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2401 [19201] math: dbl-64 remainder incorrect sign of zero result
2402 [19205] math: bits/math-finite.h conditions do not match math.h and
2404 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2405 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2406 for C99-based standards
2407 [19212] libc: features.h not -Wundef clean
2408 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2410 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2411 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2412 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2414 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2415 "inexact" exceptions
2416 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2418 [19242] libc: strtol incorrect in Turkish locales
2419 [19243] malloc: reused_arena can pick an arena on the free list, leading
2420 to an assertion failure and reference count corruption
2421 [19253] time: tzset() ineffective when temporary TZ did not include DST
2423 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2424 [19270] math: [hppa] Shared libm missing __isnanl
2425 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2426 MAP_HUGETLB and MAP_STACK defines
2427 [19313] nptl: Wrong __cpu_mask for x32
2428 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2429 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2430 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2431 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2432 [19363] time: x32: times() return value wrongly truncates/sign extends
2434 [19367] dynamic-link: Improve branch prediction on Silvermont
2435 [19369] network: Default domain name not reset by res_ninit when "search"
2436 / "domain" entry is removed from resolv.conf
2437 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2439 [19385] localedata: bg_BG: time separator should be colon, not comma
2440 [19408] libc: linux personality syscall wrapper may erroneously return an
2441 error on 32-bit architectures
2442 [19415] libc: dladdr returns wrong names on hppa
2443 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2444 IBM905, IBM907, and IBM909
2445 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2446 [19443] build: build failures with -DDEBUG
2447 [19451] build: Make check fails on test-double-vlen2
2448 [19462] libc: Glibc failed to build with -Os
2449 [19465] math: Wrong code with -Os
2450 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2452 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2454 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2455 [19486] math: S390: Math tests fail with "Exception Inexact set".
2456 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2457 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2459 [19590] math: Fail to build shared objects that use libmvec.so functions.
2463 * The following bugs are resolved with this release:
2465 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2466 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2467 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2468 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2469 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2470 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2471 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2472 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2473 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2474 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2475 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2476 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2477 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2478 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2479 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2480 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2481 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2482 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2483 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2484 18657, 18676, 18694, 18696, 18887.
2486 * Cache information can be queried via sysconf() function on s390 e.g. with
2487 _SC_LEVEL1_ICACHE_SIZE as argument.
2489 * A buffer overflow in gethostbyname_r and related functions performing DNS
2490 requests has been fixed. If the NSS functions were called with a
2491 misaligned buffer, the buffer length change due to pointer alignment was
2492 not taken into account. This could result in application crashes or,
2493 potentially arbitrary code execution, using crafted, but syntactically
2494 valid DNS responses. (CVE-2015-1781)
2496 * The time zone file parser has been made more robust against crafted time
2497 zone files, avoiding heap buffer overflows related to the processing of
2498 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2499 large time zone data files. Overly long time zone specifiers in the TZ
2500 variable no longer result in stack overflows and crashes.
2502 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2503 for LD and GD on x86 and x86-64, has been implemented. You will need
2504 binutils-2.24 or later to enable this optimization.
2506 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2507 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2508 Hat). These updates cause user visible changes, such as the fix for bug
2511 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2512 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2513 condition in some applications.
2515 * Added vector math library named libmvec with the following vectorized x86_64
2516 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2518 The library can be disabled with --disable-mathvec. Use of the functions is
2519 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2520 Shared library libmvec.so is linked in as needed when using -lm (no need to
2521 specify -lmvec explicitly for not static builds).
2522 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2524 * A new fmemopen implementation has been added with the goal of POSIX
2525 compliance. The new implementation fixes the following long-standing
2526 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2527 old implementation is still present for use be by existing binaries.
2529 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2530 releases. It has been fixed to match 2.19 and older, but binaries built
2531 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2533 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2534 Contributed by Roland McGrath (Google).
2536 * The header <regexp.h> is deprecated, and will be removed in a future
2537 release. Use of this header will trigger a deprecation warning.
2538 Application developers should update their code to use <regex.h> instead.
2540 This header was formerly part of SUS, but was deprecated in 1994 and
2541 removed from the standard in 2001. Also, the glibc implementation
2542 leaks memory. See BZ#18681 for more details.
2546 * The following bugs are resolved with this release:
2548 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2549 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2550 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2551 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2552 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2553 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2554 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2555 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2556 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2557 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2560 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2561 too little memory for the to-be-scanned arguments and overflow the
2562 allocated buffer. The implementation now correctly computes the required
2563 buffer size when using malloc, and switches to malloc from alloca as
2566 * A new semaphore algorithm has been implemented in generic C code for all
2567 machines. Previous custom assembly implementations of semaphore were
2568 difficult to reason about or ensure that they were safe. The new version
2569 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2570 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2571 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2573 * Port to Altera Nios II has been contributed by Mentor Graphics.
2575 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2576 implementations for powerpc64/powerpc64le.
2577 Implemented by Adhemerval Zanella (IBM).
2579 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2580 and powerpc64le. This may improve lock scaling of existing programs on
2581 HTM capable systems. The lock elision code is only enabled with
2582 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2583 powerpc will issue a transaction abort on every syscall to avoid side
2584 effects being visible outside transactions.
2586 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2587 AArch64. Contributed by ARM Ltd.
2589 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2591 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2592 under certain input conditions resulting in the execution of a shell for
2593 command substitution when the application did not request it. The
2594 implementation now checks WRDE_NOCMD immediately before executing the
2595 shell and returns the error WRDE_CMDSUB as expected.
2597 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2598 processing format strings with a large number of format specifiers.
2600 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2601 infinite loop if the DNS response contained a PTR record of an unexpected
2604 * The minimum GCC version that can be used to build this version of the GNU
2605 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2606 still be used to compile programs using the GNU C Library.
2608 * The GNU C Library is now built with -Werror by default. This can be
2609 disabled by configuring with --disable-werror.
2611 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2613 * The obsolete sigvec function has been removed. This was the original
2614 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2615 programs have been using instead for about 25 years. Of course, ABI
2616 compatibility for old binaries using sigvec remains intact.
2618 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2619 with newer versions of bison.
2621 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2622 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2623 registers overlay two consecutive single-precision registers. MIPS32R2
2624 introduced a new FPU mode (FR=1) where double-precision registers extend the
2625 corresponding single-precision registers which is incompatible with the
2626 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2627 require the use of FR=1 making a transition necessary. New o32 ABI
2628 extensions enable users to migrate over time from the original o32 ABI
2629 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2630 tracks the ABI of any loaded object and verifies that new objects are
2631 compatible. Mode transitions will also be requested as required and
2632 unsupportable objects will be rejected. The ABI checks include both soft and
2633 hard float ABIs for o32, n32 and n64.
2635 GCC 5 with GNU binutils 2.25 onwards:
2636 It is strongly recommended that all o32 system libraries are built using the
2637 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2638 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2639 compiler using --with-fp-32=xx to set this by default.
2643 * The following bugs are resolved with this release:
2645 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2646 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2647 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2648 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2649 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2650 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2651 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2652 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2653 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2654 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2655 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2656 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2657 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2658 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2659 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2661 * Reverted change of ABI data structures for s390 and s390x:
2662 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2663 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2664 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2665 are preserved pointing straight to the same implementation as the old ones.
2666 Given that, new callers will simply provide a too-big buffer to these
2667 functions. Any applications/libraries out there that embed jmp_buf or
2668 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2669 against 2.19 headers will have to rebuilt again. This is necessary in any
2670 case to revert the breakage in their ABI caused by the glibc change.
2672 * Support for file description locks is added to systems running the
2673 Linux kernel. The standard file locking interfaces are extended to
2674 operate on file descriptions, not file descriptors, via the use of
2675 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2676 are associated with an open file instead of a process.
2678 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2680 * The minimum Linux kernel version that this version of the GNU C Library
2681 can be used with is 2.6.32.
2683 * Running the testsuite no longer terminates as soon as a test fails.
2684 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2685 with PASS or FAIL lines for individual tests. A summary of the results is
2686 printed, including a list of failing lists, and "make check" exits with
2687 error status if there were any unexpected failures. "make check
2688 stop-on-test-failure=y" may be used to keep the old behavior.
2690 * The am33 port, which had not worked for several years, has been removed
2693 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2694 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2695 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2696 conflicted with POSIX (support for which was removed in 2.19), the
2697 interfaces those macros enabled remain available when compiling with
2698 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2699 test macros defined.
2701 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2703 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2704 This may improve lock scaling of existing programs on TX capable systems.
2705 The lock elision code is only built with --enable-lock-elision=yes and
2706 then requires a GCC version supporting the TX builtins. With lock elision
2707 default mutexes are elided via __builtin_tbegin, if the cpu supports
2708 transactions. By default lock elision is not enabled and the elision code
2711 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2712 copy the path argument. This allowed programs to cause posix_spawn to
2713 deference a dangling pointer, or use an unexpected pathname argument if
2714 the string was modified after the posix_spawn_file_actions_addopen
2717 * All supported architectures now use the main glibc sysdeps directory
2718 instead of some being in a separate "ports" directory (which was
2719 distributed separately before glibc 2.17).
2721 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2722 On configurations that support it (all Linux configurations), it's now
2723 used regardless of the --enable-add-ons switch to configure. It is no
2724 longer possible to build such configurations without pthreads support.
2726 * Locale names, including those obtained from environment variables (LANG
2727 and the LC_* variables), are more tightly checked for proper syntax.
2728 setlocale will now fail (with EINVAL) for locale names that are overly
2729 long, contain slashes without starting with a slash, or contain ".." path
2730 components. (CVE-2014-0475) Previously, some valid locale names were
2731 silently replaced with the "C" locale when running in AT_SECURE mode
2732 (e.g., in a SUID program). This is no longer necessary because of the
2735 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2736 with application code using Intel MPX instructions. (With all previous
2737 versions, the MPX register state could be clobbered when making calls
2738 into or out of a shared library.) Note that while the new dynamic
2739 linker is compatible with all known x86 hardware whether or not it
2740 supports Intel MPX, some x86 instruction-set emulators might fail to
2741 handle the new instruction encodings. This is known to affect Valgrind
2742 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2743 release), and might affect other tools that do instruction emulation.
2745 * Support for loadable gconv transliteration modules has been removed.
2746 The support for transliteration modules has been non-functional for
2747 over a decade, and the removal is prompted by security defects. The
2748 normal gconv conversion modules are still supported. Transliteration
2749 with //TRANSLIT is still possible, and the //IGNORE specifier
2750 continues to be supported. (CVE-2014-5119)
2752 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2753 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2754 resulting a denial-of-service security vulnerability in applications which
2755 use functions related to iconv. (CVE-2014-6040)
2759 * The following bugs are resolved with this release:
2761 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2762 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2763 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2764 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2765 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2766 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2767 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2768 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2769 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2770 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2771 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2772 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2773 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2774 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2775 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2776 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2777 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2778 16431, 16453, 16474, 16506, 16510, 16529
2780 * Slovenian translations for glibc messages have been contributed by the
2781 Translation Project's Slovenian team of translators.
2783 * The public headers no longer use __unused nor __block. This change is to
2784 support compiling programs that are derived from BSD sources and use
2785 __unused internally, and to support compiling with Clang's -fblock
2786 extension which uses __block.
2788 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2789 large collation sequences to optimize multiple passes. This cache
2790 computation may overflow for large collation sequences and may cause a
2791 stack or buffer overflow. This is now fixed to use a slower algorithm
2792 which does not use a cache if there is an integer overflow.
2794 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2795 rules for large collation sequences to optimize multiple passes and falls
2796 back to alloca if malloc fails, resulting in a possible stack overflow.
2797 The implementation now falls back to an uncached collation sequence lookup
2800 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2801 initialized for static applications resulting in the security feature
2802 being disabled. The pointer guard is now correctly initialized to a
2803 random value for static applications. Existing static applications need
2804 to be recompiled to take advantage of the fix (bug 15754).
2806 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2807 to the d_name member of struct dirent, or omit the terminating NUL
2808 character. (Bugzilla #14699).
2810 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2811 aligned_alloc functions could allocate too few bytes or corrupt the
2812 heap when passed very large allocation size values (Bugzilla #15855,
2815 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2816 for AF_INET6 has been fixed (Bugzilla #16072).
2818 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2819 pap_CW, quz_PE, the_NP.
2821 * Substantially revised locales: gd_GB, ht_HT
2823 * The LC_ADDRESS field was updated to support country_car for almost all
2826 * ISO 1427 definitions were updated.
2828 * ISO 3166 definitions were updated.
2830 * The localedef utility now supports --big-endian and --little-endian
2831 command-line options to generate locales for a different system from that
2832 for which the C library was built.
2834 * Binary locale files now only depend on the endianness of the system for
2835 which they are generated and not on other properties of that system. As a
2836 consequence, binary files generated with new localedef may be incompatible
2837 with old versions of the GNU C Library, and binary files generated with
2838 old localedef may be incompatible with this version of the GNU C Library,
2839 in the following circumstances:
2841 + Locale files may be incompatible on m68k systems.
2843 + Locale archive files (but not separate files for individual locales) may
2844 be incompatible on systems where plain "char" is signed.
2846 * The configure option --disable-versioning has been removed. Builds with
2847 --disable-versioning had not worked for several years.
2849 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2850 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2852 * SystemTap probes for malloc have been introduced.
2854 * SystemTap probes for slow multiple precision fallback paths of
2855 transcendental functions have been introduced.
2857 * Support for powerpc64le has been added.
2859 * The soft-float powerpc port now supports e500 processors.
2861 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2863 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2864 set of header declarations that are enabled by default, even when other
2865 feature test macros or compiler options such as -std=c99 would otherwise
2866 disable some of those declarations.
2868 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2869 conflict with POSIX. The libbsd-compat library (which was a dummy library
2870 that did nothing) has also been removed.
2872 * Preliminary documentation about Multi-Thread, Async-Signal and
2873 Async-Cancel Safety has been added.
2875 * Change of ABI data structures for s390 and s390x:
2876 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2877 allow for future hardware extensions. All library functions that accept or
2878 return these structures were versioned in order to provide backwards
2879 compatibility with existing code. However, not all cases can be handled
2880 with interface versioning. If an application embeds either structure into
2881 another structure and that structure is passed to another compilation unit
2882 which expects the newer larger-sized object then any access to the new
2883 fields will result in undefined behaviour. Similarly any access to fields
2884 that were moved by the enlarging of the structures will cause undefined
2885 behaviour. To fix the undefined behaviour all objects that are part of the
2886 final in-memory image must agree on the size of structures, and this may
2887 require recompilation.
2891 * The following bugs are resolved with this release:
2893 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2894 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2895 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2896 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2897 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2898 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2899 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2900 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2901 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2902 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2903 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2904 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2905 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2908 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2909 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2910 Distributions can re-enable building and using pt_chown via the new configure
2911 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2912 considerable security risks and should only be used if the distribution
2913 understands and accepts the risks.
2915 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2918 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2919 fixed (Bugzilla #15330).
2921 * Add support for calling C++11 thread_local object destructors on thread
2922 and program exit. This needs compiler support for offloading C++11
2923 destructor calls to glibc.
2925 * Improved worst case performance of libm functions with double inputs and
2928 * Support for priority inherited mutexes in pthread condition variables on
2929 non-x86 architectures.
2931 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2933 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2935 * Optimized string functions for ARM. Implemented by Will Newton and
2938 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2940 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2943 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2944 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2946 * Added a benchmark framework to track performance of functions in glibc.
2948 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2949 It is based on draft TS 18661 and currently enabled as a GNU extension.
2951 * On Linux, the clock function now uses the clock_gettime system call
2952 for improved precision, rather than old times system call.
2954 * Added support for version-3 tz format files. This is needed when using
2955 the tz database release 2013e or later, and affects a few unusual cases --
2956 currently only TZ='America/Godthab' for time stamps after 2037.
2958 * Added new API functions pthread_getattr_default_np and
2959 pthread_setattr_default_np to get and set the default pthread
2960 attributes of a process.
2962 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2963 This may improve lock scaling of existing programs on TSX capable systems.
2964 When the --enable-lock-elision=yes parameter is specified at configure
2965 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2968 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2969 availability). Implemented by Ryan S. Arnold.
2971 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2975 * The following bugs are resolved with this release:
2977 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2978 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2979 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2980 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2981 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2982 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2983 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2984 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2985 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2986 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2987 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2988 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2989 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2990 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2991 14879, 14889, 14893, 14898, 14914.
2993 * Optimization of memcpy for MIPS.
2995 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2996 EMFILE has been fixed (Bugzilla #14889).
2998 * The add-on ports collection is now distributed in the "ports" subdirectory
2999 of the main GNU C Library distribution, rather than separately.
3001 * Port to ARM AArch64 contributed by Linaro.
3003 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3004 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3006 Implemented by Andreas Krebbel.
3008 * The new function secure_getenv allows secure access to the environment,
3009 returning NULL if running in a SUID/SGID process. This function replaces
3010 the internal function __secure_getenv.
3012 * SystemTap static probes have been added into the dynamic linker.
3013 Implemented by Gary Benson.
3015 * Optimizations of string functions strstr, strcasestr and memmem.
3016 Implemented by Maxim Kuvyrkov.
3018 * The minimum Linux kernel version that this version of the GNU C Library
3019 can be used with is 2.6.16.
3021 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3022 powerpc POWER7. Implemented by Will Schmidt.
3024 * New configure option --disable-nscd builds the C library such that it
3025 never attempts to contact the Name Service Caching Daemon (nscd).
3026 New configure option --disable-build-nscd avoids building nscd itself;
3027 this is the default if --disable-nscd is used.
3029 * Improved support for cross-compilation, including cross-testing and
3030 bootstrap builds without a previously built glibc.
3032 * Several testsuite tests are now able to test multiple IFUNC variants of an
3033 interface, rather than just testing the one that would be chooen by
3036 * New configure options --with-bugurl and --with-pkgversion, for
3037 distributors to use to embed their bug-reporting and package version
3038 information in --help and --version output.
3040 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3041 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3042 allows creation of chroots without the procfs mounted on /proc.
3044 * The `crypt' function now fails if passed salt bytes that violate the
3045 specification for those values. On Linux, the `crypt' function will
3046 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3047 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3048 when the mode is enabled.
3050 * The `clock_*' suite of functions (declared in <time.h>) is now available
3051 directly in the main C library. Previously it was necessary to link with
3052 -lrt to use these functions. This change has the effect that a
3053 single-threaded program that uses a function such as `clock_gettime' (and
3054 is not linked with -lrt) will no longer implicitly load the pthreads
3055 library at runtime and so will not suffer the overheads associated with
3056 multi-thread support in other code such as the C++ runtime library.
3058 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3063 * The following bugs are resolved with this release:
3065 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3066 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3067 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3068 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3069 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3070 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3071 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3072 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3073 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3074 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3075 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3076 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3077 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3078 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3079 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3080 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3081 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3082 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3083 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3084 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3085 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3086 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3089 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3090 configuring glibc with:
3091 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3092 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3093 Implemented by H.J. Lu.
3097 + define static_assert
3099 + do not declare gets
3101 + declare at_quick_exit and quick_exit also for ISO C11
3103 + aligned_alloc. NB: The code is deliberately allows the size parameter
3104 to not be a multiple of the alignment. This is a moronic requirement
3105 in the standard but it is only a requirement on the caller, not the
3108 + timespec_get added
3110 + uchar.h support added
3112 + CMPLX, CMPLXF, CMPLXL added
3114 Implemented by Ulrich Drepper.
3116 * Support for the IA-64 has been moved to ports.
3118 * Remove support for anything but ELF binary format
3120 * Checking versions of poll, ppoll added.
3121 Implemented by Ulrich Drepper.
3123 * More generic and 64-bit performance optimizations to math functions.
3124 Implemented by Ulrich Drepper.
3126 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3127 headers and functions available at compile time as they were before
3128 version 2.14. This option will be removed at some time in the future
3129 after the TI-RPC library becomes fully sufficient for the needs of
3130 existing applications.
3132 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3133 Note that glibc is not expected to work with any Linux kernel version
3136 * New header <sys/auxv.h> and function getauxval allowing easy access to
3137 the AT_* key-value pairs passed from the Linux kernel. The header also
3138 defines the HWCAP_* bits associated with the AT_HWCAP key.
3140 * New locales: mag_IN
3142 * New configure option --enable-systemtap builds SystemTap static probes
3143 into libc for setjmp and longjmp and into libpthread for various operations.
3144 So far the setjmp/longjmp probes and some of the libpthread probes are
3145 provided only for i*86 and x86_64.
3146 Implemented by Roland McGrath and Rayson Ho.
3148 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3150 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3153 * More optimized functions for SPARC. Implemented by David S. Miller.
3155 * Improved support for cross-compilation, especially bootstrap builds
3156 without a previously built glibc.
3158 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3159 Chris Metcalf from Tilera.
3161 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3162 now supported for ARM processors.
3164 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3165 as the name of the dynamic linker, to distinguish it from the
3166 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3168 * Support for CRIS has been removed from ports.
3170 * A new class of installed header has been documented for low-level
3171 platform-specific functionality. PowerPC added the first instance with a
3172 function to provide time base register access. Contributed by Tulio
3173 Magno Quites Machado Filho.
3175 * ix86 configurations now install header files that are consistent with
3176 what x86-64 configurations install. These same header files can be used
3177 for -m32, -m64, or -mx32 builds.
3178 Contributed by H.J. Lu.
3180 * Math library bug fixes. A thorough audit of all open math library bugs was
3181 conducted by Joseph Myers. Significant progress was made on many math
3182 library bugs resulting in more accurate exceptions and function results.
3183 Many thanks to all those that contributed including Andreas Jaeger for his
3184 patch review and work on the x87 trigonometric instruction issues.
3186 * Timezone data is no longer installed. Timezone-related binaries and scripts
3187 will continue to be installed. Users should obtain their timezone data from
3188 their distribution provider or from the tzdata package at
3189 <ftp://munnari.oz.au/pub/>.
3193 * The following bugs are resolved with this release:
3195 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3196 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3197 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3198 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3199 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3200 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3201 13484, 13506, 13515, 13523, 13524, 13538, 13540
3203 * New program pldd to list loaded object of a process
3204 Implemented by Ulrich Drepper.
3206 * Add nss_db support back to glibc. No more dependency on Berkeley db
3207 and support for initgroups lookups.
3208 Implemented by Ulrich Drepper.
3210 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3211 Contributed by HJ Lu.
3213 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3214 Contributed by HJ Lu.
3216 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3217 on x86-32 and x86-64.
3218 Contributed by Liubov Dmitrieva.
3220 * Optimized strchr and strrchr for SSE on x86-32.
3221 Contributed by Liubov Dmitrieva.
3223 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3224 for x86-64 and x86-32.
3225 Contributed by Liubov Dmitrieva.
3227 * New interfaces: scandirat, scandirat64
3228 Implemented by Ulrich Drepper.
3230 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3231 Implemented by Ulrich Drepper.
3233 * nscd now also caches the netgroup database.
3234 Implemented by Ulrich Drepper.
3236 * Integrate libm with gcc's -ffinite-math-only option.
3237 Implemented by Ulrich Drepper.
3239 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3240 to math functions. Implemented by Ulrich Drepper.
3242 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3243 Implemented by Ulrich Drepper.
3245 * New Linux interfaces: process_vm_readv, process_vm_writev
3247 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3248 Implemented by Ulrich Drepper.
3250 * Optimized nearbyint and strcasecmp for PPC.
3251 Implemented by Adhemerval Zanella.
3253 * New locales: bho_IN, unm_US, es_CU, ta_LK
3257 * The following bugs are resolved with this release:
3259 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3260 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3261 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3262 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3263 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3264 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3265 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3266 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3267 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3268 12795, 12811, 12813, 12814, 12841
3270 * The RPC implementation in libc is obsoleted. Old programs keep working
3271 but new programs cannot be linked with the routines in libc anymore.
3272 Programs in need of RPC functionality must be linked against TI-RPC.
3273 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3275 Visible changes of this change include (obviously) the inability to link
3276 programs using RPC functions without referencing the TI-RPC library and the
3277 removal of the RPC headers from the glibc headers.
3278 Implemented by Ulrich Drepper.
3280 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3281 syncfs, setns, sendmmsg
3283 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3284 yue_HK, lij_IT, mhr_RU
3286 * New encodings: CP770, CP771, CP772, CP773, CP774
3288 * New program sotruss to trace calls through PLTs
3289 Implemented by Ulrich Drepper.
3291 * The malloc hook implementation is marked deprecated and will be removed
3292 from the default implementation in the next version. The design never
3293 worked ever since the introduction of threads. Even programs which do
3294 not create threads themselves can use multiple threads created internally.
3298 * The following bugs are resolved with this release:
3300 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3301 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3302 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3303 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3306 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3308 * POWER7 optimizations: memset, memcmp, strncmp
3310 * New optimized string functions for x86-64: strnlen (SSE2),
3311 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3312 Implemented by Ulrich Drepper.
3316 * The following bugs are resolved with this release:
3318 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3319 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3320 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3321 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3322 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3323 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3324 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3325 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3327 * New interfaces: pthread_getname_np, pthread_setname_np
3329 * New Linux interface: recvmmsg
3331 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3333 * The dynamic linker now recognizes supported ABI versions from the
3334 EI_ABIVERSION field in the ELF header.
3335 Implemented by Ulrich Drepper.
3337 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3338 will not be used to fill in password fields in the passwd.byname replies.
3339 Instead it is used to synthesize the shadow.byname table, should it be
3340 missing. This is a useful mode in some installations involving Solaris.
3341 Implemented by Ulrich Drepper.
3343 * New locales: kok_IN, sq_MK, cv_RU
3347 * The following bugs are resolved with this release:
3349 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3350 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3351 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3352 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3353 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3354 10789, 10817, 10823, 10840, 10847
3356 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3358 Implemented by Ulrich Drepper.
3360 * Checking version of longjmp added that fails if an uninitialized stack
3361 frame would be created. Implemented by Ulrich Drepper.
3363 * STT_GNU_IFUNC is now supported in static executables.
3364 Implemented by H.J. Lu.
3366 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3368 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3369 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3370 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3371 Contributed by H.J. Lu.
3373 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3374 Implemented by Ulrich Drepper.
3376 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3378 Contributed by H.J. Lu.
3380 * Support for fma instruction in AVX on x86-64.
3381 Implemented by H.J. Lu and Ulrich Drepper.
3383 * AVX support in x86-64 auditing support in ld.so.
3384 Implemented by H.J. Lu.
3386 * STB_GNU_UNIQUE support added.
3387 Implemented by Ulrich Drepper.
3389 * Implement second fallback mode for optimized DNS lookup for even more
3390 broken environments. If two requests from the same port are not
3391 handled correctly close the socket and open a new one before sending
3392 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3393 can be used to select this mode right away, instead of rediscovering the
3394 necessity is every process again.
3395 Implemented by Ulrich Drepper.
3397 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3398 Implemented by Adam Tkac.
3400 * Optimized iconv conversions for S390x.
3401 Implemented by Andreas Krebbel.
3403 * Using condvars with PI mutexes is now more efficient due to kernel
3404 support for requeueing to PI futexes. NPTL support added for x86-64.
3405 Implemented by Ulrich Drepper.
3407 * New locales: ps_AF, my_MM
3411 * The following bugs are resolved with this release:
3413 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3414 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3415 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3416 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3417 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3419 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3420 Implemented by Ulrich Drepper.
3422 * New Linux interfaces: accept4, fallocate, fallocate64.
3423 Implemented by Ulrich Drepper.
3425 * Correct declarations of string function when used in C++ code. This
3426 could lead to compile errors for invalid C++ code.
3428 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3431 * New POSIX 2008 interface: psiginfo
3432 Implemented by Ulrich Drepper.
3434 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3435 Implemented by Ulrich Drepper.
3437 * Support for selecting between multiple function definitions at runtime
3438 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3440 * The libcrypt library can now use the hash function implementations in
3441 NSS. Implemented by Ulrich Drepper.
3443 * The malloc implementation can be compiled to be less memory efficient
3444 but higher performing in multi-threaded programs.
3445 Implemented by Ulrich Drepper.
3447 * New locales: nan_TW@latin, ks_IN
3449 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3450 Implemented by Ulrich Drepper.
3452 * Extended printf hook support. It is possible to use user-defined types
3453 and extend existing format specifiers.
3454 Implemented by Ulrich Drepper.
3456 * Handling for group shadow files has been added.
3457 Implemented by Ulrich Drepper.
3459 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3460 servers (the case, e.g., for some people using the built-in DNS
3461 server in ADSL modems/routers). There is a once-per-process timeout
3462 in case of a broken server. To avoid it, users can run nscd or put
3463 'options single-request' in /etc/resolv.conf.
3464 Implemented by Ulrich Drepper.
3468 * The following bugs are resolved with this release:
3470 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3471 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3472 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3473 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3474 6968, 6974, 6980, 6995, 7008, 7009, 7029
3476 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3477 up at the same time. Implemented by Ulrich Drepper.
3479 * TLS descriptors for LD and GD on x86 and x86-64.
3480 Implemented by Alexandre Oliva.
3482 * getaddrinfo now handles DCCP and UDPlite.
3483 Implemented by Ulrich Drepper.
3485 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3486 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3487 Implemented by Ulrich Drepper.
3489 * New implementation of memmem, strstr, and strcasestr which is O(n).
3490 Implemented by Eric Blake.
3492 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3494 * Implement "e" option for popen to open file descriptor with the
3495 close-on-exec flag set. Implemented by Ulrich Drepper.
3497 * Implement "b" mode for fmemopen. In this mode writes writes don't
3498 implicitly add a NUL byte and seeks from the end of the buffer really
3499 use the buffer end, not the string length as the basis.
3500 Implemented by Ulrich Drepper.
3502 * Many functions, exported and internal, now atomically set the close-on-exec
3503 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3505 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3507 Implemented by Pravin Satpute.
3509 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3513 * The following bugs are resolved with this release:
3515 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3516 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3517 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3518 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3519 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3520 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3521 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3523 * New locales: bo_CN, bo_IN, shs_CA.
3525 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3527 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3528 Implemented by Pravin Satpute.
3530 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3532 * nscd caches now all timeouts for DNS entries
3533 Implemented by Ulrich Drepper.
3535 * nscd is more efficient and wakes up less often.
3536 Implemented by Ulrich Drepper.
3538 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3539 vdprintf, and obstack_vprintf.
3540 Implemented by Jakub Jelinek.
3542 * Faster memset for x86-64.
3543 Implemented by Harsha Jagasia and H.J. Lu.
3545 * Faster memcpy on x86.
3546 Implemented by Ulrich Drepper.
3548 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3549 Implemented by Ulrich Drepper.
3551 * Faster sqrt and sqrtf implemention for some PPC variants.
3552 Implemented by Steven Munroe.
3556 * The following bugs are resolved with this release:
3558 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3559 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3560 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3561 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3562 5071, 5103, 5104, 5112, 5113, 5184, 5186
3564 * More checking functions: fread, fread_unlocked, open*, mq_open.
3565 Implemented by Jakub Jelinek and Ulrich Drepper.
3567 * Extend fortification to C++. Implemented by Jakub Jelinek.
3569 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3570 by not recognizing 'a' as a modifier when those specs are requested.
3571 Implemented by Jakub Jelinek.
3573 * PPC optimizations to math and string functions.
3574 Implemented by Steven Munroe.
3576 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3577 options to be passed. Implemented by Ulrich Drepper.
3579 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3581 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3582 Implemented by Ulrich Drepper.
3584 * Handle private futexes in the NPTL implementation.
3585 Implemented by Jakub Jelinek and Ulrich Drepper.
3587 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3588 Implemented by Roland McGrath and Ulrich Drepper.
3590 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3592 * SHA-256 and SHA-512 based password encryption.
3593 Implemented by Ulrich Drepper.
3595 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3596 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3599 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3600 Implemented by Ulrich Drepper.
3604 * The following bugs are resolved with this release:
3606 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3607 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3608 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3609 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3610 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3611 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3612 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3615 * New Linux interfaces: epoll_pwait, sched_getcpu.
3617 * New generic interfaces: strerror_l.
3619 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3623 * The following bugs are resolved with this release:
3625 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3626 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3627 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3628 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3629 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3630 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3631 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3632 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3633 3225, 3251, 3252, 3253, 3273
3635 * For Linux, the sorting of addresses returned by getaddrinfo now also
3636 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3637 Implemented by Ulrich Drepper.
3639 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3640 Implemented by Ulrich Drepper.
3642 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3644 * New iconv module for MIK. Contributed by Alexander Shopov.
3646 * For sites with broken group and/or passwd database, the auto-propagate
3647 option of nscd can prevent creating ID lookup entries from the results
3648 of a name lookup and vice versa. This usually is no problem but some
3649 site might have problems with the default behavior.
3650 Implemented by Ulrich Drepper.
3652 * Iterating over entire database in NIS can be slow. With the
3653 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3654 to trade time for memory. The entire database will be read at once.
3655 Implemented by Ulrich Drepper.
3657 * The interfaces introduced in RFC 3542 have been implemented by
3660 * Support for the new ELF hash table format was added by Ulrich Drepper.
3662 * Support for priority inheritance mutexes added by Jakub Jelinek and
3665 * Support for priority protected mutexes added by Jakub Jelinek.
3667 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3668 or_IN, csb_PL, fy_NL, sr_ME.
3672 * More overflow detection functions.
3674 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3675 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3677 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3678 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3679 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3680 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3681 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3682 by Masahide Washizawa.
3684 * It is now possible to install audit libraries for the dynamic linker using
3685 LD_AUDIT. Implemented by Ulrich Drepper.
3687 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3688 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3689 For a libc and libpthread that works well on Linux 2.4 kernels, we
3690 recommend using the stable 2.3 branch.
3692 * The new function open_wmemstream acts like open_memstream,
3693 but uses a wchar_t wide character buffer.
3695 * The new function ppoll is an improved version of poll, similar to pselect.
3697 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3698 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3699 renameat, unlinkat, linkat, symlinkat, readlinkat.
3701 * New Linux kernel system calls: unshare,
3702 inotify_init, inotify_add_watch, inotify_rm_watch.
3704 * The euidaccess function is now also known by the alias eaccess,
3705 for compatibility with some other systems.
3707 * Timezone data updated to 2006b version.
3711 * The following bugs are resolved with this release:
3713 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3714 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3715 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3716 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3717 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3718 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3720 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3722 * As of this release, GCC 4 can be used to compile the C Library.
3724 * Timezone data updated to 2005m version.
3728 * The following bugs are resolved with this release:
3730 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3731 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3732 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3733 777, 787, 821, 822, 823, 825
3735 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3739 * Support for RFC 3678. Real implementations exist only for Linux so far.
3740 Implemented by Ulrich Drepper.
3742 * nscd can now cache entries persistently. Expiring entries are reloaded.
3743 For speedups the cache can be shared in memory with client processes.
3744 Implemented by Ulrich Drepper.
3746 * nscd can now perform SELinux checks.
3747 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3749 * getaddrinfo queries are now cached. Canonical name lookup is performed
3751 Implemented by Ulrich Drepper.
3753 * The nothrow function attribute is used when headers are used by gcc when
3754 compiling C code. This can avoid significant amounts of exception
3757 * The malloc functions perform more error checking and are stricter when
3758 it comes to reacting on errors. The default action is to terminate
3759 the process after showing an error message. Implemented by Ulrich Drepper.
3761 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3762 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3764 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3765 namespaces using the new function dlmopen(). This feature is of course,
3766 like most other dynamic loading functionality, not available in statically
3767 linked applications. Implemented by Ulrich Drepper.
3769 * Low-overhead boundary checking variants of string and some stdio functions
3770 were added. These are to be used in conjunction with a gcc patch by
3771 Jakub Jelinek which adds calls to these functions if possible.
3772 Implemented by Jakub Jelinek and Ulrich Drepper.
3774 * Old code for several operating systems and machine architectures that
3775 have not been in working condition in a long time have been removed from
3776 the main source tree maintained by the GNU C Library's maintainers.
3777 These files are now reside in the separate `ports' source module
3778 that is usable as an add-on when building the library.
3782 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3783 interrogate the dynamic linker, compatible with the Solaris interface.
3785 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3786 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3788 * getifaddrs now uses the netlink interface on Linux to get its information.
3789 Implemented by Thorsten Kukuk.
3791 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3792 Implemented by Ulrich Drepper.
3794 * support for non-executable stacks on x86 has been added. Changes mostly
3797 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3800 * getaddrinfo now performs destination address selection according to
3805 * Thread-safe interfaces for many functions that access locale data
3806 were added in version 2.3, but these features were omitted from NEWS.
3807 Many functions have variants with an `_l' suffix that take a `locale_t'
3808 object as a parameter rather than consulting the current locale.
3809 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3810 create and maintain `locale_t' objects. Additionally, the new function
3811 `uselocale' sets "the current locale" (as used by functions not so
3812 parameterized) set for an individual thread. These features were added
3813 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3815 * The functions getresuid, getresgid, setresuid, and setresgid, which
3816 have long been available on Linux, are now declared in <unistd.h>
3817 and are now also available on the Hurd.
3819 * ELF thread-local storage support (TLS) now works on x86-64.
3821 * The new dynamic string token $LIB is expanded in shared library names.
3822 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3824 * Aldy Hernandez contributed complete software floating point support for
3825 PowerPC machines with no FPU.
3827 * fexecve is implemented on Linux.
3829 * The `btowc' function should work at least twice as fast due to
3830 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3832 * With approriate thread add-ons cancelable functions are now implemented
3833 in libc.so as well. No need to call the function in libpthread. This
3834 change allowed to finally disable the incorrect and expensive handling
3835 of weak definition in ld.so.
3837 * Yet more PLT entries in libc.so have been removed. We finally arrived
3838 at the bare minimum. Startup times improved appropriately.
3840 * Support for the new Linux/x86 system call interface was added. The
3841 AT_SYSINFO auxiliary vector entry is recognized and handled.
3845 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3848 * iconv (the program and the interface) now accepts empty names (excluding
3849 options like //TRANSLIT) to mean "use charset of current locale".
3851 * localedef can now transliterate characters in strings which are not in
3852 the provided charmap. The information from the input locale is used.
3854 * Prelinking support was added for ELF targets. This requires additional
3855 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3857 * Read-only stdio streams now use mmap to speed up operation by eliminating
3858 copying and buffer underflows. To use add 'm' to the mode string of
3859 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3861 * The malloc functions were completely rewritten by Wolfram Gloger based
3862 on Doug Lea's malloc-2.7.0.c.
3864 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3865 implementation of regex.
3867 * Bruno Haible upgraded the iconv and locale implementation to support
3870 * Contents of the LC_* and LANG environment variables in the CEN style are
3871 not recognized anymore. It never was used. Change by Ulrich Drepper.
3873 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3874 thread-local storage (TLS) ABI on some platforms.
3875 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3877 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3878 EUC-JISX0213, and TSCII.
3880 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3881 BSD-compatible interface for getting all network interface addresses.
3882 Implementation for IPv4 by Roland McGrath.
3884 * Loading of locale data is faster due to the introduction of a locale
3885 archive. Implemented by Roland McGrath and Ulrich Drepper.
3887 * Startup times are significantly reduced by not using exported functions
3888 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3891 * Steven Munroe contributed a port to PowerPC64/Linux.
3895 * The Hurd now uses the GNU libio implementation of stdio rather than the
3896 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3898 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3899 with the same behavior as the Linux system call of the same name.
3903 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3904 128-bit long double format.
3906 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3907 IBM1161, and IBM1162 charsets.
3909 * Andreas Jaeger contributed a port to x86-64/Linux.
3911 * Peter Bruin contributed a port to PowerPC/Hurd.
3913 * libc functions using I/O streams now can handle wide-oriented streams
3916 * optimizations in the dynamic linker. Binaries created by recent binutils
3917 versions start up quicker due to reduced time spent on relocations.
3919 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3923 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3924 asinh, atanh, j0 for the 128-bit long double format.
3926 * Bruno Haible updated all the code handling Unicode in some form to
3927 support Unicode 3.1.
3929 * Speed of regex for single-byte locales is back to previous levels.
3930 Patch by Isamu Hasegawa.
3932 * Alpha, SPARC, and IA-64 now also using floating stacks.
3934 * Startup time of internationalized applications greatly improved through
3935 iconv cache. Use iconvconfig to generate the cache file.
3936 Contributed by Ulrich Drepper.
3938 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3939 severe performance problems. Patch by David Mosberger.
3941 * The Hurd port got a lot more functionality like AIO, various stdio
3942 extensions, etc. Mainly done by Roland McGrath.
3944 * mtrace can now lookup symbols in shared libraries.
3948 * Intel's IA-64 math library is largely integrated. It provides fast and
3949 accurate implementatations for most basic and standard math functions
3950 in float, double, and long double format.
3952 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3953 and asin for the 96-bit long double format and asin, log, tan for the
3954 128-bit long double format.
3956 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3957 The basic double functions exist today. Contributed by Abraham Ziv
3958 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3959 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3961 * An asynchronous name lookup library was added. The interface is designed
3962 after POSIX AIO. The proposal was circulated beforehand to get comments.
3963 No negative ones came in. Implemented by Ulrich Drepper.
3965 * Port to S390/64bit contributed by Martin Schwidefsky
3966 <schwidefsky@de.ibm.com>.
3968 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3969 of functions for Linux/IA-64.
3971 * The RPC code is now thread safe. Threads can now use the same service
3972 of different services at the same time. Patch by Eric Norum
3973 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3975 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3976 family of functions for Linux/S390.
3978 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3979 of functions for Linux/x86.
3981 * Port to Linux/CRIS contributed by Axis Communications.
3985 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3986 we can now check for namespace violations and missing declarations. The
3987 result is that almost all headers are now Unix-compliant (as defined in
3988 the upcoming XPG6). The negative side is that some programs might need
3989 corrections, too, if they depend on the incorrect form of the headers in
3990 previous versions which defined too many symbols and included too many
3993 * regex now handles multibyte character sets correctly.
3994 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3996 * iconv (the program) does now conform to the upcoming XPG6 and handles
3997 charmaps. Instead of the charset names the path of charmaps can be
3998 provided and the conversion happens based on this data.
3999 Contributed by Ulrich Drepper.
4001 * The locale program now provides detailed information about the installed
4006 only lists the names of the supported locales
4010 provides details such as country, language, and codeset name.
4011 Contributed by Ulrich Drepper.
4015 * The gencat program now parses the input file according to the charset
4016 selected by the LC_CTYPE category. This is important for stateful
4017 character sets. To make generating catalogs easier there is a way
4018 to overwrite the charset selected by the locale: before the first
4019 message or $ quote line the catalog can contain a line like
4021 $ codeset=ISO-8859-2
4023 to select the charset (ISO-8859-2 in this case).
4025 Implemented by Shinya Hanataka and Ulrich Drepper.
4027 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4028 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4029 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4031 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4032 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4034 * The LANGUAGE environment variable is now ignored unless the locale is
4035 changed from the default "C" locale.
4037 * The usual bug fixes.
4041 * Greg McGary added runtime support for bounds checking using gcc's
4042 new -fbounded-pointers option. ix86 target is complete. PowerPC
4045 * Thorsten Kukuk added secure mode support to nscd.
4047 * The Berkeley db libraries have been removed.
4049 Related, the nss_db module is now in a separate package since it
4050 obviously requires a database library being available.
4052 * Wide character I/O streams implemented by Ulrich Drepper.
4054 * Functions from the extended socket API added by Ulrich Drepper.
4056 * Functions feenableexcept and fedisableexcept to control the
4057 behaviour of individual exceptions have been added by Andreas Jaeger.
4059 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4061 * The resolver code has been updated from bind 8.2.3-T5B which supports
4062 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4065 This change could in some situations effect backward compatibility. Since
4066 now `_res' is a thread-local instead of a global variable, modifying it
4067 in one thread does not have any effect in other threads.
4069 The resolver library was also extended to allow IPv6 as the transport
4070 protocol for the requests. This work was done by Stig Venaas.
4072 * Compatibility code for K&R C compilers has been removed from the
4073 header files. A ISO C compiler is needed to use the library
4074 (conforming to either C89 or C99 standard).
4076 * Complete rewrite of the localedef program to support multibyte character
4077 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4078 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4079 Implemented by Ulrich Drepper.
4081 Bruno Haible significantly improved the generation and use of the data
4082 structures for the wide character tables.
4084 * Plural handling in gettext implemented by Ulrich Drepper.
4086 * The utmp daemon has been removed.
4088 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4090 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4093 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4095 * POSIX barriers implemented by Kaz Kylheku.
4097 * POSIX spawn function family implemented by Ulrich Drepper.
4099 * POSIX spinlocks are now available.
4101 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4103 * the configure option --enable-kernel=X.Y.Z allows to strip out
4104 compatibility for kernel versions before X.Y.Z. This is currently only
4105 implemented for Linux.
4107 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4108 field sin6_scope_id. This means that all programs using IPv6 should be
4109 recompiled. Don't expect binary compatibility with previous glibc
4112 * various conversion modules for IBM character sets contributed by
4115 * IA-64 port by Jes Sorensen and HJ Lu.
4129 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4131 * New xdr functions are added; some rpc functions are now 64bit clean.
4133 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4135 * Fixed known incompatibilities with glibc 2.0.
4137 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4139 * Optimized a number of functions (especially the ELF dynamic loader).
4141 * Update timezone data files.
4143 * lots of charmaps corrections
4145 * some new locale definitions and charmaps
4150 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4151 conform to POSIX member type requirements. Symbol versions have been
4152 adjusted as needed within the library, and for direct use by applications,
4153 but there is potential for problems if third-party libraries use
4154 struct timeval as part of their interface. This does not present
4155 a problem for X and other "essential" system libraries.
4157 * An additional locale model to support C++ Standard Library locale
4158 model and probably more was implemented by Ulrich Drepper.
4160 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4163 * Miles Bader provided the `argp' function family to support hierachical
4164 command line argument parsing, layered on top of getopt.
4166 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4168 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4171 * scanf recognizes the %a and %A format for scanning floating point numbers.
4173 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4174 information and interfaces for the available integer types.
4176 * about 130 new math functions were added to implement the ISO C9x math
4179 * the new header <complex.h> contains definitions of the complex math
4180 functions from ISO C 9X.
4182 * the new header <tgmath.h> defines generic macros to use complex or
4183 real valued functions.
4185 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4187 * Andreas Jaeger provided a test suite for the math library.
4189 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4191 * Ulrich Drepper added character set conversion functions (iconv).
4193 * Optimized string functions have been added.
4195 * The localedata addon is now part of glibc.
4197 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4199 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4200 daemon for NSS (nscd).
4202 Missing a better place here are some numbers on improvements. Under
4203 Linux 2.1.125 un-tar-ing the kernel sources takes
4207 using local files 12.19s 6.88s 22.91s
4209 using NIS 13.92s 8.91s 26.34s
4211 using NIS & nscd 10.37s 7.34s 25.30s
4213 using NIS+ 27.57s 30.37s 640.46s
4215 using NIS+ & nscd 10.25s 7.83s 26.51s
4217 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4219 Keep in mind that non-namelookup related operations dominate above times.
4220 It was just a common complain that using NIS+ unpacking the kernel is
4223 [1] The old nscd implementation is not available anymore since it was
4224 distributed with glibc up to version 2.0.98 and thus is now replaced.
4226 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4228 * Mark Kettenis provided a Hesiod NSS module.
4230 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4231 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4233 * rcmd can now handle netgroups (Dick Streefland).
4235 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4238 * Support for the IPv6 protocol has been added to the socket API, as per the
4239 latest draft standards.
4241 * Support for Linux 2.2 has been added.
4243 * Interface changes relative to the latest 2.0.x release:
4244 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4245 addseverity NEW: Unix98
4246 alphasort64 NEW: LFS
4247 argp_err_exit_status NEW: argp, GNU ext
4248 argp_error NEW: argp, GNU ext
4249 argp_failure NEW: argp, GNU ext
4250 argp_help NEW: argp, GNU ext
4251 argp_parse NEW: argp, GNU ext
4252 argp_program_bug_address NEW: argp, GNU ext
4253 argp_program_version NEW: argp, GNU ext
4254 argp_program_version_hook NEW: argp, GNU ext
4255 argp_state_help NEW: argp, GNU ext
4256 argp_usage NEW: argp, GNU ext
4257 authdes_create NEW: Secure RPC
4258 authdes_getucred NEW: Secure RPC
4259 authdes_pk_create NEW: Secure RPC
4260 backtrace NEW: GNU ext.
4261 backtrace_symbols NEW: GNU ext.
4262 backtrace_symbols_fd NEW: GNU ext.
4264 cacosf NEW: ISO C 9x
4265 cacosh NEW: ISO C 9x
4266 cacoshf NEW: ISO C 9x
4267 cacoshl NEW: ISO C 9x
4268 cacosl NEW: ISO C 9x
4275 casinf NEW: ISO C 9x
4276 casinh NEW: ISO C 9x
4277 casinhf NEW: ISO C 9x
4278 casinhl NEW: ISO C 9x
4279 casinl NEW: ISO C 9x
4281 catanf NEW: ISO C 9x
4282 catanh NEW: ISO C 9x
4283 catanhf NEW: ISO C 9x
4284 catanhl NEW: ISO C 9x
4285 catanl NEW: ISO C 9x
4286 cbc_crypt NEW: Secure RPC
4290 ccoshf NEW: ISO C 9x
4291 ccoshl NEW: ISO C 9x
4297 cimagf NEW: ISO C 9x
4298 cimagl NEW: ISO C 9x
4299 clearerr_locked REMOVED
4300 clntunix_create NEW: sunrpc ext
4302 clog10 NEW: ISO C 9x
4303 clog10f NEW: ISO C 9x
4304 clog10l NEW: ISO C 9x
4314 cprojf NEW: ISO C 9x
4315 cprojl NEW: ISO C 9x
4317 crealf NEW: ISO C 9x
4318 creall NEW: ISO C 9x
4323 csinhf NEW: ISO C 9x
4324 csinhl NEW: ISO C 9x
4327 csqrtf NEW: ISO C 9x
4328 csqrtl NEW: ISO C 9x
4332 ctanhf NEW: ISO C 9x
4333 ctanhl NEW: ISO C 9x
4335 des_setparity NEW: Secure RPC
4336 ecb_crypt NEW: Secure RPC
4337 endutxent NEW: Unix98
4339 exp10f NEW: ISO C 9x
4340 exp10l NEW: ISO C 9x
4344 fattach NEW: STREAMS
4345 fdetach NEW: STREAMS
4349 feclearexcept NEW: ISO C 9x
4350 fegetenv NEW: ISO C 9x
4351 fegetexceptflag NEW: ISO C 9x
4352 fegetround NEW: ISO C 9x
4353 feholdexcept NEW: ISO C 9x
4355 feraiseexcept NEW: ISO C 9x
4356 ferror_locked REMOVED
4357 fesetenv NEW: ISO C 9x
4358 fesetexceptflag NEW: ISO C 9x
4359 fesetround NEW: ISO C 9x
4360 fetestexcept NEW: ISO C 9x
4361 feupdateenv NEW: ISO C 9x
4362 fflush_locked REMOVED
4366 fgets_unlocked NEW: GNU ext.
4367 fileno_locked REMOVED
4379 fputc_locked REMOVED
4380 fputs_unlocked NEW: GNU ext.
4381 fread_unlocked NEW: GNU ext.
4386 fstatvfs NEW: Unix98
4390 ftruncate64 NEW: LFS
4392 fwrite_unlocked NEW: GNU ext.
4393 gai_strerror NEW: IPv6
4397 getchar_locked REMOVED
4399 getdate_err NEW: Unix98
4400 getdate_r NEW: GNU ext.
4402 getnameinfo NEW: IPv6
4403 getnetname NEW: Secure RPC
4404 getpmsg NEW: STREAMS
4405 getpt NEW: Unix98 PTY
4406 getrlimit64 NEW: LFS
4407 getutxent NEW: Unix98
4408 getutxid NEW: Unix98
4409 getutxline NEW: Unix98
4410 glob64 NEW: GNU ext.
4411 globfree64 NEW: GNU ext.
4412 gnu_get_libc_release NEW: GNU ext.
4413 gnu_get_libc_version NEW: GNU ext.
4414 grantpt NEW: Unix98 PTY
4415 host2netname NEW: Secure RPC
4417 iconv_close NEW: iconv
4418 iconv_open NEW: iconv
4419 if_freenameindex NEW: IPv6
4420 if_indextoname NEW: IPv6
4421 if_nameindex NEW: IPv6
4422 if_nametoindex NEW: IPv6
4423 in6addr_any NEW: IPv6
4424 in6addr_loopback NEW: IPv6
4425 inet6_isipv4mapped NEW: IPv6
4426 isastream NEW: STREAMS
4427 iswblank NEW: GNU ext.
4428 key_decryptsession NEW: Secure RPC
4429 key_decryptsession_pk NEW: Secure RPC
4430 key_encryptsession NEW: Secure RPC
4431 key_encryptsession_pk NEW: Secure RPC
4432 key_gendes NEW: Secure RPC
4433 key_get_conv NEW: Secure RPC
4434 key_secretkey_is_set NEW: Secure RPC
4435 key_setnet NEW: Secure RPC
4436 key_setsecret NEW: Secure RPC
4437 llrint NEW: ISO C 9x
4438 llrintf NEW: ISO C 9x
4439 llrintl NEW: ISO C 9x
4440 llround NEW: ISO C 9x
4441 llroundf NEW: ISO C 9x
4442 llroundl NEW: ISO C 9x
4447 lrintf NEW: ISO C 9x
4448 lrintl NEW: ISO C 9x
4449 lround NEW: ISO C 9x
4450 lroundf NEW: ISO C 9x
4451 lroundl NEW: ISO C 9x
4453 makecontext NEW: Unix98
4454 mempcpy NEW: GNU ext.
4457 modify_ldt NEW: kernel
4461 nearbyint NEW: ISO C 9x
4462 nearbyintf NEW: ISO C 9x
4463 nearbyintl NEW: ISO C 9x
4464 netname2host NEW: Secure RPC
4465 netname2user NEW: Secure RPC
4466 nexttoward NEW: ISO C 9x
4467 nexttowardf NEW: ISO C 9x
4468 nexttowardl NEW: ISO C 9x
4472 passwd2des NEW: Secure RPC
4474 pow10f NEW: GNU ext.
4475 pow10l NEW: GNU ext.
4478 printf_size NEW: GNU ext.
4479 printf_size_info NEW: GNU ext.
4480 profil_counter REMOVED
4481 pthread_mutexattr_getkind_np REPLACED
4482 pthread_mutexattr_setkind_np REPLACED
4483 ptsname NEW: Unix98 PTY
4484 ptsname_r NEW: Unix98 PTY
4486 putchar_locked REMOVED
4487 putgrent NEW: GNU ext.
4489 putpmsg NEW: STREAMS
4490 pututxline NEW: Unix98
4494 readdir64_r NEW: LFS
4495 remquo NEW: ISO C 9x
4496 remquof NEW: ISO C 9x
4497 remquol NEW: ISO C 9x
4499 roundf NEW: ISO C 9x
4500 roundl NEW: ISO C 9x
4502 scalbln NEW: ISO C 9x
4503 scalblnf NEW: ISO C 9x
4504 scalblnl NEW: ISO C 9x
4506 sendfile NEW: kernel
4507 setcontext NEW: Unix98
4508 setrlimit64 NEW: LFS
4509 setutxent NEW: Unix98
4511 sigignore NEW: Unix98
4512 sigqueue NEW: POSIX.1b
4513 sigrelse NEW: Unix98
4514 sigset NEW: POSIX.1b
4515 sigtimedwait NEW: POSIX.1b
4516 sigwaitinfo NEW: POSIX.1b
4517 sincos NEW: GNU ext.
4518 sincosf NEW: GNU ext.
4519 sincosl NEW: GNU ext.
4523 strcasestr NEW: GNU ext.
4524 strtoimax NEW: ISO C 9x
4525 strtoumax NEW: ISO C 9x
4526 strverscmp NEW: GNU ext.
4527 svcauthdes_stats NEW: Secure RPC
4528 svcunix_create NEW: sunrpc etx
4529 svcunixfd_create NEW: sunrpc ext
4530 swapcontext NEW: Unix98
4531 tcgetsid NEW: Unix98 PTY
4532 tdestroy NEW: GNU ext.
4533 tgamma NEW: ISO C 9x
4534 tgammaf NEW: ISO C 9x
4535 tgammal NEW: ISO C 9x
4539 truncf NEW: ISO C 9x
4540 truncl NEW: ISO C 9x
4542 unlockpt NEW: Unix98 PTY
4543 updwtmpx NEW: Unix98
4544 user2netname NEW: Secure RPC
4545 utmpxname NEW: Unix98
4546 versionsort NEW: GNU ext.
4547 versionsort64 NEW: GNU ext.
4549 wcscasecmp NEW: GNU ext.
4550 wcsncasecmp NEW: GNU ext.
4551 wcsnlen NEW: GNU ext.
4552 wcstoimax NEW: ISO C 9x
4553 wcstoll NEW: ISO C 9x
4554 wcstoull NEW: ISO C 9x
4555 wcstoumax NEW: ISO C 9x
4557 wordexp NEW: POSIX.2
4558 wordfree NEW: POSIX.2
4559 write_profiling REMOVED
4560 xdecrypt NEW: Secure RPC
4561 xdr_authdes_cred NEW: Secure RPC
4562 xdr_authdes_verf NEW: Secure RPC
4563 xdr_cryptkeyarg NEW: Secure RPC
4564 xdr_cryptkeyarg2 NEW: Secure RPC
4565 xdr_cryptkeyres NEW: Secure RPC
4566 xdr_getcredres NEW: Secure RPC
4567 xdr_key_netstarg NEW: Secure RPC
4568 xdr_key_netstres NEW: Secure RPC
4569 xdr_keybuf NEW: Secure RPC
4570 xdr_keystatus NEW: Secure RPC
4571 xdr_netnamestr NEW: Secure RPC
4572 xdr_sizeof NEW: Secure RPC
4573 xdr_unixcred NEW: sunrpc ext
4574 xencrypt NEW: Secure RPC
4575 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4586 * inet_ntoa is thread-safe
4588 * updwtmp is moved from libutil to libc
4590 * rewrite of cbrt function
4592 * update of timezone data
4606 * add atoll function
4608 * fix complex problems in Berkeley DB code
4610 * fix math functions
4614 * fixed lots of header problems (especially Linux/GNU specific)
4616 * dynamic loader preserves all registers
4618 * Roland McGrath provided support for handling of auxiliary objects in
4619 the ELF dynamic loader.
4621 * support for parallel builds is improved
4625 * GNU extensions are no longer declared by default. To enable them you
4626 must define the macro `_GNU_SOURCE' in your program or compile with
4629 * The library has changed from using GNU ld symbol aliases to using weak
4630 symbols where available. The ELF object file format supports weak
4631 symbols; GNU ld also supports weak symbols in the a.out format. (There
4632 is also now support for other GNU ld extensions in ELF. Use the
4633 `--with-elf' option to configure to indicate you have ELF, and
4634 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4635 of many files which contained only symbol aliases, reducing the size of
4636 the source and the compiled library; many other files were renamed to
4637 less cryptic names previously occupied by the symbol alias files.
4638 There is a new header file <elf.h> for programs which operate on
4639 files in the ELF format.
4641 * Converted to Autoconf version 2, so `configure' has more options.
4642 Run `configure --help' to see the details.
4644 * The library can now be configured to build profiling, highly-optimized
4645 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4646 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4647 `configure' enable building these extra libraries. The shared library is
4648 built by default when using both ELF and GNU ld. When shared libraries
4649 are enabled, the new library `-ldl' is available for arbitrary run-time
4650 loading of shared objects; its interface is defined in <dlfcn.h>. The
4651 new header file <link.h> gives access to the internals of the run-time
4652 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4653 application of same name on other systems and it provides information
4654 about dynamically linked binaries.
4656 * The C library now provides the run-time support code for profiling
4657 executables compiled with `-pg'. Programs can control the profiling code
4658 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4659 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4660 the support for this file format was contributed by David Mosberger-Tang.
4662 * The math code has been replaced with a math library based on fdlibm from
4663 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4664 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4665 functions. The math functions now reside in a separate library, so
4666 programs using them will need to use `-lm' their linking commands.
4668 * John C. Bowman contributed optimized ix87 assembler inline functions.
4670 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4671 mechanism similar to that found in Solaris 2. This is now used for the
4672 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4673 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4674 file controls what services are used for each individual database. This
4675 works by loading shared libraries with names specified in `nsswitch.conf',
4676 so service modules can be changed or added at any time without even
4677 relinking any program. Currently there are the file, db, and NIS based
4678 NSS services available.
4680 * The new functions `strtoq' and `strtouq' parse integer values from
4681 strings, like `strtol' and `strtoul', but they return `long long int' and
4682 `unsigned long long int' values, respectively (64-bit quantities).
4684 * The new functions `strtof' and `strtold' parse floating-point values from
4685 strings, like `strtod', but they return `float' and `long double' values,
4686 respectively (on some machines `double' and `long double' are the same).
4688 * Ulrich Drepper has contributed new implementations of the floating-point
4689 printing and reading code used in the `printf' family of functions and
4690 `strtod', `strtof', and `strtold'. These new functions are perfectly
4691 accurate, and much faster than the old ones.
4693 * The implementation of the POSIX locale model was completely rewritten by
4694 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4695 to compile the POSIX locale definition.
4697 * The former dummy implementations of the strcoll and strxfrm function are
4698 now replaced by fully functional code contributed by Ulrich Drepper. The
4699 collation information comes from the POSIX locale definitions.
4701 * The new header <langinfo.h> defines an interface for accessing
4702 various locale-dependent data (using the locale chosen with `setlocale').
4704 * Ulrich Drepper has contributed a new suite of functions for operation on
4705 wide-character and multibyte-character strings, in <wchar.h>;
4706 and classification and case conversion of wide characters, in <wctype.h>.
4707 These new functions are conforming to the ISO C, Amendement 1 specification.
4709 * There is now a second implementation of the standard I/O library available.
4710 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4711 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4713 * You can now use positional parameter specifications in format strings
4714 for the `printf' and `scanf' families of functions. For example,
4715 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4716 ``Number 6, Mr Jones''. This is mainly useful when providing different
4717 format strings for different languages, whose grammars may dictate
4718 different orderings of the values being printed. To support this
4719 feature, the interface for `register_printf_handler' has changed; see
4720 the header file <printf.h> for details.
4722 * The `printf' and `scanf' families of functions now understand a new
4723 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4724 to group numbers as indicated by the locale; for `scanf' and friends, this
4725 says to accept as valid only a number with all the proper grouping
4726 separators in the right places. In the default "C" locale, numbers are
4727 not grouped; but locales for specific countries will define the usual
4728 conventions (i.e. separate thousands with `,' in the US locale).
4730 * The pgrp functions have been regularized, slightly incompatibly but much
4731 less confusingly. The core functions are now `getpgid' and `setpgid',
4732 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4733 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4734 provided for compatibility. There is no longer an incompatible `getpgrp'
4735 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4737 * The new header file <fts.h> and suite of functions simplify programs that
4738 operate on directory trees. This code comes from 4.4 BSD.
4740 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4741 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4742 NSS scheme used in glibc.
4744 * There is a new malloc debugging hook `__memalign_hook'.
4746 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4747 `unsigned int' in <sys/types.h>. These are for compatibility only and
4748 their use is discouraged.
4750 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4751 done differently, so that it works even without GNU ld.
4753 * New function `euidaccess' checks allowed access to a file like `access',
4754 but using the effective IDs instead of the real IDs.
4756 * The time zone data files have been updated for the latest and greatest
4757 local time conventions of the countries of the world.
4759 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4762 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4763 for formatting floating-point numbers. They are provided only for
4764 compatibility; new programs should use `sprintf' instead. There are
4765 also equivalent function for the `long double' floating-point type and
4766 all functions also exist in a reentrant form.
4768 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4769 functions for maintaining the login-record files (primarily of use to
4770 system programs such as `login'), and convenient functions for
4771 allocating and initializing a pseudo-terminal (pty) device.
4773 * Ulrich Drepper has contributed new support for System V style
4774 shared memory and IPC on systems that support it.
4776 * Ulrich Drepper has contributed several miscellaneous new functions found
4777 in System V: The `hsearch' family of functions provide an effective
4778 implementation of hash tables; `a64l' and `l64a' provide a very simple
4779 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4782 * Ulrich Drepper has contributed new reentrant counterparts for the
4783 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4785 * Ulrich Drepper has contributed new, highly-optimized versions of several
4786 string functions for the i486/Pentium family of processors.
4788 * Ulrich Drepper has updated the Linux-specific code, based largely
4789 on work done in Hongjiu Lu's version of GNU libc for Linux.
4790 The GNU library now supports Linux versions 2.0.10 and later,
4791 using the ELF object file format (i[3456]86-*-linux).
4793 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4795 * David Mosberger-Tang and Richard Henderson have ported the C library
4796 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4797 dynamic linking support for ELF/Alpha.
4799 * Richard Henderson contributed several Alpha optimized assembler function
4800 for arithmetic and string handling.
4802 * Ulrich Drepper has contributed a new set of message catalog functions to
4803 support multiple languages using the <libintl.h> interface, for use with
4804 his new package GNU gettext. Translation volunteers have contributed
4805 catalogs of the library's messages in Spanish, German, and Korean.
4807 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4808 program and the `catgets' function for reading the catalog files it
4809 creates. (The <libintl.h> interface is preferred; we include the
4810 <nl_types.h> interface using `catgets' only for source compatibility with
4811 programs already written to use it.)
4813 * New header file <values.h> gives SVID-compatible names for <limits.h>
4816 * Various new macros, declarations, and small header files for compatibility
4819 * New function `group_member' is a convenient way to check if a process has
4820 a given effective group ID.
4822 * When using GCC 2.7 and later, the socket functions are now declared in a
4823 special way so that passing an argument of type `struct sockaddr_in *',
4824 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4825 `struct sockaddr *' type, does not generate a type-clash warning.
4827 * New function `error' declared in header file <error.h> is a convenient
4828 function for printing error messages and optionally exiting; this is the
4829 canonical function used in GNU programs. The new functions `err', `warn',
4830 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4831 doing the same thing.
4833 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4834 POSIX.2 `glob' function to do ~ and {...} expansion.
4836 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4837 `clearenv' which is used in POSIX.9 is also available.
4839 * New function `getsid' returns session ID number on systems that support it.
4841 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4842 files <db.h> and <mpool.h> provide a rich set of functions for several
4843 types of simple databases stored in memory and in files, and <ndbm.h> is
4844 an old `ndbm'-compatible interface using the `db' functions. Link with
4845 `-ldb' to get these functions.
4847 * New macro `strdupa' copies a string like `strdup', but uses local stack
4848 space from `alloca' instead of dynamic heap space from `malloc'.
4850 * New function `strnlen' is like `strlen' but searches only a given maximum
4851 number of characters for the null terminator. `stpncpy', `strndup' and
4852 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4855 * New function `statfs' in header <sys/statfs.h>.
4857 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4858 provide convenient functions for operating on blocks of null-terminated
4861 * A new suite of functions in <utmp.h> handle all the details of reading
4862 and writing the utmp file.
4864 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4867 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4868 completely POSIX compliant and also implemented the extended functionality
4869 to handle alternate digit representation and alternate era date formats.
4871 * Ulrich Drepper provided an implementation of the `strptime' function
4872 defined in XPG4.2 which transforms a string into a `struct tm' value.
4874 * Paul Eggert provided the tzselect shell script as part of the timezone
4875 code. The shell script makes it easy to select the correct timezone
4878 * The implementation of the malloc family of functions is completely replaced
4879 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4880 The implementation uses the mmap function (if available) and it is
4881 optimized for the use in multi threaded programs.
4883 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4884 functions. This new functionality is usable by specifying a special
4885 salt string and it is compatible with implementation on *BSD systems.
4887 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4888 `getsubopt' to handle second level command line options, `bsd_signal'
4889 to access BSD style `signal' functionality, the obsolete `regexp' style
4892 * the `lchown' function is available on system which support this
4895 * The implementation of the shadow password handling function was contributed
4898 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4900 * POSIX.1g support was added. The <sys/select.h> header is available,
4901 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4902 implementation of `getaddrinfo'.
4906 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4908 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4909 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4911 * The new function `daemon' from 4.4 BSD is useful for server programs that
4912 want to put themselves in the background.
4914 * Joel Sherrill has contributed support for several standalone boards that
4915 run without an operating system.
4917 * `printf', `scanf' and friends now accept a `q' type modifier for long
4918 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4920 * All of the code taken from BSD (notably most of the math and networking
4921 routines) has been updated from the BSD 4.4-Lite release.
4923 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4925 * The new functions `getdomainname' and `setdomainname' fetch or change the
4926 YP/NIS domain name. These are system calls which exist on systems which
4929 * The time zone data files have been updated for the latest international
4932 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4933 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4937 * The C library now includes support for Sun RPC, from Sun's free
4938 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4939 are included. (There is still no support for YP.)
4941 * Tom Quinn has contributed a port of the C library to SGI machines running
4942 Irix 4 (mips-sgi-irix4).
4944 * The new `lockf' function is a simplified interface to the locking
4945 facilities of `fcntl', included for compatibility.
4947 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4949 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4951 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4954 * The error message printed by `assert' for a failed assertion now includes
4955 the name of the program (if using GNU ld) and the name of the calling
4956 function (with versions of GCC that support this).
4958 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4960 * The library now includes the <sys/mman.h> header file and memory
4961 management functions `mmap', `munmap', `mprotect', `msync', and
4962 `madvise', on systems that support those facilities.
4964 * The interface for `mcheck' has changed slightly: the function called to
4965 abort the program when an allocation inconsistency is detected now takes
4966 an argument that indicates the type of failure. The new function
4967 `mprobe' lets you request a consistency check for a particular block at
4968 any time (checks are normally done only when you call `free' or `realloc'
4971 * It is now possible to easily cross-compile the C library, building on one
4972 system a library to run on another machine and/or operating system. All
4973 you need to do is set the variable `HOST_CC' in `configparms' to the
4974 native compiler for programs to run on the machine you are building on (a
4975 few generator programs are used on Unix systems); set `CC' to the
4978 * The new function `fexecve' (only implemented on the GNU system) executes
4979 a program file given a file descriptor already open on the file.
4983 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4984 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4986 * You can set the variable `libprefix' in `configparms' to specify a prefix
4987 to be prepended to installed library files; this makes it easy to install
4988 the GNU C library to be linked as `-lgnuc' or whatever.
4990 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4991 copies a limited number of characters from a string, and returns the
4992 address of the last character written.
4994 * You no longer need to check for whether the installed `stddef.h' is
4995 compatible with the GNU C library. configure now checks for you.
4997 * You can now define a per-stream `fileno' function to convert the
4998 stream's cookie into an integral file descriptor.
5000 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5001 allocates zero bytes of storage, and returns a unique pointer which
5002 you can pass to `realloc' or `free'. The behavior is undefined if
5003 you dereference this pointer.
5005 * The C library now runs on Sony NEWS m68k machines running either
5006 NewsOS 3 or NewsOS 4.
5008 * The new `syscall' function is a system-dependent primitive function
5009 for invoking system calls. It has the canonical behavior on Unix
5010 systems, including unreliable return values for some calls (such as
5011 `pipe', `fork' and `getppid').
5013 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5014 to `EAGAIN', which is the preferred name. On systems whose kernels
5015 use two distinct codes, the C library now translates EWOULDBLOCK to
5016 EAGAIN in every system call function.
5020 * The GNU C Library Reference Manual is now distributed with the library.
5021 `make dvi' will produce a DVI file of the printed manual.
5022 `make info' will produce Info files that you can read on line using C-h i
5023 in Emacs or the `info' program.
5024 Please send comments on the manual to bug-glibc-manual@gnu.org.
5026 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5028 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5030 * Jason Merrill has contributed a port to the Sequent Symmetry running
5031 Dynix version 3 (i386-sequent-dynix).
5033 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5034 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5036 * New function `memory_warnings' lets you arrange to get warnings when
5037 malloc is running out of memory to allocate, like Emacs gives you.
5039 * The C library now contains the relocating allocator used in Emacs 19 for
5040 its editing buffers. This allocator (ralloc) minimizes allocation
5041 overhead and fragmentation by moving allocated regions around whenever it
5042 needs to. You always refer to a ralloc'd region with a "handle" (a
5043 pointer to a pointer--an object of type `void **').
5045 * There is a new `printf' format: `%m' gives you the string corresponding
5046 to the error code in `errno'.
5048 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5049 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5050 pass, the `a' modifier says to fill in a `char **' you pass with a
5053 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5054 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5055 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5057 * `mkstemp' is a traditional Unix function to atomically create and open a
5058 uniquely-named temporary file.
5062 * The standard location for the file that says what the local timezone is
5063 has changed again. It is now `/usr/local/etc/localtime' (or more
5064 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5066 * The distribution no longer contains any files with names longer than 14
5069 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5070 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5072 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5076 * The configuration process has changed quite a bit. The `configure'
5077 script is now used just like the configuration scripts for other GNU
5078 packages. The `sysdeps' directory hierarchy is much rearranged.
5079 The file `INSTALL' explains the new scheme in detail.
5081 * The header files no longer need to be processed into ANSI C and
5082 traditional C versions. There is just one set of files to install, and
5083 it will work with ANSI or old C compilers (including `gcc -traditional').
5085 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5086 MIPS DECStation running Ultrix 4.
5088 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5089 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5090 made itself into a shared library.
5092 * Yet further improved support for the i386, running 4.3 BSD-like systems
5093 (such as Mach 3 with the Unix single-server), or System V.
5095 * New function `strncasecmp' to do case-insensitive string comparison
5096 with limited length.
5098 * New function `strsep' is a reentrant alternative to `strtok'.
5100 * New functions `scandir' and `alphasort' for searching directories.
5102 * New function `setenv' is a better interface to `putenv'.
5104 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5105 function for traversing a directory tree.
5107 * The GNU obstack package is now also part of the C library.
5108 The new function `open_obstack_stream' creates a stdio stream that
5109 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5110 formatted output directly to an obstack.
5112 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5113 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5115 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5117 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5118 things to your strings.
5120 * There are some new test programs: test-fseek, testmb, and testrand.
5122 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5123 These ports are not finished, but are a good starting place for really
5124 supporting those systems.
5126 * `/etc/localtime' is now the standard location for the file that says what
5127 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5128 This follows the general principle that `/etc' is the place for all local
5129 configuration files.
5131 * The C library header files now use `extern "C"' when used by the C++
5132 compiler, so the C library should now work with C++ code.
5134 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5135 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5138 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5139 function which uses the `merge sort' algorithm, and is said to be
5140 significantly faster than the old GNU `qsort' function. Merge sort is
5141 now the standard `qsort' function. The new algorithm can require a lot
5142 of temporary storage; so, the old sorting function is called when the
5143 required storage is not available.
5145 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5146 provides the Unix `crypt' function, plus some other entry points.
5148 * The code and header files taken from 4.4 BSD have been updated with the
5149 latest files released from Berkeley.
5151 ----------------------------------------------------------------------
5152 Copyright information:
5154 Copyright (C) 1992-2018 Free Software Foundation, Inc.
5156 Permission is granted to anyone to make or distribute verbatim copies
5157 of this document as received, in any medium, provided that the
5158 copyright notice and this permission notice are preserved,
5159 thus giving the recipient permission to redistribute in turn.
5161 Permission is granted to distribute modified versions
5162 of this document, or of portions of it,
5163 under the above conditions, provided also that they
5164 carry prominent notices stating who last changed them.
5167 version-control: never