]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
x86-64 memchr/wmemchr: Properly handle the length parameter [BZ #24097]
[thirdparty/glibc.git] / NEWS
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.
4
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
7 \f
8 Version 2.28.1
9
10 The following bugs are resolved with this release:
11
12 [19444] build failures with -O1 due to -Wmaybe-uninitialized
13 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
14 [22927] libanl: properly cleanup if first helper thread creation failed
15 [23400] stdlib/test-bz22786.c creates temporary files in glibc source tree
16 [23497] readdir64@GLIBC_2.1 cannot parse the kernel directory stream
17 [23509] CET enabled glibc is incompatible with the older linker
18 [23521] nss_files aliases database file stream leak
19 [23538] pthread_cond_broadcast: Fix waiters-after-spinning case
20 [23562] signal: Use correct type for si_band in siginfo_t
21 [23578] regex: Fix memory overread in re_compile_pattern
22 [23579] libc: Errors misreported in preadv2
23 [23606] Missing ENDBR32 in sysdeps/i386/start.S
24 [23614] powerpc: missing CFI register information in __mpn_* functions
25 [23679] gethostid: Missing NULL check for gethostbyname_r result
26 [23709] Fix CPU string flags for Haswell-type CPUs
27 [23717] Fix stack overflow in stdlib/tst-setcontext9
28 [23821] si_band in siginfo_t has wrong type long int on sparc64
29 [23822] ia64 static libm.a is missing exp2f, log2f and powf symbols
30 [23927] Linux if_nametoindex() does not close descriptor (CVE-2018-19591)
31 [23972] __old_getdents64 uses wrong d_off value on overflow
32 [24018] gettext may return NULL
33 [24022] riscv may lack <asm/syscalls.h>
34 [24024] strerror() might set errno to ENOMEM due to -fno-math-error
35 [24027] malloc: Integer overflow in realloc
36 [24034] tst-cancel21-static fails with SIGBUS on pre-ARMv7 when using GCC 8
37 [24097] Can't use 64-bit register for size_t in assembly codes for x32 (CVE-2019-6488)
38
39 Security related changes:
40
41 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
42 denial of service due to resource exhaustion when processing getaddrinfo
43 calls with crafted host names. Reported by Guido Vranken.
44
45 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
46 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
47 happened, accessing the 32-bit size_t value as the full 64-bit register
48 in the assembly string/memory functions would cause a buffer overflow.
49 Reported by H.J. Lu.
50
51 \f
52 Version 2.28
53
54 Major new features:
55
56 * The localization data for ISO 14651 is updated to match the 2016
57 Edition 4 release of the standard, this matches data provided by
58 Unicode 9.0.0. This update introduces significant improvements to the
59 collation of Unicode characters. This release deviates slightly from
60 the standard in that the collation element ordering for lowercase and
61 uppercase LATIN script characters is adjusted to ensure that regular
62 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
63 is not matched by [a-z]. With the update many locales have been
64 updated to take advantage of the new collation information. The new
65 collation information has increased the size of the compiled locale
66 archive or binary locales.
67
68 * The GNU C Library can now be compiled with support for Intel CET, AKA
69 Intel Control-flow Enforcement Technology. When the library is built
70 with --enable-cet, the resulting glibc is protected with indirect
71 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
72 compatible with all existing executables and shared libraries. This
73 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
74 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
75 capable of multi-byte NOPs, like x86-64 processors as well as Intel
76 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
77 x86_64 and x32 on non-CET processors. --enable-cet has been tested
78 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
79 validated for i686.
80
81 * The GNU C Library now has correct support for ABSOLUTE symbols
82 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
83 relocated incorrectly or in some cases discarded. The GNU linker can
84 make use of the newer semantics, but it must communicate it to the
85 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
86 field) to indicate such support is required.
87
88 * Unicode 11.0.0 Support: Character encoding, character type info, and
89 transliteration tables are all updated to Unicode 11.0.0, using
90 generator scripts contributed by Mike FABIAN (Red Hat).
91
92 * <math.h> functions that round their results to a narrower type are added
93 from TS 18661-1:2014 and TS 18661-3:2015:
94
95 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
96 fMxaddfNx functions.
97
98 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
99 fMxsubfNx functions.
100
101 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
102 fMxmulfNx functions.
103
104 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
105 fMxdivfNx functions.
106
107 * Two grammatical forms of month names are now supported for the following
108 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
109 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
110 support two grammatical forms in abbreviated month names: Catalan, Greek,
111 and Kashubian.
112
113 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
114 include the support for two grammatical forms of month names.
115
116 * Building and running on GNU/Hurd systems now works without out-of-tree
117 patches.
118
119 * The renameat2 function has been added, a variant of the renameat function
120 which has a flags argument. If the flags are zero, the renameat2 function
121 acts like renameat. If the flag is not zero and there is no kernel
122 support for renameat2, the function will fail with an errno value of
123 EINVAL. This is different from the existing gnulib function renameatu,
124 which performs a plain rename operation in case of a RENAME_NOREPLACE
125 flags and a non-existing destination (and therefore has a race condition
126 that can clobber the destination inadvertently).
127
128 * The statx function has been added, a variant of the fstatat64
129 function with an additional flags argument. If there is no direct
130 kernel support for statx, glibc provides basic stat support based on
131 the fstatat64 function.
132
133 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
134 library if installed. libidn2 version 2.0.5 or later is recommended. If
135 libidn2 is not available, internationalized domain names are not encoded
136 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
137 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
138 with an encoding error.) Flags which used to change the IDN encoding and
139 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
140 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
141 deprecated. They no longer have any effect.
142
143 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
144 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
145 range of ELF gABI expressions including such constructs as
146 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
147 have been further restricted, and where in the past a dynamic string
148 token sequence may have been interpreted as a literal string it will
149 now cause a load failure. These load failures were always considered
150 unspecified behaviour from the perspective of the dynamic loader, and
151 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
152 results in a load failure now.
153
154 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
155 implementation includes all the standard functions provided by
156 <threads.h>:
157
158 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
159 thrd_detach, thrd_exit, and thrd_join for thread management.
160
161 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
162 mtx_destroy for mutual exclusion.
163
164 - call_once for function call synchronization.
165
166 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
167 cnd_wait for conditional variables.
168
169 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
170
171 Application developers must link against libpthread to use ISO C threads.
172
173 Deprecated and removed features, and other changes affecting compatibility:
174
175 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
176 installed. Software that was using either header should be updated to
177 use standard <stdio.h> interfaces instead.
178
179 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
180 This was never required by the C standard, and the macros just expanded
181 to call alternative names for the same functions. If you hoped getc and
182 putc would provide performance improvements over fgetc and fputc, instead
183 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
184 necessary, flockfile and funlockfile.
185
186 * All stdio functions now treat end-of-file as a sticky condition. If you
187 read from a file until EOF, and then the file is enlarged by another
188 process, you must call clearerr or another function with the same effect
189 (e.g. fseek, rewind) before you can read the additional data. This
190 corrects a longstanding C99 conformance bug. It is most likely to affect
191 programs that use stdio to read interactive input from a terminal.
192 (Bug #1190.)
193
194 * The macros 'major', 'minor', and 'makedev' are now only available from
195 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
196 headers that happen to include <sys/types.h>. These macros are rarely
197 used, not part of POSIX nor XSI, and their names frequently collide with
198 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
199 further explanation.
200
201 <sys/sysmacros.h> is a GNU extension. Portable programs that require
202 these macros should first include <sys/types.h>, and then include
203 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
204
205 * The tilegx*-*-linux-gnu configurations are no longer supported.
206
207 * The obsolete function ustat is no longer available to newly linked
208 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
209 function has been deprecated in favor of fstatfs and statfs.
210
211 * The obsolete function nfsservctl is no longer available to newly linked
212 binaries. This function was specific to systems using the Linux kernel
213 and could not usefully be used with the GNU C Library on systems with
214 version 3.1 or later of the Linux kernel.
215
216 * The obsolete function name llseek is no longer available to newly linked
217 binaries. This function was specific to systems using the Linux kernel
218 and was not declared in a header. Programs should use the lseek64 name
219 for this function instead.
220
221 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
222 getaddrinfo and getnameinfo functions have been deprecated. The behavior
223 previously selected by them is now always enabled.
224
225 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
226 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
227 restriction (rejecting '_' in host names, among other things) has been
228 removed, for increased compatibility with non-IDN name resolution.
229
230 * The fcntl function now have a Long File Support variant named fcntl64. It
231 is added to fix some Linux Open File Description (OFD) locks usage on non
232 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
233 fcntl and LFS support is handled transparently. Also for Linux, the OFD
234 locks act as a cancellation entrypoint.
235
236 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
237 ecb_crypt, and des_setparity are no longer available to newly linked
238 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
239 longer installed. These functions encrypted and decrypted data with the
240 DES block cipher, which is no longer considered secure. Software that
241 still uses these functions should switch to a modern cryptography library,
242 such as libgcrypt.
243
244 * Reflecting the removal of the encrypt and setkey functions above, the
245 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
246 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
247 enabled.
248
249 * The obsolete function fcrypt is no longer available to newly linked
250 binaries. It was just another name for the standard function crypt,
251 and it has not appeared in any header file in many years.
252
253 * We have tentative plans to hand off maintenance of the passphrase-hashing
254 library, libcrypt, to a separate development project that will, we hope,
255 keep up better with new passphrase-hashing algorithms. We will continue
256 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
257 'crypt_r' should not need to change at all; however, distributions will
258 need to install <crypt.h> and libcrypt from a separate project.
259
260 In this release, if the configure option --disable-crypt is used, glibc
261 will not install <crypt.h> or libcrypt, making room for the separate
262 project's versions of these files. The plan is to make this the default
263 behavior in a future release.
264
265 Changes to build and runtime requirements:
266
267 GNU make 4.0 or later is now required to build glibc.
268
269 Security related changes:
270
271 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
272 been fixed by removing the glibc-internal IDNA implementation and using
273 the system-provided libidn2 library instead. Originally reported by Hanno
274 Böck and Christian Weisgerber.
275
276 CVE-2017-18269: An SSE2-based memmove implementation for the i386
277 architecture could corrupt memory. Reported by Max Horn.
278
279 CVE-2018-11236: Very long pathname arguments to realpath function could
280 result in an integer overflow and buffer overflow. Reported by Alexey
281 Izbyshev.
282
283 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
284 architecture could write beyond the target buffer, resulting in a buffer
285 overflow. Reported by Andreas Schwab.
286
287 The following bugs are resolved with this release:
288
289 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
290 [6889] manual: 'PWD' mentioned but not specified
291 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
292 when __WORDSIZE != 64
293 [13762] regex: re_search etc. should return -2 on memory exhaustion
294 [13888] build: /tmp usage during testing
295 [13932] math: dbl-64 pow unexpectedly slow for some inputs
296 [14092] nptl: Support C11 threads
297 [14095] localedata: Review / update collation data from Unicode / ISO
298 14651
299 [14508] libc: -Wformat warnings
300 [14553] libc: Namespace pollution loff_t in sys/types.h
301 [14890] libc: Make NT_PRFPREG canonical.
302 [15105] libc: Extra PLT references with -Os
303 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
304 conversion is given
305 [16335] manual: Feature test macro documentation incomplete and out of
306 date
307 [16552] libc: Unify umount implementations in terms of umount2
308 [17082] libc: htons et al.: statement-expressions prevent use on global
309 scope with -O1 and higher
310 [17343] libc: Signed integer overflow in /stdlib/random_r.c
311 [17438] localedata: pt_BR: wrong d_fmt delimiter
312 [17662] libc: please implement binding for the new renameat2 syscall
313 [17721] libc: __restrict defined as /* Ignore */ even in c11
314 [17979] libc: inconsistency between uchar.h and stdint.h
315 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
316 [18023] libc: extend_alloca is broken (questionable pointer comparison,
317 horrible machine code)
318 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
319 last constant.
320 [18471] libc: llseek should be a compat symbol
321 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
322 symbols
323 [18991] nss: nss_files skips large entry in database
324 [19239] libc: Including stdlib.h ends up with macros major and minor being
325 defined
326 [19463] libc: linknamespace failures when compiled with -Os
327 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
328 [19527] locale: Normalized charset name not recognized by setlocale
329 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
330 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
331 fpscr.c'
332 [19728] network: out of bounds stack read in libidn function
333 idna_to_ascii_4i (CVE-2016-6261)
334 [19729] network: out of bounds heap read on invalid utf-8 inputs in
335 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
336 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
337 the base address
338 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
339 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
340 [20419] dynamic-link: files with large allocated notes crash in
341 open_verify
342 [20530] libc: bswap_16 should use __builtin_bswap16() when available
343 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
344 [20980] manual: CFLAGS environment variable replaces vital options
345 [21163] regex: Assertion failure in pop_fail_stack when executing a
346 malformed regexp (CVE-2015-8985)
347 [21234] manual: use of CFLAGS makes glibc detect no optimization
348 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
349 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
350 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
351 [21508] locale: intl/tst-gettext failure with latest msgfmt
352 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
353 [21812] network: getifaddrs() returns entries with ifa_name == NULL
354 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
355 dlopen
356 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
357 AT_SECURE=1
358 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
359 [22247] network: Integer overflow in the decode_digit function in
360 puny_decode.c in libidn (CVE-2017-14062)
361 [22342] nscd: NSCD not properly caching netgroup
362 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
363 [22550] localedata: es_ES locale (and other es_* locales): collation
364 should treat ñ as a primary different character, sync the collation
365 for Spanish with CLDR
366 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
367 by gcc configured with --enable-default-pie
368 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
369 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
370 crossing 2GB threshold (CVE-2017-18269)
371 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
372 es_BO
373 [22735] time: Misleading typo in time.h source comment regarding
374 CLOCKS_PER_SECOND
375 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
376 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
377 `__assert_perror_fail ()' to `__assert_fail_base ()'
378 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
379 dlopen failures
380 [22786] libc: Stack buffer overflow in realpath() if input size is close
381 to SSIZE_MAX (CVE-2018-11236)
382 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
383 through an absolute DT_NEEDED path
384 [22792] build: tcb-offsets.h dependency dropped
385 [22797] libc: pkey_get() uses non-reserved name of argument
386 [22807] libc: PTRACE_* constants missing for powerpc
387 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
388 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
389 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
390 [22848] localedata: ca_ES: update date definitions from CLDR
391 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
392 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
393 [22896] localedata: Update locale data for an_ES
394 [22902] math: float128 test failures with GCC 8
395 [22918] libc: multiple common of `__nss_shadow_database'
396 [22919] libc: sparc32: backtrace yields infinite backtrace with
397 makecontext
398 [22926] libc: FTBFS on powerpcspe
399 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
400 required
401 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
402 [22947] libc: FAIL: misc/tst-preadvwritev2
403 [22963] localedata: cs_CZ: Add alternative month names
404 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
405 [22996] localedata: change LC_PAPER to en_US in es_BO locale
406 [22998] dynamic-link: execstack tests are disabled when SELinux is
407 disabled
408 [23005] network: Crash in __res_context_send after memory allocation
409 failure
410 [23007] math: strtod cannot handle -nan
411 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
412 [23036] regex: regex equivalence class regression
413 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
414 [23069] libc: sigaction broken on riscv64-linux-gnu
415 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
416 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
417 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
418 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
419 and libc build with -Os)
420 [23140] localedata: More languages need two forms of month names
421 [23145] libc: _init/_fini aren't marked as hidden
422 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
423 [23171] math: C++ iseqsig for long double converts arguments to double
424 [23178] nscd: sudo will fail when it is run in concurrent with commands
425 that changes /etc/passwd
426 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
427 (CVE-2018-11237)
428 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
429 [23208] localedata: New locale - Lower Sorbian (dsb)
430 [23233] regex: Memory leak in build_charclass_op function in file
431 posix/regcomp.c
432 [23236] stdio: Harden function pointers in _IO_str_fields
433 [23250] nptl: Offset of __private_ss differs from GCC
434 [23253] math: tgamma test suite failures on i686 with -march=x86-64
435 -mtune=generic -mfpmath=sse
436 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
437 AT_SECURE
438 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
439 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
440 parser.c
441 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
442 [23277] math: nan function should not have const attribute
443 [23279] math: scanf and strtod wrong for some hex floating-point
444 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
445 directed rounding
446 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
447 [23303] build: undefined reference to symbol
448 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
449 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
450 lookup
451 [23313] stdio: libio vtables validation and standard file object
452 interposition
453 [23329] libc: The __libc_freeres infrastructure is not properly run across
454 DSO boundaries.
455 [23349] libc: Various glibc headers no longer compatible with
456 <linux/time.h>
457 [23351] malloc: Remove unused code related to heap dumps and malloc
458 checking
459 [23363] stdio: stdio-common/tst-printf.c has non-free license
460 [23396] regex: Regex equivalence regression in single-byte locales
461 [23422] localedata: oc_FR: More updates of locale data
462 [23442] build: New warning with GCC 8
463 [23448] libc: Out of bounds access in IBM-1390 converter
464 [23456] libc: Wrong index_cpu_LZCNT
465 [23458] build: tst-get-cpu-features-static isn't added to tests
466 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
467 processors
468 [23467] dynamic-link: x86/CET: A property note parser bug
469
470 \f
471 Version 2.27
472
473 Major new features:
474
475 * The GNU C Library can now be compiled with support for building static
476 PIE executables (See --enable-static-pie in INSTALL). These static PIE
477 executables are like static executables but can be loaded at any address
478 and provide additional security hardening benefits at the cost of some
479 memory and performance. When the library is built with --enable-static-pie
480 the resulting libc.a is usable with GCC 8 and above to create static PIE
481 executables using the GCC option '-static-pie'. This feature is currently
482 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
483 aarch64 with binutils 2.30 or later.
484
485 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
486 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
487 H.J. Lu from Intel.
488
489 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
490
491 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
492
493 * In order to support faster and safer process termination the malloc API
494 family of functions will no longer print a failure address and stack
495 backtrace after detecting heap corruption. The goal is to minimize the
496 amount of work done after corruption is detected and to avoid potential
497 security issues in continued process execution. Reducing shutdown time
498 leads to lower overall process restart latency, so there is benefit both
499 from a security and performance perspective.
500
501 * The abort function terminates the process immediately, without flushing
502 stdio streams. Previous glibc versions used to flush streams, resulting
503 in deadlocks and further data corruption. This change also affects
504 process aborts as the result of assertion failures.
505
506 * On platforms where long double has the IEEE binary128 format (aarch64,
507 alpha, mips64, riscv, s390 and sparc), the math library now implements
508 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
509 These are the same interfaces added in version 2.26 for some platforms where
510 this format is supported but is not the format of long double.
511
512 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
513 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
514 implements interfaces for that type, as defined by ISO/IEC TS
515 18661-3:2015. These are corresponding interfaces to those supported for
516 _Float128.
517
518 * The math library now implements interfaces for the _Float32, _Float64 and
519 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
520 corresponding interfaces to those supported for _Float128.
521
522 * glibc now implements the memfd_create and mlock2 functions on Linux.
523
524 * Support for memory protection keys was added. The <sys/mman.h> header now
525 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
526 pkey_get.
527
528 * The copy_file_range function was added.
529
530 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
531
532 * The ldconfig utility now processes `include' directives using the C/POSIX
533 collation ordering. Previous glibc versions used locale-specific
534 ordering, the change might break systems that relied on that.
535
536 * Support for two grammatical forms of month names has been added.
537 In a call to strftime, the "%B" and "%b" format specifiers will now
538 produce the grammatical form required when the month is used as part
539 of a complete date. New "%OB" and "%Ob" specifiers produce the form
540 required when the month is named by itself. For instance, in Greek
541 and in many Slavic and Baltic languages, "%B" will produce the month
542 in genitive case, and "%OB" will produce the month in nominative case.
543
544 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
545 are all valid and will all accept any known form of month
546 name---standalone or complete, abbreviated or full. In a call to
547 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
548 the strings used by "%B" and "%b", respectively. New query
549 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
550 used by "%OB" and "%Ob", respectively.
551
552 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
553 define the strings for %OB and %Ob, respectively; these have the
554 same syntax as "mon" and "abmon". These arrays are optional; if they
555 are not provided then they have the same content as "mon" and "abmon",
556 respectively.
557
558 These features are provided for locales which define "alt_mon" and/or
559 "ab_alt_mon" in their locale source data. This release includes such
560 alternative month name data for the following languages: Belarusian,
561 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
562
563 This feature is currently a GNU extension, but it is expected to
564 be added to the next revision of POSIX, and it is also already
565 available on some BSD-derived operating systems.
566
567 This feature will cause existing statically compiled applications
568 to fail to load locales and fall back to the builtin C/POSIX locales.
569 See notes below for other changes affecting compatibility.
570
571 * Support for the RISC-V ISA running on Linux has been added. This port
572 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
573 for the following ISA and ABI pairs:
574
575 - rv64imac lp64
576 - rv64imafdc lp64
577 - rv64imafdc lp64d
578
579 Deprecated and removed features, and other changes affecting compatibility:
580
581 * Statically compiled applications attempting to load locales compiled for the
582 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
583 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
584 support for two grammatical forms of the month names, also extends the locale
585 data binary format. Static applications needing locale support must be
586 recompiled to match the runtime and data they are deployed with. In some
587 distributions there is an upgrade window where dynamically linked applications
588 may use a new library but the old locale data and also fall back to the
589 builtin C/POSIX locales; restarting the application process is sufficient to
590 fix this.
591
592 * Support for statically linked applications which call dlopen is deprecated
593 and will be removed in a future version of glibc. Applications which call
594 dlopen need to be linked dynamically instead.
595
596 * Support for old programs which use internal stdio data structures and
597 functions is deprecated. This includes programs which use the C++ streams
598 provided by libstdc++ in GCC 2.95. Programs which use the internal
599 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
600 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
601 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
602 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
603 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
604 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
605 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
606 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
607 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
608 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
609 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
610 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
611 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
612 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
613 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
614 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
615 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
616 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
617 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
618 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
619 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
620 future version of glibc. Unlike other symbol removals, these old
621 applications will not be supported using compatibility symbols.
622
623 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
624 defined by <sys/ptrace.h>.
625
626 * libm no longer supports SVID error handling (calling a user-provided
627 matherr function on error) or the _LIB_VERSION variable to control error
628 handling. (SVID error handling and the _LIB_VERSION variable still work
629 for binaries linked against older versions of the GNU C Library.) The
630 libieee.a library is no longer provided. math.h no longer defines struct
631 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
632 TLOSS, PLOSS and HUGE.
633
634 * The libm functions pow10, pow10f and pow10l are no longer supported for
635 new programs. Programs should use the standard names exp10, exp10f and
636 exp10l for these functions instead.
637
638 * The mcontext_t type is no longer the same as struct sigcontext. On
639 platforms where it was previously the same, this changes the C++ name
640 mangling for interfaces involving this type.
641
642 * The add-ons mechanism for building additional packages at the same time as
643 glibc has been removed. The --enable-add-ons configure option is now
644 ignored.
645
646 * The --without-fp configure option is now ignored. Whether hardware
647 floating-point instructions are used is now configured based on whether
648 the compiler used at configure time (without any options implied by a
649 --with-cpu= configure option) uses such instructions.
650
651 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
652 the specified string can be parsed as a domain name.
653
654 * In the malloc_info output, the <heap> element may contain another <aspace>
655 element, "subheaps", which contains the number of sub-heaps.
656
657 * The libresolv function p_secstodate is no longer supported for new
658 programs.
659
660 * The tilepro-*-linux-gnu configuration is no longer supported.
661
662 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
663 and will be removed in a future release. Software that is still using
664 either header should be updated to use standard <stdio.h> interfaces
665 instead.
666
667 libio.h was originally the header for a set of supported GNU extensions,
668 but they have not been maintained as such in many years, they are now
669 standing in the way of improvements to stdio, and we don't think there are
670 any remaining external users. _G_config.h was never intended for public
671 use, but predates the bits convention.
672
673 Changes to build and runtime requirements:
674
675 * bison version 2.7 or later is required to generate code in the 'intl'
676 subdirectory.
677
678 Security related changes:
679
680 CVE-2009-5064: The ldd script would sometimes run the program under
681 examination directly, without preventing code execution through the
682 dynamic linker. (The glibc project disputes that this is a security
683 vulnerability; only trusted binaries must be examined using the ldd
684 script.)
685
686 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
687 suffered from a one-byte overflow during ~ operator processing (either
688 on the stack or the heap, depending on the length of the user name).
689 Reported by Tim Rühsen.
690
691 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
692 would sometimes fail to free memory allocated during ~ operator
693 processing, leading to a memory leak and, potentially, to a denial
694 of service.
695
696 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
697 without GLOB_NOESCAPE, could write past the end of a buffer while
698 unescaping user names. Reported by Tim Rühsen.
699
700 CVE-2017-17426: The malloc function, when called with an object size near
701 the value SIZE_MAX, would return a pointer to a buffer which is too small,
702 instead of NULL. This was a regression introduced with the new malloc
703 thread cache in glibc 2.26. Reported by Iain Buclaw.
704
705 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
706 to the allocation of too much memory. (This is not a security bug per se,
707 it is mentioned here only because of the CVE assignment.) Reported by
708 Qualys.
709
710 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
711 of the number of search path components. (This is not a security
712 vulnerability per se because no trust boundary is crossed if the fix for
713 CVE-2017-1000366 has been applied, but it is mentioned here only because
714 of the CVE assignment.) Reported by Qualys.
715
716 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
717 for AT_SECURE or SUID binaries could be used to load libraries from the
718 current directory.
719
720 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
721 succeeds without returning an absolute path due to unexpected behaviour
722 of the Linux kernel getcwd syscall. Reported by halfdog.
723
724 CVE-2018-6485: The posix_memalign and memalign functions, when called with
725 an object size near the value of SIZE_MAX, would return a pointer to a
726 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
727
728 CVE-2018-6551: The malloc function, when called with an object size near
729 the value of SIZE_MAX, would return a pointer to a buffer which is too
730 small, instead of NULL.
731
732 The following bugs are resolved with this release:
733
734 [866] glob: glob should match dangling symlinks
735 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
736 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
737 locale
738 [5997] math: Very slow execution of sinf function
739 [10580] localedata: hr_HR: updated locale
740 [10871] locale: 'mon' array should contain both nominative and genitive
741 cases
742 [12349] localedata: eu_ES: incorrect thousands separator
743 [13605] localedata: shn_MM: new Shan locale
744 [13805] localedata: ru_RU: currency should use ',' as radix point
745 [13953] localedata: km_KH: locale update
746 [13994] localedata: mjw_IN: new locale
747 [14121] build: make writes .mo files in po directory
748 [14333] libc: Fix the race between atexit() and exit()
749 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
750 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
751 "Bangla"
752 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
753 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
754 width Latin characters
755 [15332] localedata: es_CU: locale update
756 [15436] stdio: Don't close or flush stdio streams on abort
757 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
758 letters
759 [16148] localedata: ca_ES: incorrect thousands separator
760 [16750] dynamic-link: ldd should not try to execute the binaries
761 (CVE-2009-5064)
762 [16777] localedata: pl_PL: incorrect thousands separator in locale
763 [16905] localedata: hanzi: new collation
764 [17563] localedata: cmn_TW: add hanzi collation
765 [17750] localedata: wrong collation order of diacritics in most locales
766 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
767 if malloc succeeded
768 [17956] build: Build fails on missing definitions from header file
769 nss/nss.h when Mozilla NSS is used for cryptography
770 [18203] libc: realpath() does not handle unreachable paths correctly
771 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
772 [18812] localedata: kab_DZ: new Kabyle Algeria locale
773 [18822] libc: Internal functions are called via PLT
774 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
775 [19170] libc: __gmon_start__ defined in hppa in crtn.S
776 [19574] libc: glibc should support building static PIE binaries
777 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
778 U+309A
779 [19971] glob: glob: Do not skip entries with zero d_ino values
780 [19982] localedata: fr.po: spelling mistake for error code EXDEV
781 [20008] localedata: km_KH: convert to translit_neutral
782 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
783 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
784 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
785 [20482] localedata: de_CH: abbreviated weekdays should be two letters
786 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
787 for Nicaragua
788 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
789 they're set, wrong errors returned by gaih_inet when lookup functions
790 are not found.
791 [20756] localedata: [PATCH] Use Unicode wise thousands separator
792 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
793 access
794 [20952] localedata: yuw_PG: new locale
795 [21084] localedata: charmaps/IBM858: new codepage
796 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
797 [21242] libc: assert gives pedantic warning in old gcc versions
798 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
799 __regcall calling convention
800 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
801 [21326] libc: C99 functions are not declared for C++11 and later if
802 _GNU_SOURCE is not predefined by g++
803 [21457] libc: sys/ucontext.h namespace
804 [21530] libc: tmpfile() should be implemented using O_TMPFILE
805 [21660] math: GCC fails to compile a formula with tgmath.h
806 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
807 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
808 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
809 [21684] math: tgmath.h handling of complex integers
810 [21685] math: tgmath.h handling of bit-fields
811 [21686] math: tgmath.h handling of __int128
812 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
813 locale
814 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
815 protector=all
816 [21750] localedata: column width of characters incompatible with classical
817 wcwidth
818 [21754] malloc: malloc: Perform as little work as possible after heap
819 consistency check failures
820 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
821 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
822 [21791] string: Unused XXX_chk_XXX functions in libc.a
823 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
824 PIE
825 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
826 locales
827 [21845] localedata: Added new Locale bho_NP
828 [21853] localedata: Fix abday Which looks same as day in zh_SG
829 [21854] localedata: Added New Locale en_SC
830 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
831 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
832 _dl_runtime_resolve_avx_slow
833 [21885] network: getaddrinfo: gethosts does not release resolver context
834 on memory allocation failure
835 [21899] libc: XPG4.2 sigaction namespace
836 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
837 the suspect)
838 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
839 default (i386)
840 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
841 errno value, causing getaddrinfo to fail
842 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
843 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
844 instead of EAI_NODATA
845 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
846 constant PTRACE_SEIZE_DEVEL
847 [21930] math: C-only gcc builtins used in <math.h> isinf
848 [21932] network: Unpaired __resolv_context_get in generic get*_r
849 implementation
850 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
851 [21944] libc: sigval namespace
852 [21951] localedata: Update hanzi collation by stroke
853 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
854 sysdeps/x86_64/fpu/e_expf.S
855 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
856 allocation in syscall loops)
857 [21959] localedata: Fix Country name for xh_ZA
858 [21960] localedata: Fix abmon for bem_ZM
859 [21966] math: AVX2 mathvec functions use FMA without checking
860 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
861 [21971] localedata: Added New Locale for mfe_MU
862 [21972] libc: assert macro requires operator== (int) for its argument type
863 [21973] math: [sparc] libm missing sqrtl compat symbol
864 [21974] libc: Remove __bb_init_func and __bb_exit_func
865 [21982] string: stratcliff.c: error: assuming signed overflow does not
866 occur with -O3
867 [21986] stdio: __guess_grouping is called incorrectly
868 [21987] math: [sparc32] wrong bits/long-double.h installed
869 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
870 amounts)
871 [22022] localedata: Missing country_name for mni_IN
872 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
873 niu_NZ
874 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
875 checks
876 [22026] locale: iconv_open: heap overflow on gconv_init failure
877 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
878 [22035] math: [m68k] bits/math-inline.h macro namespace
879 [22038] localedata: Fix abbreviated weeks and months for Somali
880 [22044] localedata: Remove redundant data for Limburgish Language
881 [22050] malloc: Linking with -lmcheck does not hook
882 __malloc_initialize_hook correctly
883 [22051] libc: zero terminator in the middle of glibc's .eh_frame
884 [22052] malloc: malloc failed to compile with GCC 7 and -O3
885 [22070] localedata: charmaps/UTF-8: wcwidth for
886 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
887 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
888 Jungseong and Jongseong) should be 0
889 [22078] nss: nss_files performance issue in multi mode
890 [22082] math: bits/math-finite.h exp10 condition
891 [22086] libc: pcprofiledump incorrect cross-endian condition
892 [22093] dynamic-link: ld.so no longer searches in .../x86_64
893 [22095] network: Name server address allocation memory leak in resolv.conf
894 parsing after OOM
895 [22096] network: __resolv_conf_attach can incorrectly free passed conf
896 object
897 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
898 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
899 e.g. ET_GNU_DEBUG_*
900 [22111] malloc: malloc: per thread cache is not returned when thread exits
901 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
902 [22134] libc: [linux] implement fexecve with execveat
903 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
904 ppc64le
905 [22145] libc: ttyname() gives up too early in the face of namespaces
906 [22146] math: C++ build issue with float128 on x86_64
907 [22153] nptl: nptl: save error code before process termination
908 [22156] libc: [hppa,ia64,microblaze] Executable stack default
909 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
910 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
911 [22165] libc: [hppa] Text relocations in libc.so
912 [22180] libc: destructor registered via __cxa_atexit is called twice
913 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
914 symlinks") cause "make" segfaults
915 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
916 math_force_eval
917 [22207] libc: FAIL: stdlib/test-atexit-race
918 [22225] math: nearbyint arithmetic moved before feholdexcept
919 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
920 [22235] math: iscanonical in C++ and float128
921 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
922 the svid compat wrapper
923 [22244] math: ynf and yn are wrong without the svid compat wrapper
924 [22273] libc: Improper assert in Linux posix_spawn implementation
925 [22284] libc: -pg -pie doesn't work
926 [22292] locale: localedef exits with error 4 when it should be error 1
927 [22294] locale: Allow "" for int_currency_symbol definition in locales.
928 [22295] locale: Don't warn on non-symbolic characters in locale sources in
929 --verbose.
930 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
931 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
932 to 2.26
933 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
934 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
935 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
936 [22322] libc: [mips64] wrong bits/long-double.h installed
937 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
938 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
939 (CVE-2017-15804)
940 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
941 [22343] malloc: Integer overflow in posix_memalign
942 [22347] libc: getrandom() returns the number of bytes that were copied to
943 the buffer even though the comments say "Return 0 on success and -1 on
944 failure."
945 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
946 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
947 -m32
948 [22370] dynamic-link: Incorrect note padding check
949 [22375] libc: malloc returns pointer from tcache_get when should return
950 NULL (CVE-2017-17426)
951 [22377] math: iseqsig, float128 and C++
952 [22382] localedata: Error in tpi_PG locale
953 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
954 inside the ASCII printable range
955 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
956 -mlong-double-64
957 [22403] localedata: Slash needs escaping in some locales
958 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
959 [22409] network: res_hnok does not accept some host names used on the
960 Internet
961 [22412] network: res_dnok, res_hnok should perform syntax checks
962 [22413] network: ns_name_pton ignores syntactically invalid trailing
963 backslash
964 [22415] stdio: setvbuf can lead to invalid free/segfault
965 [22432] build: Non-deterministic build
966 [22439] malloc: malloc_info should compute summary statistics for all sub-
967 heaps in an arena
968 [22442] network: if_nametoindex could report index for the wrong
969 networking interface
970 [22446] build: aliasing violation calling readlink in handle_request
971 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
972 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
973 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
974 __nscd_hash/__nss_hash
975 [22463] network: p_secstodate overflow handling
976 [22469] localedata: pl_PL LC_COLLATE does not use i18n
977 [22478] libc: sigwait can fail with EINTR
978 [22505] libc: ldconfig processes include directive in locale-specific
979 order
980 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
981 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
982 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
983 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
984 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
985 [22534] localedata: Collation rules for Serbian and Bosnian should be the
986 same as for Croatian
987 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
988 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
989 [22577] libc: missing newline after "cannot allocate TLS data structures
990 for initial thread"
991 [22588] manual: manual/conf.texi: missing underscore in front of
992 SC_SSIZE_MAX
993 [22593] math: nextafter and nexttoward are declared with const attribute
994 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
995 [22603] string: ia64 memchr overflows internal pointer check
996 [22605] libc: SH clone does not set the exit code correctly
997 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
998 (CVE-2017-1000408)
999 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1000 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1001 modified in case of success
1002 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1003 [22615] manual: manual: ambiguous wording about errno value in case of
1004 success
1005 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1006 stack-protector=all
1007 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1008 binaries or if /proc is not mounted (CVE-2017-16997)
1009 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1010 [22630] build: $(no-pie-ldflag) is no longer effective
1011 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1012 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1013 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1014 [22637] nptl: guard size is subtracted from thread stack size instead of
1015 adding it on top
1016 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1017 [22657] localedata: hu_HU: Avoid double space in date
1018 [22660] math: fmax, fmin sNaN handling on alpha
1019 [22664] libc: New warning of GCC8
1020 [22665] math: alpha: ceil and floor raise inexact exceptions
1021 [22666] math: alpha: trunc raise inexact exceptions
1022 [22667] libc: makecontext lacks stack alignment on i386
1023 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1024 [22679] libc: getcwd(3) can succeed without returning an absolute path
1025 (CVE-2018-1000001)
1026 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1027 elision and tunables
1028 [22687] math: [powerpc-nofpu] complex long double functions spurious
1029 "invalid" exception
1030 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1031 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1032 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1033 exception
1034 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1035 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1036 power4
1037 [22701] nis: Incomplete removal of libnsl
1038 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1039 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1040 [22715] dynamic-link: FAIL: elf/tst-audit10
1041 [22719] libc: Backtrace tests fail on hppa
1042 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1043 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1044 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1045 before the first call to crypt_r () in crypt/badsalttest.c
1046
1047 \f
1048 Version 2.26
1049
1050 Major new features:
1051
1052 * A per-thread cache has been added to malloc. Access to the cache requires
1053 no locks and therefore significantly accelerates the fast path to allocate
1054 and free small amounts of memory. Refilling an empty cache requires locking
1055 the underlying arena. Performance measurements show significant gains in a
1056 wide variety of user workloads. Workloads were captured using a special
1057 instrumented malloc and analyzed with a malloc simulator. Contributed by
1058 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1059
1060 * Unicode 10.0.0 Support: Character encoding, character type info, and
1061 transliteration tables are all updated to Unicode 10.0.0, using
1062 generator scripts contributed by Mike FABIAN (Red Hat).
1063 These updates cause user visible changes, especially the changes in
1064 wcwidth for many emoji characters cause problems when emoji sequences
1065 are rendered with pango, see for example:
1066 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1067
1068 * Collation of Hungarian has been overhauled and is now consistent with "The
1069 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1070 Egmont Koblinger.
1071
1072 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1073
1074 - The GNU C Library will now detect when /etc/resolv.conf has been
1075 modified and reload the changed configuration. The new resolver option
1076 “no-reload” (RES_NORELOAD) disables this behavior.
1077
1078 - The GNU C Library now supports an arbitrary number of search domains
1079 (configured using the “search” directive in /etc/resolv.conf);
1080 previously, there was a hard limit of six domains. For backward
1081 compatibility, applications that directly modify the ‘_res’ global
1082 object are still limited to six search domains.
1083
1084 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1085 Library will now randomly pick a name server from the configuration as a
1086 starting point. (Previously, the second name server was always used.)
1087
1088 * The tunables feature is now enabled by default. This allows users to tweak
1089 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1090
1091 * New function reallocarray, which resizes an allocated block (like realloc)
1092 to the product of two sizes, with a guaranteed clean failure upon integer
1093 overflow in the multiplication. Originally from OpenBSD, contributed by
1094 Dennis Wölfing and Rüdiger Sonderfeld.
1095
1096 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1097 These are extended versions of preadv and pwritev, respectively, taking an
1098 additional flags argument. The set of supported flags depends on the
1099 running kernel; full support currently requires kernel 4.7 or later.
1100
1101 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1102 create a new session ID for the spawned process. This feature is
1103 scheduled to be added to the next major revision of POSIX; for the time
1104 being, it is available under _GNU_SOURCE.
1105
1106 * errno.h is now safe to use from C-preprocessed assembly language on all
1107 supported operating systems. In this context, it will only define the
1108 Exxxx constants, as preprocessor macros expanding to integer literals.
1109
1110 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1111 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1112 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1113 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1114
1115 To compile programs that use this feature, the compiler must support
1116 128-bit floating point with the type name _Float128 (as defined by TS
1117 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1118 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1119 must be defined to make the new interfaces visible.
1120
1121 The new functions and macros correspond to those present for other
1122 floating-point types (except for a few obsolescent interfaces not
1123 supported for the new type), with F128 or f128 suffixes; for example,
1124 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1125 printf or scanf formats for the new type; the strfromf128 and strtof128
1126 interfaces should be used instead.
1127
1128 Deprecated and removed features, and other changes affecting compatibility:
1129
1130 * The synchronization that pthread_spin_unlock performs has been changed to
1131 now be equivalent to a C11 atomic store with release memory order to the
1132 spin lock's memory location. Previously, several (but not all)
1133 architectures used stronger synchronization (e.g., containing what is
1134 often called a full barrier). This change can improve performance, but
1135 may affect odd fringe uses of spin locks that depend on the previous
1136 behavior (e.g., using spin locks as atomic variables to try to implement
1137 Dekker's mutual exclusion algorithm).
1138
1139 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1140 removed.
1141
1142 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1143 will only be built and installed when the GNU C Library is configured with
1144 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1145 as TIRPC or rpcsvc-proto, to be used.
1146
1147 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1148 libnss_compat, are deprecated, and will not be built or installed by
1149 default.
1150
1151 The NIS(+) support library, libnsl, is also deprecated. By default, a
1152 compatibility shared library will be built and installed, but not headers
1153 or development libraries. Only a few NIS-related programs require this
1154 library. (In particular, the GNU C Library has never required programs
1155 that use 'gethostbyname' to be linked with libnsl.)
1156
1157 Replacement implementations based on TIRPC, which additionally support
1158 IPv6, are available from <https://github.com/thkukuk/>. The configure
1159 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1160 name service modules, to be built and installed.
1161
1162 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1163 support is enabled, the configured recursive resolver must support EDNS.
1164 (Responding to EDNS-enabled queries with responses which are not
1165 EDNS-enabled is fine, but FORMERR responses are not.)
1166
1167 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1168 servers have not supported this opcode for a long time.
1169
1170 * The _res_opcodes variable has been removed from libresolv. It had been
1171 exported by accident.
1172
1173 * <string.h> no longer includes inline versions of any string functions,
1174 as this kind of optimization is better done by the compiler. The macros
1175 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1176
1177 * The nonstandard header <xlocale.h> has been removed. Most programs should
1178 use <locale.h> instead. If you have a specific need for the definition of
1179 locale_t with no other declarations, please contact
1180 libc-alpha@sourceware.org and explain.
1181
1182 * The obsolete header <sys/ultrasound.h> has been removed.
1183
1184 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1185
1186 * The obsolete function cfree has been removed. Applications should use
1187 free instead.
1188
1189 * The stack_t type no longer has the name struct sigaltstack. This changes
1190 the C++ name mangling for interfaces involving this type.
1191
1192 * The ucontext_t type no longer has the name struct ucontext. This changes
1193 the C++ name mangling for interfaces involving this type.
1194
1195 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1196 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1197 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1198 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1199 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1200 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1201 fpregset_t type no longer has the name struct fpu. This changes the C++
1202 name mangling for interfaces involving those types.
1203
1204 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1205 synced with the kernel:
1206
1207 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1208 are not supported on this architecture and have been removed.
1209
1210 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1211 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1212 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1213
1214 Programs that assume the GET/SETREGS ptrace requests are universally
1215 available will now fail to build, instead of malfunctioning at runtime.
1216
1217 Changes to build and runtime requirements:
1218
1219 * Linux kernel 3.2 or later is required at runtime, on all architectures
1220 supported by that kernel. (This is a change from version 2.25 only for
1221 x86-32 and x86-64.)
1222
1223 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1224
1225 * On most architectures, GCC 4.9 or later is required to build the GNU C
1226 Library. On powerpc64le, GCC 6.2 or later is required.
1227
1228 Older GCC versions and non-GNU compilers are still supported when
1229 compiling programs that use the GNU C Library. (We do not know exactly
1230 how old, and some GNU extensions to C may be _de facto_ required. If you
1231 are interested in helping us make this statement less vague, please
1232 contact libc-alpha@sourceware.org.)
1233
1234 Security related changes:
1235
1236 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1237 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1238
1239 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1240 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1241
1242 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1243 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1244 (CVE-2010-3192).
1245
1246 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1247 fixed (CVE-2017-12133).
1248
1249 The following bugs are resolved with this release:
1250
1251 [984] network: Respond to changed resolv.conf in gethostbyname
1252 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1253 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1254 fields
1255 [12189] libc: __stack_chk_fail should not attempt a backtrace
1256 (CVE-2010-3192)
1257 [14096] time: Race condition on timezone/tst-timezone.out
1258 [14172] localedata: az_IR: new locale
1259 [14995] build: glibc fails to build if gold is the default linker, even if
1260 ld.bfd is available
1261 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1262 [16637] network: inet_pton function is accepting IPv6 with bad format
1263 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1264 x86 and other generic code
1265 [16875] localedata: ko_KR: fix lang_name
1266 [17225] localedata: ar_SY: localized month names for May and June are
1267 incorrect
1268 [17297] localedata: da_DK: wrong date_fmt string
1269 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1270 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1271 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1272 [19066] localedata: ar_SA abbreviated day and month names are in English
1273 [19569] network: resolv: Support an arbitrary number of search domains
1274 [19570] network: Implement random DNS server selection in the stub
1275 resolver
1276 [19838] locale: localedef fails on PA-RISC
1277 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1278 order of 0D36 and 0D37
1279 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1280 chillu characters
1281 [20098] libc: FAIL: debug/backtrace-tst on hppa
1282 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1283 receiving data
1284 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1285 whitespace
1286 [20313] localedata: Update locale data to Unicode 9.0
1287 [20424] manual: Document how to provide a malloc replacement
1288 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1289 for Peru
1290 [20686] locale: Add el_GR@euro to SUPPORTED.
1291 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1292 failures consistently
1293 [21015] dynamic-link: Document and fix --enable-bind-now
1294 [21016] nptl: pthread_cond support is broken on hppa
1295 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1296 frame-pointer on i386
1297 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1298 register
1299 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1300 [21088] libc: Build fails with --enable-static-nss
1301 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1302 generic c code is used
1303 [21109] libc: Tunables broken on big-endian
1304 [21112] math: powf has large ulp errors with base close to 1 and exponent
1305 around 4000
1306 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1307 (CVE-2017-12133)
1308 [21120] malloc: glibc malloc is incompatible with GCC 7
1309 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1310 with -lieee
1311 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1312 (-0.0) when linking with -lieee
1313 [21171] math: log10, log2 and lgamma return incorrect results
1314 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1315 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1316 [21207] localedata: ce_RU: update weekdays from CLDR
1317 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1318 [21217] localedata: Update months from CLDR-31
1319 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1320 [21243] libc: support_delete_temp_file should issue warning for failed
1321 remove()
1322 [21244] libc: support resolv_test_start() socket fd close should be
1323 checked for errors.
1324 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1325 new posix_spawn implementation
1326 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1327 leads to lower CPU frequency
1328 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1329 [21261] libc: [sparc64] bits/setjmp.h namespace
1330 [21267] network: [mips] bits/socket.h IOC* namespace
1331 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1332 [21270] libc: mmap64 silently truncates large offset values
1333 [21275] libc: posix_spawn always crashes on ia64 now
1334 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1335 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1336 [21289] libc: Incorrect declaration for 32-bit platforms with
1337 _FILE_OFFSET_BITS=64 causes build error
1338 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1339 not support gethostbyname4_r
1340 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1341 switching
1342 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1343 arenas
1344 [21340] libc: Support POSIX_SPAWN_SETSID
1345 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1346 [21359] network: ns_name_pack needs additional byte in destination buffer
1347 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1348 against fragmentation attacks (CVE-2017-12132)
1349 [21369] network: resolv: Remove EDNS fallback
1350 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1351 and _POSIX_C_SOURCE
1352 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1353 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1354 [21393] stdio: Missing dup3 error check in freopen, freopen64
1355 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1356 [21399] localedata: Bad description for U00EC in
1357 localedata/charmaps/CP1254
1358 [21411] malloc: realloc documentation error
1359 [21426] network: sys/socket.h uio.h namespace
1360 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1361 [21445] libc: signal.h bsd_signal namespace
1362 [21455] network: Network headers stdint.h namespace
1363 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1364 retry value
1365 [21475] network: resolv: Overlong search path is truncated mid-label
1366 [21511] libc: sigstack namespace
1367 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1368 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1369 bits/syscall.h: No such file or directory
1370 [21517] libc: struct sigaltstack namespace
1371 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1372 [21533] localedata: Update locale data to Unicode 10.0
1373 [21537] libc:
1374 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1375 junk at end of line, first unrecognized character is `@'
1376 [21538] libc: SIG_HOLD missing for XPG4
1377 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1378 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1379 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1380 [21543] libc: sigevent namespace
1381 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1382 size for general purpose registers in mcontext_t structure
1383 [21550] libc: sigwait namespace
1384 [21552] libc: XPG4 bsd_signal namespace
1385 [21554] libc: sigpause namespace
1386 [21560] libc: sys/wait.h signal.h namespace
1387 [21561] libc: waitid namespace
1388 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1389 '/usr/include/stdlib.h': Permission denied
1390 [21575] libc: sys/wait.h missing struct rusage definition
1391 [21584] libc: sigaltstack etc namespace
1392 [21597] libc: siginterrupt namespace
1393 [21607] math: hppa: FAIL: math/test-tgmath
1394 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1395 issue on x86-64
1396 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1397 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1398 stack and heap (CVE-2017-1000366)
1399 [21625] libc: wait3 namespace
1400 [21654] nss: Incorrect pointer alignment in NSS group merge result
1401 construction
1402 [21657] network: Parse interface zone id for node-local multicast
1403 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1404 between_2_3
1405 [21666] libc: .symver is used on common symbol
1406 [21668] network: resolv: res_init cross-thread broadcast introduces race
1407 conditions
1408 [21687] math: tgmath.h totalorder, totalordermag return type
1409 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1410 Hindi Locale
1411 [21696] libc: Incorrect assumption of of __cpu_mask in
1412 posix/sched_cpucount.c
1413 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1414 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1415 locale
1416 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1417 leading to relocation crash
1418 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1419 `resolv_conf_matches (resp, conf)' failed.
1420 [21710] localedata: Added Samoan language locale for Samoa
1421 [21711] localedata: Pashto yesstr/nostr locale are missing
1422 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1423 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1424 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1425 [21723] localedata: yesstr/nostr missing for Chinese language locale
1426 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1427 locale
1428 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1429 locale
1430 [21728] localedata: New Locale for Tongan language
1431 [21729] localedata: incorrect LC_NAME fields for hi_IN
1432 [21733] localedata: yesstr and nostr are missing for zh_HK
1433 [21734] localedata: Missing yesstr and nostr are for kw_GB
1434 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1435 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1436 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1437 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1438 strings
1439 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1440 protector=all
1441 [21749] localedata: Wrong abbreviated day name (“abday”) for
1442 ar_JO/ar_LB/ar_SY
1443 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1444 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1445 [21759] localedata: missing yesstr and nostr for Tigrinya
1446 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1447 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1448 [21767] localedata: Missing Bislama locales
1449 [21768] localedata: Missing yesstr and nostr for aa_ET
1450 [21770] localedata: Missing Field in li_NL
1451 [21778] nptl: Robust mutex may deadlock
1452 [21779] libc: MicroBlaze segfaults when loading libpthread
1453 [21783] localedata: Fix int_select international_call_prefixes
1454 [21784] localedata: Inconsistency in country_isbn
1455 [21788] localedata: Missing Country Postal Abbreviations
1456 [21794] localedata: Added-country_isbn-for-Italy
1457 [21795] localedata: Add/Fix country_isbn for France
1458 [21796] localedata: Added country_isbn for Republic of Korea
1459 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1460 [21799] localedata: Added int_select international_call_prefixes
1461 [21801] localedata: Added int_select international_call_prefixes
1462 [21804] nptl: Double semicolon in thread-shared-types.h
1463 [21807] localedata: LC_ADDRESS fix for pap_CW
1464 [21808] localedata: Fix LC_ADDRESS for pap_AW
1465 [21821] localedata: Added country_name in mai_IN
1466 [21822] localedata: Fix LC_TIME for mai_IN
1467 [21823] localedata: missing yesstr, nostr for sa_IN
1468 [21825] localedata: Fix name_mrs for mag_IN
1469 [21828] localedata: 2.26 changelog should mention user visible changes
1470 with unicode 9.0
1471 [21835] localedata: Added Maithili language locale for Nepal
1472 [21838] localedata: Removed redundant data for the_NP
1473 [21839] localedata: Fix LC_MONETARY for ta_LK
1474 [21844] localedata: Fix Latin characters and Months Sequence.
1475 [21848] localedata: Fix mai_NP Title Name
1476
1477 \f
1478 Version 2.25
1479
1480 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1481 24731-2:2010, is supported to enable declarations of functions from that
1482 TR. Note that not all functions from that TR are supported by the GNU C
1483 Library.
1484
1485 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1486 18661-1:2014, is supported to enable declarations of functions and macros
1487 from that TS. Note that not all features from that TS are supported by
1488 the GNU C Library.
1489
1490 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1491 18661-4:2015, is supported to enable declarations of functions and macros
1492 from that TS. Note that most features from that TS are not supported by
1493 the GNU C Library.
1494
1495 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1496 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1497 Since the GNU C Library defaults to a much newer revision of POSIX, this
1498 will only affect programs that specifically request an old conformance
1499 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1500 see a change in the visible declarations, but a program compiled with
1501 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1502 will not.
1503
1504 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1505 defined by all multithreaded code, but glibc has not required this for
1506 many years.
1507
1508 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1509 means that in a future release, the macros “major”, “minor”, and “makedev”
1510 will only be available from <sys/sysmacros.h>.
1511
1512 These macros are not part of POSIX nor XSI, and their names frequently
1513 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1514 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1515 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1516 problem.
1517
1518 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1519 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1520 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1521
1522 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1523 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1524 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1525 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1526 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1527 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1528 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1529 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1530 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1531 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1532 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1533 WINT_WIDTH.
1534
1535 * New <math.h> features are added from TS 18661-1:2014:
1536
1537 - Signaling NaN macros: SNANF, SNAN, SNANL.
1538
1539 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1540 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1541 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1542
1543 - llogb functions: the llogb, llogbf and llogbl functions, and the
1544 FP_LLOGB0 and FP_LLOGBNAN macros.
1545
1546 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1547 fminmagf, fminmagl.
1548
1549 - Comparison macros: iseqsig.
1550
1551 - Classification macros: iscanonical, issubnormal, iszero.
1552
1553 - Total order functions: totalorder, totalorderf, totalorderl,
1554 totalordermag, totalordermagf, totalordermagl.
1555
1556 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1557
1558 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1559 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1560
1561 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1562 are added to libc. They convert a floating-point number into string.
1563
1564 * Most of glibc can now be built with the stack smashing protector enabled.
1565 It is recommended to build glibc with --enable-stack-protector=strong.
1566 Implemented by Nick Alcock (Oracle).
1567
1568 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1569 intended to be used instead of memset() to erase sensitive data after use;
1570 the compiler will not optimize out calls to explicit_bzero even if they
1571 are "unnecessary" (in the sense that no _correct_ program can observe the
1572 effects of the memory clear).
1573
1574 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1575 to float instead of double. This does not affect the ABI of any libraries
1576 that are part of the GNU C Library, but may affect the ABI of other
1577 libraries that use this type in their interfaces.
1578
1579 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1580 float_t and double_t types are now defined to long double instead of float
1581 and double. These options are not the default, and this does not affect
1582 the ABI of any libraries that are part of the GNU C Library, but it may
1583 affect the ABI of other libraries that use this type in their interfaces,
1584 if they are compiled or used with those options.
1585
1586 * The getentropy and getrandom functions, and the <sys/random.h> header file
1587 have been added.
1588
1589 * The buffer size for byte-oriented stdio streams is now limited to 8192
1590 bytes by default. Previously, on Linux, the default buffer size on most
1591 file systems was 4096 bytes (and thus remains unchanged), except on
1592 network file systems, where the buffer size was unpredictable and could be
1593 as large as several megabytes.
1594
1595 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1596 for the Linux quota interface which predates kernel version 2.4.22 has
1597 been removed.
1598
1599 * The malloc_get_state and malloc_set_state functions have been removed.
1600 Already-existing binaries that dynamically link to these functions will
1601 get a hidden implementation in which malloc_get_state is a stub. As far
1602 as we know, these functions are used only by GNU Emacs and this change
1603 will not adversely affect already-built Emacs executables. Any undumped
1604 Emacs executables, which normally exist only during an Emacs build, should
1605 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1606
1607 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1608 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1609 “no-ip6-dotint” had already been the default, and support for the
1610 “ip6-dotint” option was removed from the Internet in 2006.
1611
1612 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1613 flag from <resolv.h> have been removed. The option relied on a
1614 backwards-incompatible DNS extension which was never deployed on the
1615 Internet.
1616
1617 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1618 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1619 They were already unimplemented.
1620
1621 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1622 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1623 removed again from the IETF name lookup interface specification in RFC
1624 2553. Applications should use getaddrinfo instead.
1625
1626 * DNSSEC-related declarations and definitions have been removed from the
1627 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1628 decode the data part of DNSSEC record types. Previous versions of glibc
1629 only implemented minimal support for the previous version of DNSSEC, which
1630 is incompatible with the currently deployed version.
1631
1632 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1633 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1634 <arpa/nameser.h> header file because the distinction between RR types and
1635 meta-RR types is not officially standardized, subject to revision, and
1636 thus not suitable for encoding in a macro.
1637
1638 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1639 and rhook members of the res_state type in <resolv.h> have been removed.
1640 The glibc stub resolver did not support these hooks, but the header file
1641 did not reflect that.
1642
1643 * For multi-arch support it is recommended to use a GCC which has
1644 been built with support for GNU indirect functions. This ensures
1645 that correct debugging information is generated for functions
1646 selected by IFUNC resolvers. This support can either be enabled by
1647 configuring GCC with '--enable-gnu-indirect-function', or by
1648 enabling it by default by setting 'default_gnu_indirect_function'
1649 variable for a particular architecture in the GCC source file
1650 'gcc/config.gcc'.
1651
1652 * GDB pretty printers have been added for mutex and condition variable
1653 structures in POSIX Threads. When installed and loaded in gdb these pretty
1654 printers show various pthread variables in human-readable form when read
1655 using the 'print' or 'display' commands in gdb.
1656
1657 * Tunables feature added to allow tweaking of the runtime for an application
1658 program. This feature can be enabled with the '--enable-tunables' configure
1659 flag. The GNU C Library manual has details on usage and README.tunables has
1660 instructions on adding new tunables to the library.
1661
1662 * A new version of condition variables functions have been implemented in
1663 the NPTL implementation of POSIX Threads to provide stronger ordering
1664 guarantees.
1665
1666 * A new version of pthread_rwlock functions have been implemented to use a more
1667 scalable algorithm primarily through not using a critical section anymore to
1668 make state changes.
1669
1670 Security related changes:
1671
1672 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1673 have been created with makecontext could fail to terminate due to a
1674 missing .cantunwind annotation. This has been observed to lead to a hang
1675 (denial of service) in some Go applications compiled with gccgo. Reported
1676 by Andreas Schwab. (CVE-2016-6323)
1677
1678 * The DNS stub resolver functions would crash due to a NULL pointer
1679 dereference when processing a query with a valid DNS question type which
1680 was used internally in the implementation. The stub resolver now uses a
1681 question type which is outside the range of valid question type values.
1682 (CVE-2015-5180)
1683
1684 The following bugs are resolved with this release:
1685
1686 [4099] stdio: Overly agressive caching by stream i/o functions.
1687 [7065] build: Support building glibc with -fstack-protector or -fstack-
1688 protector-all
1689 [9842] localedata: en_CA: incorrect date format
1690 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1691 before it started waiting
1692 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1693 [15765] nptl: sem_open is wrongly a cancellation point
1694 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1695 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1696 wrong type
1697 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1698 library linked with pthread
1699 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1700 "FIXME: Ingo" issue)
1701 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1702 [17252] libc: getrandom and getentropy syscall
1703 [17730] malloc: thread-local storage is sometimes improperly free()'d
1704 after being __libc_memalign()'d
1705 [18241] stdio: failed fseek on memstream does not set errno and can fail
1706 when it shouldnt
1707 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1708 not
1709 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1710 PTHREAD_PRIO_INHERIT mutex on ARM
1711 [18784] network: res_query and related function crash for special record
1712 type queries (CVE-2015-5180)
1713 [19380] math: strtod does not raise "inexact"
1714 [19387] string: Integer overflow in memchr
1715 [19390] string: Integer overflow in strncat
1716 [19398] build: linknamespace tests fail with massively parallel build
1717 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1718 termination
1719 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1720 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1721 [19514] libc: [PATCH] Fix spelling errors in spelling
1722 "implement"/"implementation" in several places
1723 [19582] network: Deprecate RES_USE_INET6
1724 [19673] manual: clog10 docs appear to be erroneous
1725 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1726 causes a segmentation fault
1727 [19826] libc: invalid pointer returned from __tls_get_addr with static
1728 linking
1729 [20016] network: resolv: Remove hooks support from the API
1730 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1731 uninitialized GOT
1732 [20033] math: [x86_64] vectorized math function don't call the __finite
1733 versions
1734 [20116] nptl: use after free in pthread_create
1735 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1736 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1737 is always true.
1738 [20311] nptl: please install proc_service.h
1739 [20366] libc: Compilation errors in installed headers in strict-compliance
1740 modes
1741 [20370] malloc: malloc: Arena free list management is still racy
1742 (incorrect fix in bug 19243)
1743 [20386] libc: assert (X = 0) does not result in GCC warning
1744 [20432] malloc: malloc: Minimize interface required for interposition
1745 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1746 in _Unwind_Backtrace (CVE-2016-6323)
1747 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1748 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1749 libpthread.a
1750 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1751 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1752 [20477] network: resolv: incorrect double-checked locking related to
1753 _res_hconf
1754 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1755 information.
1756 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1757 penalty
1758 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1759 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1760 cause transition penalty
1761 [20517] math: sparcv9 missing fdiml compat symbol
1762 [20524] manual: strverscmp is inconsistent
1763 [20525] libc: <sys/quota.h> should be based on kernel headers
1764 [20539] math: GCC 7 -static -lm fails to link at -O3
1765 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1766 mcount.oS)
1767 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1768 Checking
1769 [20591] network: Remove obsolete DNSSEC support
1770 [20592] network: DNS resource record type classification macros in
1771 <arpa/nameser.h> are incorrect
1772 [20593] network: Update DNS RR type definitions
1773 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1774 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1775 multi-arch
1776 [20629] network: libresolv: Remove support for bitstring labels
1777 (RES_USEBSTRING)
1778 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1779 assertion: "maxidx >= 2"
1780 [20660] math: [arm] Use VSQRT
1781 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1782 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1783 [20689] libc: Test for FMA should also check for AVX.
1784 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1785 [20715] math: iszero macro breaks existing code
1786 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1787 [20728] libc: powerpc: Missing TOC stub in clone
1788 [20729] build: glibc-2.24 fails to build for i486 with -Os
1789 [20750] build: Build fails with default PIE enabled gcc-6.x
1790 [20768] math: [alpha] sqrt fegetenv namespace
1791 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1792 glibc
1793 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1794 __FLT_EVAL_METHOD__ is 2
1795 [20790] network: rpcgen buffer overrun in get_prog_declaration
1796 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1797 [20829] libc: crypt snprintf namespace
1798 [20847] libc: tst-vfork3 failure
1799 [20855] math: Default bits/mathdef.h has inappropriate float_t
1800 [20859] math: [sh4] FP_ILOGB0 invalid
1801 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1802 U+20AC), not same as GBK
1803 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1804 [20916] math: pow handling of sNaN arguments
1805 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1806 [20919] math: Bad pow (qNaN, 0) result with -lieee
1807 [20940] math: hypot sNaN handling
1808 [20947] math: fmax, fmin sNaN handling
1809 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1810 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1811 RES_USE_INET6
1812 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1813 [20973] nptl: robust mutexes: Lost wake-ups
1814 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1815 [20978] nis: strlen on null pointer in nss_nisplus
1816 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1817 wrong condition
1818 [21014] string: i686 memchr overflows internal pointer check
1819 [21019] libc: [mips] n32 lseek incorrect on overflow
1820 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1821 [21026] libc: [MIPS] readahead syscall is broken on n64
1822 [21028] math: Fallback fesetexceptflag should always succeed
1823 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1824 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1825 clang
1826 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1827 [21061] librt: [microblaze] librt lost clock_* exports
1828 [21073] libc: tunables: insecure environment variables passed to
1829 subprocesses with AT_SECURE
1830 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1831 \f
1832 Version 2.24
1833
1834 * The minimum Linux kernel version that this version of the GNU C Library
1835 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1836 kernel version 2.6.32 or later suffices (on architectures that already
1837 required kernel versions more recent than 3.2, those requirements remain
1838 unchanged). Linux 3.2 or later kernel headers are required on all
1839 architectures.
1840
1841 * The pap_AN locale has been deleted. This has been deprecated for a long
1842 time. It has been replaced by pap_AW & pap_CW, both of which have long
1843 been included in previous releases.
1844
1845 * The readdir_r and readdir64_r functions have been deprecated. It is
1846 recommended to use readdir and readdir64 instead.
1847
1848 * The type “union wait” has been removed. It was deprecated in the early
1849 1990s and never part of POSIX. Application code should use the int type
1850 instead of “union wait”.
1851
1852 * A new NSS action is added to facilitate large distributed system
1853 administration. The action, MERGE, allows remote user stores like LDAP
1854 to be merged into local user stores like /etc/groups in order to provide
1855 easy to use, updated, and managed sets of merged credentials. The new
1856 action can be used by configuring it in /etc/nsswitch.conf:
1857 group: files [SUCCESS=merge] nis
1858 Implemented by Stephen Gallagher (Red Hat).
1859
1860 * The deprecated __malloc_initialize_hook variable has been removed from the
1861 API.
1862
1863 * The long unused localedef --old-style option has been removed. It hasn't
1864 done anything in over 16 years. Scripts using this option can safely
1865 drop it.
1866
1867 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1868 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1869 return the next representable value in the direction of positive infinity
1870 and the nextdown functions return the next representable value in the
1871 direction of negative infinity. These are currently enabled as GNU
1872 extensions.
1873
1874 Security related changes:
1875
1876 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1877 could result in a stack overflow when getnetbyname was called with an
1878 overly long name. (CVE-2016-3075)
1879
1880 * Previously, getaddrinfo copied large amounts of address data to the stack,
1881 even after the fix for CVE-2013-4458 has been applied, potentially
1882 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1883 instead. Reported by Michael Petlan. (CVE-2016-3706)
1884
1885 * The glob function suffered from a stack-based buffer overflow when it was
1886 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1887 Reported by Alexander Cherepanov. (CVE-2016-1234)
1888
1889 * The Sun RPC UDP client could exhaust all available stack space when
1890 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1891 alloca plugin for GCC. (CVE-2016-4429)
1892
1893 * The IPv6 name server management code in libresolv could result in a memory
1894 leak for each thread which is created, performs a failing naming lookup,
1895 and exits. Over time, this could result in a denial of service due to
1896 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1897
1898 The following bugs are resolved with this release:
1899
1900 [1170] localedata: ne_NP: update Nepali locale definition file
1901 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1902 of MS-DOS.
1903 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1904 [6796] math: fdim() does not set errno on overflow
1905 [10354] libc: posix_spawn should use vfork() in more cases than presently
1906 [11213] localedata: localedata: add copyright disclaimer to locale files
1907 [12143] localedata: chr_US: new Cherokee locale
1908 [12450] localedata: sgs_LT: new locale
1909 [12676] localedata: ln_CD: new locale
1910 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1911 CLDR data
1912 [13304] math: fma, fmaf, fmal produce wrong results
1913 [14259] build: --localedir arg to configure is ignored
1914 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1915 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1916 [14934] localedata: es_CL: wrong first weekday chilean locale
1917 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1918 romanisation
1919 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1920 and +/-
1921 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1922 [15368] nptl: raise() is not async-signal-safe
1923 [15479] math: ceil, floor, round and trunc raise inexact exception
1924 [15578] localedata: kk_KZ: various updates
1925 [16003] localedata: pap_AN: punt old locale
1926 [16137] localedata: iw_IL: punt old locale
1927 [16190] localedata: eo: new esperanto locale
1928 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1929 [16742] malloc: race condition: pthread_atfork() called before first
1930 malloc() results in unexpected locking behaviour/deadlocks
1931 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1932 all locales
1933 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1934 [17565] localedata: pt_PT: wrong (work-)week start
1935 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1936 [17950] build: Build fails with -msse
1937 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1938 [18433] libc: posix_spawn does not return correctly upon failure to
1939 execute
1940 [18453] localedata: charmaps/IBM875: incorrect codes
1941 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1942 -Wsystem-headers
1943 [18896] localedata: he_IL: improvements for currency
1944 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1945 Romanian locale data
1946 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1947 symbol
1948 [19056] libc: Deprecate readdir_r
1949 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1950 language
1951 [19198] localedata: nl_NL: small improvements for Dutch locales
1952 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1953 nameservers (CVE-2016-5417)
1954 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1955 machine
1956 [19400] locale: Language missing in "iso-639.def", trivial fix in
1957 description
1958 [19431] malloc: Deadlock between fflush, getdelim, and fork
1959 [19505] libc: Incorrect file descriptor validity checks in
1960 posix_spawn_file_actions_add{open,close,dup2}
1961 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1962 when using RTLD_NEXT
1963 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1964 `intl/{gettextP,loadinfo}.h'
1965 [19534] libc: execle, execlp may use malloc
1966 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1967 [19573] network: res_nclose and __res_maybe_init disagree about name
1968 server initialization, breaking Hesiod
1969 [19575] localedata: Status of GB18030 tables
1970 [19581] localedata: sr_* date_fmt string contains additional newline
1971 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1972 Excavator core
1973 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1974 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1975 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1976 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1977 double range
1978 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1979 part zero incorrect
1980 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1981 equality tests
1982 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1983 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1984 [19633] locale: strfmon_l applies global locale to number formatting
1985 [19642] network: Memory leak in getnameinfo
1986 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1987 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1988 glibc-2.22
1989 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1990 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1991 'tst-numeric.c'
1992 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1993 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1994 low part
1995 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1996 result
1997 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1998 unaligned stack
1999 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2000 pointers and lengths in error-case.
2001 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2002 report errors on UTF-16 surrogates values.
2003 [19755] nscd: nscd assertion failure in gc
2004 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2005 [19759] libc: mempcpy shouldn't be inlined
2006 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2007 [19765] libc: s390 needs an optimized mempcpy
2008 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2009 NAME_MAX limit assumption (CVE-2016-1234)
2010 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2011 [19787] network: Missing and incorrect truncation checks in getnameinfo
2012 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2013 modes
2014 [19791] network: Assertion failure in res_query.c with un-connectable name
2015 server addresses
2016 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2017 [19822] math: libm.so install clobbers old version
2018 [19825] network: resolv: send_vc can return uninitialized data in second
2019 response to getaddrinfo
2020 [19830] network: nss_dns: should check RDATA length against buffer length
2021 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2022 confronted with A/AAAA records of invalid size
2023 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2024 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2025 [19853] stdio: Printing IBM long double in decimal with high precision is
2026 sometimes incorrect
2027 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2028 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2029 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2030 record types
2031 [19865] network: Assertion failure or memory leak in
2032 _nss_dns_getcanonname_r
2033 [19868] network: nss_dns: netent code does not skip over non-PTR records
2034 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2035 (CVE-2016-3075)
2036 [19881] string: Improve x86-64 memset
2037 [19907] string: Incorrect memcpy tests
2038 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2039 resolving symbols
2040 [19925] libc: termios.h XCASE namespace
2041 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2042 [19929] libc: limits.h NL_NMAX namespace
2043 [19931] stdio: Memory leak in vfprintf
2044 [19957] libc: clone(CLONE_VM) access invalid parent memory
2045 [19963] localedata: en_IL: New locale
2046 [19989] stdio: stdio.h cuserid namespace
2047 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2048 gethosts
2049 [19996] locale: langinfo.h nl_langinfo_l namespace
2050 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2051 [20010] network: getaddrinfo: Stack overflow in hostent translation
2052 (CVE-2016-3706)
2053 [20012] stdio: libio: fmemopen append mode failure
2054 [20014] stdio: stdio.h namespace for pre-threads POSIX
2055 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2056 [20023] libc: fcntl.h timespec namespace
2057 [20024] math: [x86_64] vectorized sincos trashes the stack
2058 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2059 [20041] time: sys/time.h timespec namespace
2060 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2061 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2062 [20051] libc: ttyslot in wrong header under wrong conditions
2063 [20054] libc: gethostname not declared for XPG4
2064 [20055] libc: termios.h missing tcgetsid for XPG4
2065 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2066 executable
2067 [20073] libc: sys/stat.h fchmod namespace
2068 [20074] libc: stdlib.h rand_r namespace
2069 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2070 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2071 XPG3
2072 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2073 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2074 (CVE-2016-4429)
2075 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2076 [20119] libc: Wrong mask for processors level type from CPUID
2077 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2078 AS not supporting AVX512
2079 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2080 sNaN argument
2081 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2082 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2083 argument
2084 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2085 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2086 [20178] libc: posix_spawn{p} should not call exit
2087 [20191] stdio: libio: vtables hardening
2088 [20195] string: FMA4 detection requires CPUID execution with register
2089 eax=0x80000001
2090 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2091 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2092 subnormals
2093 [20212] math: acos (sNaN) returns sNaN
2094 [20213] math: asin (sNaN) returns sNaN
2095 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2096 [20218] math: [i386] asinhl (sNaN) returns sNaN
2097 [20219] math: [i386] atanhl (sNaN) returns sNaN
2098 [20222] stdio: fopencookie: Mangle function pointers
2099 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2100 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2101 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2102 input
2103 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2104 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2105 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2106 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2107 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2108 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2109 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2110 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2111 [20240] math: modf (sNaN) returns sNaN
2112 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2113 [20250] math: frexp (sNaN) returns sNaN
2114 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2115 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2116 double rounding
2117 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2118 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2119 subscript is below array bounds [-Werror=array-bounds]
2120 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2121 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2122 (Only arm/linux)
2123 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2124 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2125 fallbacks
2126 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2127 "invalid" exceptions
2128 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2129 [20316] localedata: id_ID: Februari instead of Pebruari
2130 [20327] string: POWER8 strcasecmp returns incorrect result
2131 [20347] math: Failure: Test: j0_downward (0xap+0)
2132 [20348] libc: FAIL: misc/tst-preadvwritev64
2133 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2134 [20350] libc: There is no test for p{read,write}64
2135 [20357] math: Incorrect cos result for 1.5174239687223976
2136 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2137 \f
2138 Version 2.23
2139
2140 * Unicode 8.0.0 Support: Character encoding, character type info, and
2141 transliteration tables are all updated to Unicode 8.0.0, using new
2142 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2143 These updates cause user visible changes, such as the fixes for bugs
2144 89, 16061, and 18568.
2145
2146 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2147 kernel-internal CPU set size. This means that requests that change the
2148 CPU affinity which failed before (for example, an all-ones CPU mask) will
2149 now succeed. Applications that need to determine the effective CPU
2150 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2151 setting it because the kernel can adjust it (and the previous size check
2152 would not detect this in the majority of cases).
2153
2154 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2155 the following new symbols are used: fts64_children, fts64_close,
2156 fts64_open, fts64_read and fts64_set.
2157
2158 * getaddrinfo now detects certain invalid responses on an internal netlink
2159 socket. If such responses are received, an affected process will
2160 terminate with an error message of "Unexpected error <number> on netlink
2161 descriptor <number>" or "Unexpected netlink response of size <number> on
2162 descriptor <number>". The most likely cause for these errors is a
2163 multi-threaded application which erroneously closes and reuses the netlink
2164 file descriptor while it is used by getaddrinfo.
2165
2166 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2167 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2168 unnecessary serialization of memory allocation requests across threads.
2169 The defect is now corrected. Users should see a substantial increase in
2170 the concurent throughput of allocation requests for applications which
2171 trigger this bug. Affected applications typically create and
2172 destroy threads frequently. (Bug 19048 was reported and analyzed by
2173 Ericsson.)
2174
2175 * There is now a --disable-timezone-tools configure option for disabling the
2176 building and installing of the timezone related utilities (zic, zdump, and
2177 tzselect). This is useful for people who build the timezone data and code
2178 independent of the GNU C Library.
2179
2180 * The obsolete header <regexp.h> has been removed. Programs that require
2181 this header must be updated to use <regex.h> instead.
2182
2183 * The obsolete functions bdflush, create_module, get_kernel_syms,
2184 query_module and uselib are no longer available to newly linked binaries;
2185 the header <sys/kdaemon.h> has been removed. These functions and header
2186 were specific to systems using the Linux kernel and could not usefully be
2187 used with the GNU C Library on systems with version 2.6 or later of the
2188 Linux kernel.
2189
2190 * Optimized string, wcsmbs and memory functions for IBM z13.
2191 Implemented by Stefan Liebler.
2192
2193 * Newly linked programs that define a variable called signgam will no longer
2194 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2195 require signgam to be set by those functions must ensure that they use the
2196 variable provided by the GNU C Library and declared in <math.h>, without
2197 defining their own copy.
2198
2199 * The minimum GCC version that can be used to build this version of the GNU
2200 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2201 still be used to compile programs using the GNU C Library.
2202
2203 Security related changes:
2204
2205 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2206 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2207
2208 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2209 the pointer guard feature. It is always enabled. Previously,
2210 LD_POINTER_GUARD could be used to disable security hardening in binaries
2211 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2212 (CVE-2015-8777)
2213
2214 * An integer overflow in hcreate and hcreate_r could lead to an
2215 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2216
2217 * The catopen function no longer has unbounded stack usage. Reported by
2218 Max. (CVE-2015-8779)
2219
2220 * The nan, nanf and nanl functions no longer have unbounded stack usage
2221 depending on the length of the string passed as an argument to the
2222 functions. Reported by Joseph Myers. (CVE-2014-9761)
2223
2224 * A stack-based buffer overflow was found in libresolv when invoked from
2225 libnss_dns, allowing specially crafted DNS responses to seize control
2226 of execution flow in the DNS client. The buffer overflow occurs in
2227 the functions send_dg (send datagram) and send_vc (send TCP) for the
2228 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2229 family. The use of AF_UNSPEC triggers the low-level resolver code to
2230 send out two parallel queries for A and AAAA. A mismanagement of the
2231 buffers used for those queries could result in the response of a query
2232 writing beyond the alloca allocated buffer created by
2233 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2234 the overflow. Thanks to the Google Security Team and Red Hat for
2235 reporting the security impact of this issue, and Robert Holiday of
2236 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2237
2238 The following bugs are resolved with this release:
2239
2240 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2241 [887] math: Math library function "logb" and "nextafter" inconsistent
2242 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2243 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2244 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2245 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2246 use `mkstemp'
2247 [4404] localedata: German translation of "Alarm clock" is misleading
2248 [6799] math: nextafter() and nexttoward() doen't set errno on
2249 overflow/underflow errors
2250 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2251 overflow/underflow
2252 [10432] nis: _nss_nis_setnetgrent assertion failure
2253 [11460] libc: fts has no LFS support
2254 [12926] network: getaddrinfo()/make_request() may spin forever
2255 [13065] nptl: Race condition in pthread barriers
2256 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2257 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2258 are not contiguous
2259 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2260 [14912] libc: Rename non-installed bits/*.h headers
2261 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2262 [15367] math: Let gcc use __builtin_isinf
2263 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2264 [15421] math: lgamma wrongly sets signgam for ISO C
2265 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2266 FE_INVALID with argument out of range
2267 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2268 all exceptions
2269 [15786] dynamic-link: ifunc resolver functions can smash function
2270 arguments
2271 [15918] math: Unnecessary check for equality in hypotf()
2272 [16061] localedata: Review / update transliteration data
2273 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2274 should include
2275 [16141] time: strptime %z offset restriction
2276 [16171] math: drem should be alias of remainder
2277 [16296] math: fegetround is pure?
2278 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2279 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2280 GNU/Linux
2281 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2282 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2283 arguments
2284 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2285 "Exception "Inexact" set" on ppc32
2286 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2287 [16517] math: Missing underflow exception from tanf/tan/tanl
2288 [16519] math: Missing underflow exception from sinhf
2289 [16520] math: Missing underflow exception from tanhf
2290 [16521] math: Missing underflow exception from exp2
2291 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2292 rounding results
2293 [16734] stdio: fopen calls mmap to allocate its buffer
2294 [16961] math: nan function incorrect handling of bad sequences
2295 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2296 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2297 [16985] locale: localedef: confusing error message when opening output
2298 fails
2299 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2300 [17197] locale: Redundant shift character in iconv conversion output at
2301 block boundary
2302 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2303 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2304 [17250] dynamic-link: static linking breaks nss loading
2305 (getaddrinfo/getpwnam/etc...)
2306 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2307 4.7?
2308 [17441] math: isnan() should use __builtin_isnan() in GCC
2309 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2310 (related to lock elision)
2311 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2312 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2313 [17887] time: strptime should be able to parse "+01:00" style timezones
2314 [17905] libc: catopen() Multiple unbounded stack allocations
2315 (CVE-2015-8779)
2316 [18084] libc: backtrace (..., 0) dumps core on x86
2317 [18086] libc: nice() sets errno to 0 on success
2318 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2319 is too large (CVE-2015-8778)
2320 [18251] dynamic-link: SONAME missing when audit modules provides path
2321 [18265] libc: add attributes for wchar string and memory functions
2322 [18370] math: csqrt missing underflows
2323 [18421] libc: [hppa] read-only segment has dynamic relocations
2324 [18472] libc: Obsolete syscall wrappers should be compat symbols
2325 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2326 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2327 [18525] localedata: Remove locale timezone information
2328 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2329 [18568] localedata: Update locale data to Unicode 8.0
2330 [18589] locale: sort-test.sh fails at random
2331 [18595] math: ctan, ctanh missing underflows
2332 [18604] libc: assert macro-expands its argument
2333 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2334 contains a vector instruction exception.
2335 [18611] math: j1, jn missing errno setting on underflow
2336 [18618] localedata: sync Chechen locale definitions with other *_RU
2337 locales
2338 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2339 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2340 [18665] network: In send_dg, the recvfrom function is NOT always using the
2341 buffer size of a newly created buffer (CVE-2015-7547)
2342 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2343 missing break ?
2344 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2345 32bit processes
2346 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2347 [18699] math: tilegx cproj() for various complex infinities does not yield
2348 infinity
2349 [18724] libc: Harden put*ent functions against data injection
2350 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2351 [18755] build: build errors with -DNDEBUG
2352 [18757] stdio: fmemopen fails to set errno on failure
2353 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2354 be forced unloaded
2355 [18781] libc: openat64 lacks O_LARGEFILE
2356 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2357 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2358 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2359 [18790] math: [ldbl-128ibm] tanhl inaccurate
2360 [18795] libc: stpncpy fortification misses buffer lengths that are
2361 statically too large
2362 [18796] build: build fails for --disable-mathvec
2363 [18803] math: hypot missing underflows
2364 [18820] stdio: fmemopen may leak memory on failure
2365 [18823] math: csqrt spurious underflows
2366 [18824] math: fma spurious underflows
2367 [18825] math: pow missing underflows
2368 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2369 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2370 [18870] build: sem_open.c fails to compile with missing symbol
2371 FUTEX_SHARED
2372 [18872] stdio: Fix memory leak in printf_positional
2373 [18873] libc: posix_fallocate overflow check ineffective
2374 [18875] math: Excess precision leads incorrect libm
2375 [18877] libc: arm: mmap offset regression
2376 [18887] libc: memory corruption when using getmntent on blank lines
2377 [18918] localedata: hu_HU: change time to HH:MM:SS format
2378 [18921] libc: Regression: extraneous stat() and fstat() performed by
2379 opendir()
2380 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2381 binaries (CVE-2015-8777)
2382 [18951] math: tgamma missing underflows
2383 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2384 signgam
2385 [18953] localedata: lt_LT: change currency symbol to the euro
2386 [18956] math: powf inaccuracy
2387 [18961] math: [i386] exp missing underflows
2388 [18966] math: [i386] exp10 missing underflows
2389 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2390 [18969] build: multiple string test failures due to missing locale
2391 dependencies
2392 [18970] libc: Reference of pthread_setcancelstate in libc.a
2393 [18977] math: float / long double Bessel functions not in XSI POSIX
2394 [18980] math: i386 libm functions return with excess range and precision
2395 [18981] math: i386 scalb*, ldexp return with excess range and precision
2396 [18982] stdio: va_list and vprintf
2397 [18985] time: Passing out of range data to strftime() causes a segfault
2398 (CVE-2015-8776)
2399 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2400 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2401 [19012] locale: iconv_open leaks memory on error path
2402 [19016] math: clog, clog10 inaccuracy
2403 [19018] nptl: Mangle function pointers in tls_dtor_list
2404 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2405 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2406 [19048] malloc: malloc: arena free list can become cyclic, increasing
2407 contention
2408 [19049] math: [powerpc] erfc incorrect zero sign
2409 [19050] math: [powerpc] log* incorrect zero sign
2410 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2411 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2412 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2413 [19074] network: Data race in _res_hconf_reorder_addrs
2414 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2415 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2416 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2417 rounding modes
2418 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2419 ILP32
2420 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2421 [19086] manual: posix_fallocate64 documented argument order is wrong.
2422 [19088] math: lround, llround missing exceptions close to overflow
2423 threshold
2424 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2425 [19095] math: dbl-64 lrint incorrect for 64-bit long
2426 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2427 [19124] dynamic-link: ld.so failed to build with older assmebler
2428 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2429 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2430 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2431 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2432 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2433 pthread_setaffinity_np
2434 [19156] math: [ldbl-128] j0l spurious underflows
2435 [19164] nptl: tst-getcpu fails with many possible CPUs
2436 [19168] math: math/test-ildoubl and math/test-ldouble failure
2437 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2438 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2439 prelink
2440 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2441 (FE_NOMASK_ENV) do not clear SSE exceptions
2442 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2443 _int_new_arena/reused_arena
2444 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2445 [19201] math: dbl-64 remainder incorrect sign of zero result
2446 [19205] math: bits/math-finite.h conditions do not match math.h and
2447 bits/mathcalls.h
2448 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2449 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2450 for C99-based standards
2451 [19212] libc: features.h not -Wundef clean
2452 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2453 math-only
2454 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2455 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2456 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2457 disabled
2458 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2459 "inexact" exceptions
2460 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2461 arguments
2462 [19242] libc: strtol incorrect in Turkish locales
2463 [19243] malloc: reused_arena can pick an arena on the free list, leading
2464 to an assertion failure and reference count corruption
2465 [19253] time: tzset() ineffective when temporary TZ did not include DST
2466 rules
2467 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2468 [19270] math: [hppa] Shared libm missing __isnanl
2469 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2470 MAP_HUGETLB and MAP_STACK defines
2471 [19313] nptl: Wrong __cpu_mask for x32
2472 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2473 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2474 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2475 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2476 [19363] time: x32: times() return value wrongly truncates/sign extends
2477 from 32bit
2478 [19367] dynamic-link: Improve branch prediction on Silvermont
2479 [19369] network: Default domain name not reset by res_ninit when "search"
2480 / "domain" entry is removed from resolv.conf
2481 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2482 subnormals
2483 [19385] localedata: bg_BG: time separator should be colon, not comma
2484 [19408] libc: linux personality syscall wrapper may erroneously return an
2485 error on 32-bit architectures
2486 [19415] libc: dladdr returns wrong names on hppa
2487 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2488 IBM905, IBM907, and IBM909
2489 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2490 [19443] build: build failures with -DDEBUG
2491 [19451] build: Make check fails on test-double-vlen2
2492 [19462] libc: Glibc failed to build with -Os
2493 [19465] math: Wrong code with -Os
2494 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2495 -Os
2496 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2497 CPU's.
2498 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2499 [19486] math: S390: Math tests fail with "Exception Inexact set".
2500 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2501 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2502 architectures
2503 [19590] math: Fail to build shared objects that use libmvec.so functions.
2504 \f
2505 Version 2.22
2506
2507 * The following bugs are resolved with this release:
2508
2509 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2510 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2511 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2512 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2513 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2514 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2515 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2516 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2517 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2518 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2519 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2520 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2521 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2522 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2523 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2524 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2525 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2526 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2527 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2528 18657, 18676, 18694, 18696, 18887.
2529
2530 * Cache information can be queried via sysconf() function on s390 e.g. with
2531 _SC_LEVEL1_ICACHE_SIZE as argument.
2532
2533 * A buffer overflow in gethostbyname_r and related functions performing DNS
2534 requests has been fixed. If the NSS functions were called with a
2535 misaligned buffer, the buffer length change due to pointer alignment was
2536 not taken into account. This could result in application crashes or,
2537 potentially arbitrary code execution, using crafted, but syntactically
2538 valid DNS responses. (CVE-2015-1781)
2539
2540 * The time zone file parser has been made more robust against crafted time
2541 zone files, avoiding heap buffer overflows related to the processing of
2542 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2543 large time zone data files. Overly long time zone specifiers in the TZ
2544 variable no longer result in stack overflows and crashes.
2545
2546 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2547 for LD and GD on x86 and x86-64, has been implemented. You will need
2548 binutils-2.24 or later to enable this optimization.
2549
2550 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2551 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2552 Hat). These updates cause user visible changes, such as the fix for bug
2553 17998.
2554
2555 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2556 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2557 condition in some applications.
2558
2559 * Added vector math library named libmvec with the following vectorized x86_64
2560 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2561 pow, powf.
2562 The library can be disabled with --disable-mathvec. Use of the functions is
2563 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2564 Shared library libmvec.so is linked in as needed when using -lm (no need to
2565 specify -lmvec explicitly for not static builds).
2566 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2567
2568 * A new fmemopen implementation has been added with the goal of POSIX
2569 compliance. The new implementation fixes the following long-standing
2570 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2571 old implementation is still present for use be by existing binaries.
2572
2573 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2574 releases. It has been fixed to match 2.19 and older, but binaries built
2575 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2576
2577 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2578 Contributed by Roland McGrath (Google).
2579
2580 * The header <regexp.h> is deprecated, and will be removed in a future
2581 release. Use of this header will trigger a deprecation warning.
2582 Application developers should update their code to use <regex.h> instead.
2583
2584 This header was formerly part of SUS, but was deprecated in 1994 and
2585 removed from the standard in 2001. Also, the glibc implementation
2586 leaks memory. See BZ#18681 for more details.
2587 \f
2588 Version 2.21
2589
2590 * The following bugs are resolved with this release:
2591
2592 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2593 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2594 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2595 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2596 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2597 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2598 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2599 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2600 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2601 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2602 17892.
2603
2604 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2605 too little memory for the to-be-scanned arguments and overflow the
2606 allocated buffer. The implementation now correctly computes the required
2607 buffer size when using malloc, and switches to malloc from alloca as
2608 intended.
2609
2610 * A new semaphore algorithm has been implemented in generic C code for all
2611 machines. Previous custom assembly implementations of semaphore were
2612 difficult to reason about or ensure that they were safe. The new version
2613 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2614 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2615 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2616
2617 * Port to Altera Nios II has been contributed by Mentor Graphics.
2618
2619 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2620 implementations for powerpc64/powerpc64le.
2621 Implemented by Adhemerval Zanella (IBM).
2622
2623 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2624 and powerpc64le. This may improve lock scaling of existing programs on
2625 HTM capable systems. The lock elision code is only enabled with
2626 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2627 powerpc will issue a transaction abort on every syscall to avoid side
2628 effects being visible outside transactions.
2629
2630 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2631 AArch64. Contributed by ARM Ltd.
2632
2633 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2634
2635 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2636 under certain input conditions resulting in the execution of a shell for
2637 command substitution when the application did not request it. The
2638 implementation now checks WRDE_NOCMD immediately before executing the
2639 shell and returns the error WRDE_CMDSUB as expected.
2640
2641 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2642 processing format strings with a large number of format specifiers.
2643
2644 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2645 infinite loop if the DNS response contained a PTR record of an unexpected
2646 format.
2647
2648 * The minimum GCC version that can be used to build this version of the GNU
2649 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2650 still be used to compile programs using the GNU C Library.
2651
2652 * The GNU C Library is now built with -Werror by default. This can be
2653 disabled by configuring with --disable-werror.
2654
2655 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2656
2657 * The obsolete sigvec function has been removed. This was the original
2658 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2659 programs have been using instead for about 25 years. Of course, ABI
2660 compatibility for old binaries using sigvec remains intact.
2661
2662 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2663 with newer versions of bison.
2664
2665 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2666 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2667 registers overlay two consecutive single-precision registers. MIPS32R2
2668 introduced a new FPU mode (FR=1) where double-precision registers extend the
2669 corresponding single-precision registers which is incompatible with the
2670 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2671 require the use of FR=1 making a transition necessary. New o32 ABI
2672 extensions enable users to migrate over time from the original o32 ABI
2673 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2674 tracks the ABI of any loaded object and verifies that new objects are
2675 compatible. Mode transitions will also be requested as required and
2676 unsupportable objects will be rejected. The ABI checks include both soft and
2677 hard float ABIs for o32, n32 and n64.
2678
2679 GCC 5 with GNU binutils 2.25 onwards:
2680 It is strongly recommended that all o32 system libraries are built using the
2681 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2682 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2683 compiler using --with-fp-32=xx to set this by default.
2684 \f
2685 Version 2.20
2686
2687 * The following bugs are resolved with this release:
2688
2689 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2690 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2691 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2692 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2693 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2694 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2695 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2696 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2697 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2698 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2699 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2700 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2701 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2702 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2703 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2704
2705 * Reverted change of ABI data structures for s390 and s390x:
2706 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2707 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2708 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2709 are preserved pointing straight to the same implementation as the old ones.
2710 Given that, new callers will simply provide a too-big buffer to these
2711 functions. Any applications/libraries out there that embed jmp_buf or
2712 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2713 against 2.19 headers will have to rebuilt again. This is necessary in any
2714 case to revert the breakage in their ABI caused by the glibc change.
2715
2716 * Support for file description locks is added to systems running the
2717 Linux kernel. The standard file locking interfaces are extended to
2718 operate on file descriptions, not file descriptors, via the use of
2719 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2720 are associated with an open file instead of a process.
2721
2722 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2723
2724 * The minimum Linux kernel version that this version of the GNU C Library
2725 can be used with is 2.6.32.
2726
2727 * Running the testsuite no longer terminates as soon as a test fails.
2728 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2729 with PASS or FAIL lines for individual tests. A summary of the results is
2730 printed, including a list of failing lists, and "make check" exits with
2731 error status if there were any unexpected failures. "make check
2732 stop-on-test-failure=y" may be used to keep the old behavior.
2733
2734 * The am33 port, which had not worked for several years, has been removed
2735 from ports.
2736
2737 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2738 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2739 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2740 conflicted with POSIX (support for which was removed in 2.19), the
2741 interfaces those macros enabled remain available when compiling with
2742 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2743 test macros defined.
2744
2745 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2746
2747 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2748 This may improve lock scaling of existing programs on TX capable systems.
2749 The lock elision code is only built with --enable-lock-elision=yes and
2750 then requires a GCC version supporting the TX builtins. With lock elision
2751 default mutexes are elided via __builtin_tbegin, if the cpu supports
2752 transactions. By default lock elision is not enabled and the elision code
2753 is not built.
2754
2755 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2756 copy the path argument. This allowed programs to cause posix_spawn to
2757 deference a dangling pointer, or use an unexpected pathname argument if
2758 the string was modified after the posix_spawn_file_actions_addopen
2759 invocation.
2760
2761 * All supported architectures now use the main glibc sysdeps directory
2762 instead of some being in a separate "ports" directory (which was
2763 distributed separately before glibc 2.17).
2764
2765 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2766 On configurations that support it (all Linux configurations), it's now
2767 used regardless of the --enable-add-ons switch to configure. It is no
2768 longer possible to build such configurations without pthreads support.
2769
2770 * Locale names, including those obtained from environment variables (LANG
2771 and the LC_* variables), are more tightly checked for proper syntax.
2772 setlocale will now fail (with EINVAL) for locale names that are overly
2773 long, contain slashes without starting with a slash, or contain ".." path
2774 components. (CVE-2014-0475) Previously, some valid locale names were
2775 silently replaced with the "C" locale when running in AT_SECURE mode
2776 (e.g., in a SUID program). This is no longer necessary because of the
2777 additional checks.
2778
2779 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2780 with application code using Intel MPX instructions. (With all previous
2781 versions, the MPX register state could be clobbered when making calls
2782 into or out of a shared library.) Note that while the new dynamic
2783 linker is compatible with all known x86 hardware whether or not it
2784 supports Intel MPX, some x86 instruction-set emulators might fail to
2785 handle the new instruction encodings. This is known to affect Valgrind
2786 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2787 release), and might affect other tools that do instruction emulation.
2788
2789 * Support for loadable gconv transliteration modules has been removed.
2790 The support for transliteration modules has been non-functional for
2791 over a decade, and the removal is prompted by security defects. The
2792 normal gconv conversion modules are still supported. Transliteration
2793 with //TRANSLIT is still possible, and the //IGNORE specifier
2794 continues to be supported. (CVE-2014-5119)
2795
2796 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2797 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2798 resulting a denial-of-service security vulnerability in applications which
2799 use functions related to iconv. (CVE-2014-6040)
2800 \f
2801 Version 2.19
2802
2803 * The following bugs are resolved with this release:
2804
2805 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2806 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2807 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2808 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2809 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2810 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2811 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2812 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2813 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2814 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2815 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2816 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2817 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2818 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2819 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2820 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2821 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2822 16431, 16453, 16474, 16506, 16510, 16529
2823
2824 * Slovenian translations for glibc messages have been contributed by the
2825 Translation Project's Slovenian team of translators.
2826
2827 * The public headers no longer use __unused nor __block. This change is to
2828 support compiling programs that are derived from BSD sources and use
2829 __unused internally, and to support compiling with Clang's -fblock
2830 extension which uses __block.
2831
2832 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2833 large collation sequences to optimize multiple passes. This cache
2834 computation may overflow for large collation sequences and may cause a
2835 stack or buffer overflow. This is now fixed to use a slower algorithm
2836 which does not use a cache if there is an integer overflow.
2837
2838 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2839 rules for large collation sequences to optimize multiple passes and falls
2840 back to alloca if malloc fails, resulting in a possible stack overflow.
2841 The implementation now falls back to an uncached collation sequence lookup
2842 if malloc fails.
2843
2844 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2845 initialized for static applications resulting in the security feature
2846 being disabled. The pointer guard is now correctly initialized to a
2847 random value for static applications. Existing static applications need
2848 to be recompiled to take advantage of the fix (bug 15754).
2849
2850 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2851 to the d_name member of struct dirent, or omit the terminating NUL
2852 character. (Bugzilla #14699).
2853
2854 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2855 aligned_alloc functions could allocate too few bytes or corrupt the
2856 heap when passed very large allocation size values (Bugzilla #15855,
2857 #15856, #15857).
2858
2859 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2860 for AF_INET6 has been fixed (Bugzilla #16072).
2861
2862 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2863 pap_CW, quz_PE, the_NP.
2864
2865 * Substantially revised locales: gd_GB, ht_HT
2866
2867 * The LC_ADDRESS field was updated to support country_car for almost all
2868 supported locales.
2869
2870 * ISO 1427 definitions were updated.
2871
2872 * ISO 3166 definitions were updated.
2873
2874 * The localedef utility now supports --big-endian and --little-endian
2875 command-line options to generate locales for a different system from that
2876 for which the C library was built.
2877
2878 * Binary locale files now only depend on the endianness of the system for
2879 which they are generated and not on other properties of that system. As a
2880 consequence, binary files generated with new localedef may be incompatible
2881 with old versions of the GNU C Library, and binary files generated with
2882 old localedef may be incompatible with this version of the GNU C Library,
2883 in the following circumstances:
2884
2885 + Locale files may be incompatible on m68k systems.
2886
2887 + Locale archive files (but not separate files for individual locales) may
2888 be incompatible on systems where plain "char" is signed.
2889
2890 * The configure option --disable-versioning has been removed. Builds with
2891 --disable-versioning had not worked for several years.
2892
2893 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2894 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2895
2896 * SystemTap probes for malloc have been introduced.
2897
2898 * SystemTap probes for slow multiple precision fallback paths of
2899 transcendental functions have been introduced.
2900
2901 * Support for powerpc64le has been added.
2902
2903 * The soft-float powerpc port now supports e500 processors.
2904
2905 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2906
2907 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2908 set of header declarations that are enabled by default, even when other
2909 feature test macros or compiler options such as -std=c99 would otherwise
2910 disable some of those declarations.
2911
2912 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2913 conflict with POSIX. The libbsd-compat library (which was a dummy library
2914 that did nothing) has also been removed.
2915
2916 * Preliminary documentation about Multi-Thread, Async-Signal and
2917 Async-Cancel Safety has been added.
2918
2919 * Change of ABI data structures for s390 and s390x:
2920 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2921 allow for future hardware extensions. All library functions that accept or
2922 return these structures were versioned in order to provide backwards
2923 compatibility with existing code. However, not all cases can be handled
2924 with interface versioning. If an application embeds either structure into
2925 another structure and that structure is passed to another compilation unit
2926 which expects the newer larger-sized object then any access to the new
2927 fields will result in undefined behaviour. Similarly any access to fields
2928 that were moved by the enlarging of the structures will cause undefined
2929 behaviour. To fix the undefined behaviour all objects that are part of the
2930 final in-memory image must agree on the size of structures, and this may
2931 require recompilation.
2932 \f
2933 Version 2.18
2934
2935 * The following bugs are resolved with this release:
2936
2937 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2938 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2939 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2940 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2941 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2942 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2943 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2944 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2945 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2946 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2947 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2948 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2949 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2950 15755, 15759.
2951
2952 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2953 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2954 Distributions can re-enable building and using pt_chown via the new configure
2955 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2956 considerable security risks and should only be used if the distribution
2957 understands and accepts the risks.
2958
2959 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2960 #15078).
2961
2962 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2963 fixed (Bugzilla #15330).
2964
2965 * Add support for calling C++11 thread_local object destructors on thread
2966 and program exit. This needs compiler support for offloading C++11
2967 destructor calls to glibc.
2968
2969 * Improved worst case performance of libm functions with double inputs and
2970 output.
2971
2972 * Support for priority inherited mutexes in pthread condition variables on
2973 non-x86 architectures.
2974
2975 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2976
2977 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2978
2979 * Optimized string functions for ARM. Implemented by Will Newton and
2980 Richard Henderson.
2981
2982 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2983
2984 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2985 Richard Henderson.
2986
2987 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2988 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2989
2990 * Added a benchmark framework to track performance of functions in glibc.
2991
2992 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2993 It is based on draft TS 18661 and currently enabled as a GNU extension.
2994
2995 * On Linux, the clock function now uses the clock_gettime system call
2996 for improved precision, rather than old times system call.
2997
2998 * Added support for version-3 tz format files. This is needed when using
2999 the tz database release 2013e or later, and affects a few unusual cases --
3000 currently only TZ='America/Godthab' for time stamps after 2037.
3001
3002 * Added new API functions pthread_getattr_default_np and
3003 pthread_setattr_default_np to get and set the default pthread
3004 attributes of a process.
3005
3006 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3007 This may improve lock scaling of existing programs on TSX capable systems.
3008 When the --enable-lock-elision=yes parameter is specified at configure
3009 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3010 mutexes.
3011
3012 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3013 availability). Implemented by Ryan S. Arnold.
3014
3015 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3016 \f
3017 Version 2.17
3018
3019 * The following bugs are resolved with this release:
3020
3021 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3022 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3023 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3024 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3025 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3026 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3027 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3028 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3029 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3030 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3031 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3032 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3033 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3034 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3035 14879, 14889, 14893, 14898, 14914.
3036
3037 * Optimization of memcpy for MIPS.
3038
3039 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3040 EMFILE has been fixed (Bugzilla #14889).
3041
3042 * The add-on ports collection is now distributed in the "ports" subdirectory
3043 of the main GNU C Library distribution, rather than separately.
3044
3045 * Port to ARM AArch64 contributed by Linaro.
3046
3047 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3048 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3049 zEnterprise z196.
3050 Implemented by Andreas Krebbel.
3051
3052 * The new function secure_getenv allows secure access to the environment,
3053 returning NULL if running in a SUID/SGID process. This function replaces
3054 the internal function __secure_getenv.
3055
3056 * SystemTap static probes have been added into the dynamic linker.
3057 Implemented by Gary Benson.
3058
3059 * Optimizations of string functions strstr, strcasestr and memmem.
3060 Implemented by Maxim Kuvyrkov.
3061
3062 * The minimum Linux kernel version that this version of the GNU C Library
3063 can be used with is 2.6.16.
3064
3065 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3066 powerpc POWER7. Implemented by Will Schmidt.
3067
3068 * New configure option --disable-nscd builds the C library such that it
3069 never attempts to contact the Name Service Caching Daemon (nscd).
3070 New configure option --disable-build-nscd avoids building nscd itself;
3071 this is the default if --disable-nscd is used.
3072
3073 * Improved support for cross-compilation, including cross-testing and
3074 bootstrap builds without a previously built glibc.
3075
3076 * Several testsuite tests are now able to test multiple IFUNC variants of an
3077 interface, rather than just testing the one that would be chooen by
3078 default.
3079
3080 * New configure options --with-bugurl and --with-pkgversion, for
3081 distributors to use to embed their bug-reporting and package version
3082 information in --help and --version output.
3083
3084 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3085 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3086 allows creation of chroots without the procfs mounted on /proc.
3087
3088 * The `crypt' function now fails if passed salt bytes that violate the
3089 specification for those values. On Linux, the `crypt' function will
3090 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3091 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3092 when the mode is enabled.
3093
3094 * The `clock_*' suite of functions (declared in <time.h>) is now available
3095 directly in the main C library. Previously it was necessary to link with
3096 -lrt to use these functions. This change has the effect that a
3097 single-threaded program that uses a function such as `clock_gettime' (and
3098 is not linked with -lrt) will no longer implicitly load the pthreads
3099 library at runtime and so will not suffer the overheads associated with
3100 multi-thread support in other code such as the C++ runtime library.
3101
3102 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3103 sat_IN, and szl_PL.
3104 \f
3105 Version 2.16
3106
3107 * The following bugs are resolved with this release:
3108
3109 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3110 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3111 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3112 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3113 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3114 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3115 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3116 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3117 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3118 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3119 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3120 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3121 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3122 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3123 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3124 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3125 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3126 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3127 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3128 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3129 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3130 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3131 14277, 14278.
3132
3133 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3134 configuring glibc with:
3135 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3136 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3137 Implemented by H.J. Lu.
3138
3139 * ISO C11 support:
3140
3141 + define static_assert
3142
3143 + do not declare gets
3144
3145 + declare at_quick_exit and quick_exit also for ISO C11
3146
3147 + aligned_alloc. NB: The code is deliberately allows the size parameter
3148 to not be a multiple of the alignment. This is a moronic requirement
3149 in the standard but it is only a requirement on the caller, not the
3150 implementation.
3151
3152 + timespec_get added
3153
3154 + uchar.h support added
3155
3156 + CMPLX, CMPLXF, CMPLXL added
3157
3158 Implemented by Ulrich Drepper.
3159
3160 * Support for the IA-64 has been moved to ports.
3161
3162 * Remove support for anything but ELF binary format
3163
3164 * Checking versions of poll, ppoll added.
3165 Implemented by Ulrich Drepper.
3166
3167 * More generic and 64-bit performance optimizations to math functions.
3168 Implemented by Ulrich Drepper.
3169
3170 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3171 headers and functions available at compile time as they were before
3172 version 2.14. This option will be removed at some time in the future
3173 after the TI-RPC library becomes fully sufficient for the needs of
3174 existing applications.
3175
3176 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3177 Note that glibc is not expected to work with any Linux kernel version
3178 before 2.6.
3179
3180 * New header <sys/auxv.h> and function getauxval allowing easy access to
3181 the AT_* key-value pairs passed from the Linux kernel. The header also
3182 defines the HWCAP_* bits associated with the AT_HWCAP key.
3183
3184 * New locales: mag_IN
3185
3186 * New configure option --enable-systemtap builds SystemTap static probes
3187 into libc for setjmp and longjmp and into libpthread for various operations.
3188 So far the setjmp/longjmp probes and some of the libpthread probes are
3189 provided only for i*86 and x86_64.
3190 Implemented by Roland McGrath and Rayson Ho.
3191
3192 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3193
3194 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3195 and Will Schmidt.
3196
3197 * More optimized functions for SPARC. Implemented by David S. Miller.
3198
3199 * Improved support for cross-compilation, especially bootstrap builds
3200 without a previously built glibc.
3201
3202 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3203 Chris Metcalf from Tilera.
3204
3205 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3206 now supported for ARM processors.
3207
3208 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3209 as the name of the dynamic linker, to distinguish it from the
3210 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3211
3212 * Support for CRIS has been removed from ports.
3213
3214 * A new class of installed header has been documented for low-level
3215 platform-specific functionality. PowerPC added the first instance with a
3216 function to provide time base register access. Contributed by Tulio
3217 Magno Quites Machado Filho.
3218
3219 * ix86 configurations now install header files that are consistent with
3220 what x86-64 configurations install. These same header files can be used
3221 for -m32, -m64, or -mx32 builds.
3222 Contributed by H.J. Lu.
3223
3224 * Math library bug fixes. A thorough audit of all open math library bugs was
3225 conducted by Joseph Myers. Significant progress was made on many math
3226 library bugs resulting in more accurate exceptions and function results.
3227 Many thanks to all those that contributed including Andreas Jaeger for his
3228 patch review and work on the x87 trigonometric instruction issues.
3229
3230 * Timezone data is no longer installed. Timezone-related binaries and scripts
3231 will continue to be installed. Users should obtain their timezone data from
3232 their distribution provider or from the tzdata package at
3233 <ftp://munnari.oz.au/pub/>.
3234 \f
3235 Version 2.15
3236
3237 * The following bugs are resolved with this release:
3238
3239 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3240 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3241 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3242 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3243 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3244 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3245 13484, 13506, 13515, 13523, 13524, 13538, 13540
3246
3247 * New program pldd to list loaded object of a process
3248 Implemented by Ulrich Drepper.
3249
3250 * Add nss_db support back to glibc. No more dependency on Berkeley db
3251 and support for initgroups lookups.
3252 Implemented by Ulrich Drepper.
3253
3254 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3255 Contributed by HJ Lu.
3256
3257 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3258 Contributed by HJ Lu.
3259
3260 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3261 on x86-32 and x86-64.
3262 Contributed by Liubov Dmitrieva.
3263
3264 * Optimized strchr and strrchr for SSE on x86-32.
3265 Contributed by Liubov Dmitrieva.
3266
3267 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3268 for x86-64 and x86-32.
3269 Contributed by Liubov Dmitrieva.
3270
3271 * New interfaces: scandirat, scandirat64
3272 Implemented by Ulrich Drepper.
3273
3274 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3275 Implemented by Ulrich Drepper.
3276
3277 * nscd now also caches the netgroup database.
3278 Implemented by Ulrich Drepper.
3279
3280 * Integrate libm with gcc's -ffinite-math-only option.
3281 Implemented by Ulrich Drepper.
3282
3283 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3284 to math functions. Implemented by Ulrich Drepper.
3285
3286 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3287 Implemented by Ulrich Drepper.
3288
3289 * New Linux interfaces: process_vm_readv, process_vm_writev
3290
3291 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3292 Implemented by Ulrich Drepper.
3293
3294 * Optimized nearbyint and strcasecmp for PPC.
3295 Implemented by Adhemerval Zanella.
3296
3297 * New locales: bho_IN, unm_US, es_CU, ta_LK
3298 \f
3299 Version 2.14
3300
3301 * The following bugs are resolved with this release:
3302
3303 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3304 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3305 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3306 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3307 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3308 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3309 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3310 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3311 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3312 12795, 12811, 12813, 12814, 12841
3313
3314 * The RPC implementation in libc is obsoleted. Old programs keep working
3315 but new programs cannot be linked with the routines in libc anymore.
3316 Programs in need of RPC functionality must be linked against TI-RPC.
3317 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3318
3319 Visible changes of this change include (obviously) the inability to link
3320 programs using RPC functions without referencing the TI-RPC library and the
3321 removal of the RPC headers from the glibc headers.
3322 Implemented by Ulrich Drepper.
3323
3324 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3325 syncfs, setns, sendmmsg
3326
3327 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3328 yue_HK, lij_IT, mhr_RU
3329
3330 * New encodings: CP770, CP771, CP772, CP773, CP774
3331
3332 * New program sotruss to trace calls through PLTs
3333 Implemented by Ulrich Drepper.
3334
3335 * The malloc hook implementation is marked deprecated and will be removed
3336 from the default implementation in the next version. The design never
3337 worked ever since the introduction of threads. Even programs which do
3338 not create threads themselves can use multiple threads created internally.
3339 \f
3340 Version 2.13
3341
3342 * The following bugs are resolved with this release:
3343
3344 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3345 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3346 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3347 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3348 12378, 12394, 12397
3349
3350 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3351
3352 * POWER7 optimizations: memset, memcmp, strncmp
3353
3354 * New optimized string functions for x86-64: strnlen (SSE2),
3355 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3356 Implemented by Ulrich Drepper.
3357 \f
3358 Version 2.12
3359
3360 * The following bugs are resolved with this release:
3361
3362 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3363 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3364 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3365 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3366 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3367 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3368 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3369 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3370
3371 * New interfaces: pthread_getname_np, pthread_setname_np
3372
3373 * New Linux interface: recvmmsg
3374
3375 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3376
3377 * The dynamic linker now recognizes supported ABI versions from the
3378 EI_ABIVERSION field in the ELF header.
3379 Implemented by Ulrich Drepper.
3380
3381 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3382 will not be used to fill in password fields in the passwd.byname replies.
3383 Instead it is used to synthesize the shadow.byname table, should it be
3384 missing. This is a useful mode in some installations involving Solaris.
3385 Implemented by Ulrich Drepper.
3386
3387 * New locales: kok_IN, sq_MK, cv_RU
3388 \f
3389 Version 2.11
3390
3391 * The following bugs are resolved with this release:
3392
3393 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3394 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3395 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3396 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3397 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3398 10789, 10817, 10823, 10840, 10847
3399
3400 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3401 mkostemps64
3402 Implemented by Ulrich Drepper.
3403
3404 * Checking version of longjmp added that fails if an uninitialized stack
3405 frame would be created. Implemented by Ulrich Drepper.
3406
3407 * STT_GNU_IFUNC is now supported in static executables.
3408 Implemented by H.J. Lu.
3409
3410 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3411
3412 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3413 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3414 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3415 Contributed by H.J. Lu.
3416
3417 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3418 Implemented by Ulrich Drepper.
3419
3420 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3421 strstr, strcasestr.
3422 Contributed by H.J. Lu.
3423
3424 * Support for fma instruction in AVX on x86-64.
3425 Implemented by H.J. Lu and Ulrich Drepper.
3426
3427 * AVX support in x86-64 auditing support in ld.so.
3428 Implemented by H.J. Lu.
3429
3430 * STB_GNU_UNIQUE support added.
3431 Implemented by Ulrich Drepper.
3432
3433 * Implement second fallback mode for optimized DNS lookup for even more
3434 broken environments. If two requests from the same port are not
3435 handled correctly close the socket and open a new one before sending
3436 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3437 can be used to select this mode right away, instead of rediscovering the
3438 necessity is every process again.
3439 Implemented by Ulrich Drepper.
3440
3441 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3442 Implemented by Adam Tkac.
3443
3444 * Optimized iconv conversions for S390x.
3445 Implemented by Andreas Krebbel.
3446
3447 * Using condvars with PI mutexes is now more efficient due to kernel
3448 support for requeueing to PI futexes. NPTL support added for x86-64.
3449 Implemented by Ulrich Drepper.
3450
3451 * New locales: ps_AF, my_MM
3452 \f
3453 Version 2.10
3454
3455 * The following bugs are resolved with this release:
3456
3457 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3458 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3459 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3460 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3461 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3462
3463 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3464 Implemented by Ulrich Drepper.
3465
3466 * New Linux interfaces: accept4, fallocate, fallocate64.
3467 Implemented by Ulrich Drepper.
3468
3469 * Correct declarations of string function when used in C++ code. This
3470 could lead to compile errors for invalid C++ code.
3471
3472 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3473 now in POSIX.
3474
3475 * New POSIX 2008 interface: psiginfo
3476 Implemented by Ulrich Drepper.
3477
3478 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3479 Implemented by Ulrich Drepper.
3480
3481 * Support for selecting between multiple function definitions at runtime
3482 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3483
3484 * The libcrypt library can now use the hash function implementations in
3485 NSS. Implemented by Ulrich Drepper.
3486
3487 * The malloc implementation can be compiled to be less memory efficient
3488 but higher performing in multi-threaded programs.
3489 Implemented by Ulrich Drepper.
3490
3491 * New locales: nan_TW@latin, ks_IN
3492
3493 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3494 Implemented by Ulrich Drepper.
3495
3496 * Extended printf hook support. It is possible to use user-defined types
3497 and extend existing format specifiers.
3498 Implemented by Ulrich Drepper.
3499
3500 * Handling for group shadow files has been added.
3501 Implemented by Ulrich Drepper.
3502
3503 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3504 servers (the case, e.g., for some people using the built-in DNS
3505 server in ADSL modems/routers). There is a once-per-process timeout
3506 in case of a broken server. To avoid it, users can run nscd or put
3507 'options single-request' in /etc/resolv.conf.
3508 Implemented by Ulrich Drepper.
3509 \f
3510 Version 2.9
3511
3512 * The following bugs are resolved with this release:
3513
3514 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3515 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3516 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3517 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3518 6968, 6974, 6980, 6995, 7008, 7009, 7029
3519
3520 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3521 up at the same time. Implemented by Ulrich Drepper.
3522
3523 * TLS descriptors for LD and GD on x86 and x86-64.
3524 Implemented by Alexandre Oliva.
3525
3526 * getaddrinfo now handles DCCP and UDPlite.
3527 Implemented by Ulrich Drepper.
3528
3529 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3530 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3531 Implemented by Ulrich Drepper.
3532
3533 * New implementation of memmem, strstr, and strcasestr which is O(n).
3534 Implemented by Eric Blake.
3535
3536 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3537
3538 * Implement "e" option for popen to open file descriptor with the
3539 close-on-exec flag set. Implemented by Ulrich Drepper.
3540
3541 * Implement "b" mode for fmemopen. In this mode writes writes don't
3542 implicitly add a NUL byte and seeks from the end of the buffer really
3543 use the buffer end, not the string length as the basis.
3544 Implemented by Ulrich Drepper.
3545
3546 * Many functions, exported and internal, now atomically set the close-on-exec
3547 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3548
3549 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3550 Sinhala)
3551 Implemented by Pravin Satpute.
3552
3553 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3554 \f
3555 Version 2.8
3556
3557 * The following bugs are resolved with this release:
3558
3559 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3560 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3561 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3562 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3563 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3564 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3565 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3566
3567 * New locales: bo_CN, bo_IN, shs_CA.
3568
3569 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3570
3571 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3572 Implemented by Pravin Satpute.
3573
3574 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3575
3576 * nscd caches now all timeouts for DNS entries
3577 Implemented by Ulrich Drepper.
3578
3579 * nscd is more efficient and wakes up less often.
3580 Implemented by Ulrich Drepper.
3581
3582 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3583 vdprintf, and obstack_vprintf.
3584 Implemented by Jakub Jelinek.
3585
3586 * Faster memset for x86-64.
3587 Implemented by Harsha Jagasia and H.J. Lu.
3588
3589 * Faster memcpy on x86.
3590 Implemented by Ulrich Drepper.
3591
3592 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3593 Implemented by Ulrich Drepper.
3594
3595 * Faster sqrt and sqrtf implemention for some PPC variants.
3596 Implemented by Steven Munroe.
3597 \f
3598 Version 2.7
3599
3600 * The following bugs are resolved with this release:
3601
3602 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3603 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3604 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3605 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3606 5071, 5103, 5104, 5112, 5113, 5184, 5186
3607
3608 * More checking functions: fread, fread_unlocked, open*, mq_open.
3609 Implemented by Jakub Jelinek and Ulrich Drepper.
3610
3611 * Extend fortification to C++. Implemented by Jakub Jelinek.
3612
3613 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3614 by not recognizing 'a' as a modifier when those specs are requested.
3615 Implemented by Jakub Jelinek.
3616
3617 * PPC optimizations to math and string functions.
3618 Implemented by Steven Munroe.
3619
3620 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3621 options to be passed. Implemented by Ulrich Drepper.
3622
3623 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3624
3625 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3626 Implemented by Ulrich Drepper.
3627
3628 * Handle private futexes in the NPTL implementation.
3629 Implemented by Jakub Jelinek and Ulrich Drepper.
3630
3631 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3632 Implemented by Roland McGrath and Ulrich Drepper.
3633
3634 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3635
3636 * SHA-256 and SHA-512 based password encryption.
3637 Implemented by Ulrich Drepper.
3638
3639 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3640 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3641 yo_NG.
3642
3643 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3644 Implemented by Ulrich Drepper.
3645 \f
3646 Version 2.6
3647
3648 * The following bugs are resolved with this release:
3649
3650 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3651 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3652 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3653 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3654 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3655 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3656 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3657 4702, 4858
3658
3659 * New Linux interfaces: epoll_pwait, sched_getcpu.
3660
3661 * New generic interfaces: strerror_l.
3662
3663 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3664 \f
3665 Version 2.5
3666
3667 * The following bugs are resolved with this release:
3668
3669 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3670 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3671 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3672 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3673 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3674 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3675 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3676 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3677 3225, 3251, 3252, 3253, 3273
3678
3679 * For Linux, the sorting of addresses returned by getaddrinfo now also
3680 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3681 Implemented by Ulrich Drepper.
3682
3683 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3684 Implemented by Ulrich Drepper.
3685
3686 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3687
3688 * New iconv module for MIK. Contributed by Alexander Shopov.
3689
3690 * For sites with broken group and/or passwd database, the auto-propagate
3691 option of nscd can prevent creating ID lookup entries from the results
3692 of a name lookup and vice versa. This usually is no problem but some
3693 site might have problems with the default behavior.
3694 Implemented by Ulrich Drepper.
3695
3696 * Iterating over entire database in NIS can be slow. With the
3697 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3698 to trade time for memory. The entire database will be read at once.
3699 Implemented by Ulrich Drepper.
3700
3701 * The interfaces introduced in RFC 3542 have been implemented by
3702 Ulrich Drepper.
3703
3704 * Support for the new ELF hash table format was added by Ulrich Drepper.
3705
3706 * Support for priority inheritance mutexes added by Jakub Jelinek and
3707 Ulrich Drepper.
3708
3709 * Support for priority protected mutexes added by Jakub Jelinek.
3710
3711 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3712 or_IN, csb_PL, fy_NL, sr_ME.
3713 \f
3714 Version 2.4
3715
3716 * More overflow detection functions.
3717
3718 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3719 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3720
3721 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3722 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3723 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3724 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3725 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3726 by Masahide Washizawa.
3727
3728 * It is now possible to install audit libraries for the dynamic linker using
3729 LD_AUDIT. Implemented by Ulrich Drepper.
3730
3731 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3732 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3733 For a libc and libpthread that works well on Linux 2.4 kernels, we
3734 recommend using the stable 2.3 branch.
3735
3736 * The new function open_wmemstream acts like open_memstream,
3737 but uses a wchar_t wide character buffer.
3738
3739 * The new function ppoll is an improved version of poll, similar to pselect.
3740
3741 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3742 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3743 renameat, unlinkat, linkat, symlinkat, readlinkat.
3744
3745 * New Linux kernel system calls: unshare,
3746 inotify_init, inotify_add_watch, inotify_rm_watch.
3747
3748 * The euidaccess function is now also known by the alias eaccess,
3749 for compatibility with some other systems.
3750
3751 * Timezone data updated to 2006b version.
3752 \f
3753 Version 2.3.6
3754
3755 * The following bugs are resolved with this release:
3756
3757 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3758 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3759 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3760 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3761 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3762 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3763
3764 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3765
3766 * As of this release, GCC 4 can be used to compile the C Library.
3767
3768 * Timezone data updated to 2005m version.
3769 \f
3770 Version 2.3.5
3771
3772 * The following bugs are resolved with this release:
3773
3774 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3775 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3776 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3777 777, 787, 821, 822, 823, 825
3778
3779 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3780 \f
3781 Version 2.3.4
3782
3783 * Support for RFC 3678. Real implementations exist only for Linux so far.
3784 Implemented by Ulrich Drepper.
3785
3786 * nscd can now cache entries persistently. Expiring entries are reloaded.
3787 For speedups the cache can be shared in memory with client processes.
3788 Implemented by Ulrich Drepper.
3789
3790 * nscd can now perform SELinux checks.
3791 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3792
3793 * getaddrinfo queries are now cached. Canonical name lookup is performed
3794 efficiently.
3795 Implemented by Ulrich Drepper.
3796
3797 * The nothrow function attribute is used when headers are used by gcc when
3798 compiling C code. This can avoid significant amounts of exception
3799 handling data.
3800
3801 * The malloc functions perform more error checking and are stricter when
3802 it comes to reacting on errors. The default action is to terminate
3803 the process after showing an error message. Implemented by Ulrich Drepper.
3804
3805 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3806 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3807
3808 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3809 namespaces using the new function dlmopen(). This feature is of course,
3810 like most other dynamic loading functionality, not available in statically
3811 linked applications. Implemented by Ulrich Drepper.
3812
3813 * Low-overhead boundary checking variants of string and some stdio functions
3814 were added. These are to be used in conjunction with a gcc patch by
3815 Jakub Jelinek which adds calls to these functions if possible.
3816 Implemented by Jakub Jelinek and Ulrich Drepper.
3817
3818 * Old code for several operating systems and machine architectures that
3819 have not been in working condition in a long time have been removed from
3820 the main source tree maintained by the GNU C Library's maintainers.
3821 These files are now reside in the separate `ports' source module
3822 that is usable as an add-on when building the library.
3823 \f
3824 Version 2.3.3
3825
3826 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3827 interrogate the dynamic linker, compatible with the Solaris interface.
3828
3829 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3830 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3831
3832 * getifaddrs now uses the netlink interface on Linux to get its information.
3833 Implemented by Thorsten Kukuk.
3834
3835 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3836 Implemented by Ulrich Drepper.
3837
3838 * support for non-executable stacks on x86 has been added. Changes mostly
3839 by Roland McGrath.
3840
3841 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3842 and Ulrich Drepper.
3843
3844 * getaddrinfo now performs destination address selection according to
3845 RFC 3484.
3846 \f
3847 Version 2.3.2
3848
3849 * Thread-safe interfaces for many functions that access locale data
3850 were added in version 2.3, but these features were omitted from NEWS.
3851 Many functions have variants with an `_l' suffix that take a `locale_t'
3852 object as a parameter rather than consulting the current locale.
3853 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3854 create and maintain `locale_t' objects. Additionally, the new function
3855 `uselocale' sets "the current locale" (as used by functions not so
3856 parameterized) set for an individual thread. These features were added
3857 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3858
3859 * The functions getresuid, getresgid, setresuid, and setresgid, which
3860 have long been available on Linux, are now declared in <unistd.h>
3861 and are now also available on the Hurd.
3862
3863 * ELF thread-local storage support (TLS) now works on x86-64.
3864
3865 * The new dynamic string token $LIB is expanded in shared library names.
3866 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3867
3868 * Aldy Hernandez contributed complete software floating point support for
3869 PowerPC machines with no FPU.
3870
3871 * fexecve is implemented on Linux.
3872
3873 * The `btowc' function should work at least twice as fast due to
3874 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3875
3876 * With approriate thread add-ons cancelable functions are now implemented
3877 in libc.so as well. No need to call the function in libpthread. This
3878 change allowed to finally disable the incorrect and expensive handling
3879 of weak definition in ld.so.
3880
3881 * Yet more PLT entries in libc.so have been removed. We finally arrived
3882 at the bare minimum. Startup times improved appropriately.
3883
3884 * Support for the new Linux/x86 system call interface was added. The
3885 AT_SYSINFO auxiliary vector entry is recognized and handled.
3886 \f
3887 Version 2.3
3888
3889 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3890 charsets.
3891
3892 * iconv (the program and the interface) now accepts empty names (excluding
3893 options like //TRANSLIT) to mean "use charset of current locale".
3894
3895 * localedef can now transliterate characters in strings which are not in
3896 the provided charmap. The information from the input locale is used.
3897
3898 * Prelinking support was added for ELF targets. This requires additional
3899 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3900
3901 * Read-only stdio streams now use mmap to speed up operation by eliminating
3902 copying and buffer underflows. To use add 'm' to the mode string of
3903 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3904
3905 * The malloc functions were completely rewritten by Wolfram Gloger based
3906 on Doug Lea's malloc-2.7.0.c.
3907
3908 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3909 implementation of regex.
3910
3911 * Bruno Haible upgraded the iconv and locale implementation to support
3912 Unicode 3.2.
3913
3914 * Contents of the LC_* and LANG environment variables in the CEN style are
3915 not recognized anymore. It never was used. Change by Ulrich Drepper.
3916
3917 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3918 thread-local storage (TLS) ABI on some platforms.
3919 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3920
3921 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3922 EUC-JISX0213, and TSCII.
3923
3924 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3925 BSD-compatible interface for getting all network interface addresses.
3926 Implementation for IPv4 by Roland McGrath.
3927
3928 * Loading of locale data is faster due to the introduction of a locale
3929 archive. Implemented by Roland McGrath and Ulrich Drepper.
3930
3931 * Startup times are significantly reduced by not using exported functions
3932 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3933 and Ulrich Drepper.
3934
3935 * Steven Munroe contributed a port to PowerPC64/Linux.
3936 \f
3937 Version 2.2.6
3938
3939 * The Hurd now uses the GNU libio implementation of stdio rather than the
3940 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3941
3942 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3943 with the same behavior as the Linux system call of the same name.
3944 \f
3945 Version 2.2.5
3946
3947 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3948 128-bit long double format.
3949
3950 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3951 IBM1161, and IBM1162 charsets.
3952
3953 * Andreas Jaeger contributed a port to x86-64/Linux.
3954
3955 * Peter Bruin contributed a port to PowerPC/Hurd.
3956
3957 * libc functions using I/O streams now can handle wide-oriented streams
3958 as well.
3959
3960 * optimizations in the dynamic linker. Binaries created by recent binutils
3961 versions start up quicker due to reduced time spent on relocations.
3962
3963 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3964 \f
3965 Version 2.2.4
3966
3967 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3968 asinh, atanh, j0 for the 128-bit long double format.
3969
3970 * Bruno Haible updated all the code handling Unicode in some form to
3971 support Unicode 3.1.
3972
3973 * Speed of regex for single-byte locales is back to previous levels.
3974 Patch by Isamu Hasegawa.
3975
3976 * Alpha, SPARC, and IA-64 now also using floating stacks.
3977
3978 * Startup time of internationalized applications greatly improved through
3979 iconv cache. Use iconvconfig to generate the cache file.
3980 Contributed by Ulrich Drepper.
3981
3982 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3983 severe performance problems. Patch by David Mosberger.
3984
3985 * The Hurd port got a lot more functionality like AIO, various stdio
3986 extensions, etc. Mainly done by Roland McGrath.
3987
3988 * mtrace can now lookup symbols in shared libraries.
3989 \f
3990 Version 2.2.3
3991
3992 * Intel's IA-64 math library is largely integrated. It provides fast and
3993 accurate implementatations for most basic and standard math functions
3994 in float, double, and long double format.
3995
3996 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3997 and asin for the 96-bit long double format and asin, log, tan for the
3998 128-bit long double format.
3999
4000 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4001 The basic double functions exist today. Contributed by Abraham Ziv
4002 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4003 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4004
4005 * An asynchronous name lookup library was added. The interface is designed
4006 after POSIX AIO. The proposal was circulated beforehand to get comments.
4007 No negative ones came in. Implemented by Ulrich Drepper.
4008
4009 * Port to S390/64bit contributed by Martin Schwidefsky
4010 <schwidefsky@de.ibm.com>.
4011
4012 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4013 of functions for Linux/IA-64.
4014
4015 * The RPC code is now thread safe. Threads can now use the same service
4016 of different services at the same time. Patch by Eric Norum
4017 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4018
4019 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4020 family of functions for Linux/S390.
4021
4022 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4023 of functions for Linux/x86.
4024
4025 * Port to Linux/CRIS contributed by Axis Communications.
4026 \f
4027 Version 2.2.2
4028
4029 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4030 we can now check for namespace violations and missing declarations. The
4031 result is that almost all headers are now Unix-compliant (as defined in
4032 the upcoming XPG6). The negative side is that some programs might need
4033 corrections, too, if they depend on the incorrect form of the headers in
4034 previous versions which defined too many symbols and included too many
4035 other headers.
4036
4037 * regex now handles multibyte character sets correctly.
4038 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4039
4040 * iconv (the program) does now conform to the upcoming XPG6 and handles
4041 charmaps. Instead of the charset names the path of charmaps can be
4042 provided and the conversion happens based on this data.
4043 Contributed by Ulrich Drepper.
4044
4045 * The locale program now provides detailed information about the installed
4046 locales. While
4047
4048 locale -a
4049
4050 only lists the names of the supported locales
4051
4052 locale -a --verbose
4053
4054 provides details such as country, language, and codeset name.
4055 Contributed by Ulrich Drepper.
4056 \f
4057 Version 2.2.1
4058
4059 * The gencat program now parses the input file according to the charset
4060 selected by the LC_CTYPE category. This is important for stateful
4061 character sets. To make generating catalogs easier there is a way
4062 to overwrite the charset selected by the locale: before the first
4063 message or $ quote line the catalog can contain a line like
4064
4065 $ codeset=ISO-8859-2
4066
4067 to select the charset (ISO-8859-2 in this case).
4068
4069 Implemented by Shinya Hanataka and Ulrich Drepper.
4070
4071 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4072 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4073 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4074
4075 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4076 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4077
4078 * The LANGUAGE environment variable is now ignored unless the locale is
4079 changed from the default "C" locale.
4080
4081 * The usual bug fixes.
4082 \f
4083 Version 2.2
4084
4085 * Greg McGary added runtime support for bounds checking using gcc's
4086 new -fbounded-pointers option. ix86 target is complete. PowerPC
4087 is in progress.
4088
4089 * Thorsten Kukuk added secure mode support to nscd.
4090
4091 * The Berkeley db libraries have been removed.
4092
4093 Related, the nss_db module is now in a separate package since it
4094 obviously requires a database library being available.
4095
4096 * Wide character I/O streams implemented by Ulrich Drepper.
4097
4098 * Functions from the extended socket API added by Ulrich Drepper.
4099
4100 * Functions feenableexcept and fedisableexcept to control the
4101 behaviour of individual exceptions have been added by Andreas Jaeger.
4102
4103 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4104
4105 * The resolver code has been updated from bind 8.2.3-T5B which supports
4106 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4107 and Mark Kettenis.
4108
4109 This change could in some situations effect backward compatibility. Since
4110 now `_res' is a thread-local instead of a global variable, modifying it
4111 in one thread does not have any effect in other threads.
4112
4113 The resolver library was also extended to allow IPv6 as the transport
4114 protocol for the requests. This work was done by Stig Venaas.
4115
4116 * Compatibility code for K&R C compilers has been removed from the
4117 header files. A ISO C compiler is needed to use the library
4118 (conforming to either C89 or C99 standard).
4119
4120 * Complete rewrite of the localedef program to support multibyte character
4121 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4122 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4123 Implemented by Ulrich Drepper.
4124
4125 Bruno Haible significantly improved the generation and use of the data
4126 structures for the wide character tables.
4127
4128 * Plural handling in gettext implemented by Ulrich Drepper.
4129
4130 * The utmp daemon has been removed.
4131
4132 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4133
4134 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4135 and Yutaka Niibe.
4136
4137 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4138
4139 * POSIX barriers implemented by Kaz Kylheku.
4140
4141 * POSIX spawn function family implemented by Ulrich Drepper.
4142
4143 * POSIX spinlocks are now available.
4144
4145 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4146
4147 * the configure option --enable-kernel=X.Y.Z allows to strip out
4148 compatibility for kernel versions before X.Y.Z. This is currently only
4149 implemented for Linux.
4150
4151 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4152 field sin6_scope_id. This means that all programs using IPv6 should be
4153 recompiled. Don't expect binary compatibility with previous glibc
4154 versions.
4155
4156 * various conversion modules for IBM character sets contributed by
4157 Masahide Washizawa.
4158
4159 * IA-64 port by Jes Sorensen and HJ Lu.
4160 \f
4161 Version 2.1.3
4162
4163 * bug fixes
4164
4165 \f
4166 Version 2.1.2
4167
4168 * bug fixes
4169
4170 \f
4171 Version 2.1.1
4172
4173 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4174
4175 * New xdr functions are added; some rpc functions are now 64bit clean.
4176
4177 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4178
4179 * Fixed known incompatibilities with glibc 2.0.
4180
4181 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4182
4183 * Optimized a number of functions (especially the ELF dynamic loader).
4184
4185 * Update timezone data files.
4186
4187 * lots of charmaps corrections
4188
4189 * some new locale definitions and charmaps
4190
4191 \f
4192 Version 2.1
4193
4194 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4195 conform to POSIX member type requirements. Symbol versions have been
4196 adjusted as needed within the library, and for direct use by applications,
4197 but there is potential for problems if third-party libraries use
4198 struct timeval as part of their interface. This does not present
4199 a problem for X and other "essential" system libraries.
4200
4201 * An additional locale model to support C++ Standard Library locale
4202 model and probably more was implemented by Ulrich Drepper.
4203
4204 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4205 symbol level.
4206
4207 * Miles Bader provided the `argp' function family to support hierachical
4208 command line argument parsing, layered on top of getopt.
4209
4210 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4211
4212 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4213 numbers.
4214
4215 * scanf recognizes the %a and %A format for scanning floating point numbers.
4216
4217 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4218 information and interfaces for the available integer types.
4219
4220 * about 130 new math functions were added to implement the ISO C9x math
4221 library.
4222
4223 * the new header <complex.h> contains definitions of the complex math
4224 functions from ISO C 9X.
4225
4226 * the new header <tgmath.h> defines generic macros to use complex or
4227 real valued functions.
4228
4229 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4230
4231 * Andreas Jaeger provided a test suite for the math library.
4232
4233 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4234
4235 * Ulrich Drepper added character set conversion functions (iconv).
4236
4237 * Optimized string functions have been added.
4238
4239 * The localedata addon is now part of glibc.
4240
4241 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4242
4243 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4244 daemon for NSS (nscd).
4245
4246 Missing a better place here are some numbers on improvements. Under
4247 Linux 2.1.125 un-tar-ing the kernel sources takes
4248
4249 user system wall
4250
4251 using local files 12.19s 6.88s 22.91s
4252
4253 using NIS 13.92s 8.91s 26.34s
4254
4255 using NIS & nscd 10.37s 7.34s 25.30s
4256
4257 using NIS+ 27.57s 30.37s 640.46s
4258
4259 using NIS+ & nscd 10.25s 7.83s 26.51s
4260
4261 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4262
4263 Keep in mind that non-namelookup related operations dominate above times.
4264 It was just a common complain that using NIS+ unpacking the kernel is
4265 horribly slow.
4266
4267 [1] The old nscd implementation is not available anymore since it was
4268 distributed with glibc up to version 2.0.98 and thus is now replaced.
4269
4270 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4271
4272 * Mark Kettenis provided a Hesiod NSS module.
4273
4274 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4275 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4276
4277 * rcmd can now handle netgroups (Dick Streefland).
4278
4279 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4280 Bambrough.
4281
4282 * Support for the IPv6 protocol has been added to the socket API, as per the
4283 latest draft standards.
4284
4285 * Support for Linux 2.2 has been added.
4286
4287 * Interface changes relative to the latest 2.0.x release:
4288 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4289 addseverity NEW: Unix98
4290 alphasort64 NEW: LFS
4291 argp_err_exit_status NEW: argp, GNU ext
4292 argp_error NEW: argp, GNU ext
4293 argp_failure NEW: argp, GNU ext
4294 argp_help NEW: argp, GNU ext
4295 argp_parse NEW: argp, GNU ext
4296 argp_program_bug_address NEW: argp, GNU ext
4297 argp_program_version NEW: argp, GNU ext
4298 argp_program_version_hook NEW: argp, GNU ext
4299 argp_state_help NEW: argp, GNU ext
4300 argp_usage NEW: argp, GNU ext
4301 authdes_create NEW: Secure RPC
4302 authdes_getucred NEW: Secure RPC
4303 authdes_pk_create NEW: Secure RPC
4304 backtrace NEW: GNU ext.
4305 backtrace_symbols NEW: GNU ext.
4306 backtrace_symbols_fd NEW: GNU ext.
4307 cacos NEW: ISO C 9x
4308 cacosf NEW: ISO C 9x
4309 cacosh NEW: ISO C 9x
4310 cacoshf NEW: ISO C 9x
4311 cacoshl NEW: ISO C 9x
4312 cacosl NEW: ISO C 9x
4313 capget NEW: kernel
4314 capset NEW: kernel
4315 carg NEW: ISO C 9x
4316 cargf NEW: ISO C 9x
4317 cargl NEW: ISO C 9x
4318 casin NEW: ISO C 9x
4319 casinf NEW: ISO C 9x
4320 casinh NEW: ISO C 9x
4321 casinhf NEW: ISO C 9x
4322 casinhl NEW: ISO C 9x
4323 casinl NEW: ISO C 9x
4324 catan NEW: ISO C 9x
4325 catanf NEW: ISO C 9x
4326 catanh NEW: ISO C 9x
4327 catanhf NEW: ISO C 9x
4328 catanhl NEW: ISO C 9x
4329 catanl NEW: ISO C 9x
4330 cbc_crypt NEW: Secure RPC
4331 ccos NEW: ISO C 9x
4332 ccosf NEW: ISO C 9x
4333 ccosh NEW: ISO C 9x
4334 ccoshf NEW: ISO C 9x
4335 ccoshl NEW: ISO C 9x
4336 ccosl NEW: ISO C 9x
4337 cexp NEW: ISO C 9x
4338 cexpf NEW: ISO C 9x
4339 cexpl NEW: ISO C 9x
4340 cimag NEW: ISO C 9x
4341 cimagf NEW: ISO C 9x
4342 cimagl NEW: ISO C 9x
4343 clearerr_locked REMOVED
4344 clntunix_create NEW: sunrpc ext
4345 clog NEW: ISO C 9x
4346 clog10 NEW: ISO C 9x
4347 clog10f NEW: ISO C 9x
4348 clog10l NEW: ISO C 9x
4349 clogf NEW: ISO C 9x
4350 clogl NEW: ISO C 9x
4351 conj NEW: ISO C 9x
4352 conjf NEW: ISO C 9x
4353 conjl NEW: ISO C 9x
4354 cpow NEW: ISO C 9x
4355 cpowf NEW: ISO C 9x
4356 cpowl NEW: ISO C 9x
4357 cproj NEW: ISO C 9x
4358 cprojf NEW: ISO C 9x
4359 cprojl NEW: ISO C 9x
4360 creal NEW: ISO C 9x
4361 crealf NEW: ISO C 9x
4362 creall NEW: ISO C 9x
4363 creat64 NEW: LFS
4364 csin NEW: ISO C 9x
4365 csinf NEW: ISO C 9x
4366 csinh NEW: ISO C 9x
4367 csinhf NEW: ISO C 9x
4368 csinhl NEW: ISO C 9x
4369 csinl NEW: ISO C 9x
4370 csqrt NEW: ISO C 9x
4371 csqrtf NEW: ISO C 9x
4372 csqrtl NEW: ISO C 9x
4373 ctan NEW: ISO C 9x
4374 ctanf NEW: ISO C 9x
4375 ctanh NEW: ISO C 9x
4376 ctanhf NEW: ISO C 9x
4377 ctanhl NEW: ISO C 9x
4378 ctanl NEW: ISO C 9x
4379 des_setparity NEW: Secure RPC
4380 ecb_crypt NEW: Secure RPC
4381 endutxent NEW: Unix98
4382 exp10 NEW: ISO C 9x
4383 exp10f NEW: ISO C 9x
4384 exp10l NEW: ISO C 9x
4385 exp2 NEW: ISO C 9x
4386 exp2f NEW: ISO C 9x
4387 exp2l NEW: ISO C 9x
4388 fattach NEW: STREAMS
4389 fdetach NEW: STREAMS
4390 fdim NEW: ISO C 9x
4391 fdimf NEW: ISO C 9x
4392 fdiml NEW: ISO C 9x
4393 feclearexcept NEW: ISO C 9x
4394 fegetenv NEW: ISO C 9x
4395 fegetexceptflag NEW: ISO C 9x
4396 fegetround NEW: ISO C 9x
4397 feholdexcept NEW: ISO C 9x
4398 feof_locked REMOVED
4399 feraiseexcept NEW: ISO C 9x
4400 ferror_locked REMOVED
4401 fesetenv NEW: ISO C 9x
4402 fesetexceptflag NEW: ISO C 9x
4403 fesetround NEW: ISO C 9x
4404 fetestexcept NEW: ISO C 9x
4405 feupdateenv NEW: ISO C 9x
4406 fflush_locked REMOVED
4407 ffsl NEW: GNU ext.
4408 ffsll NEW: GNU ext.
4409 fgetpos64 NEW: LFS
4410 fgets_unlocked NEW: GNU ext.
4411 fileno_locked REMOVED
4412 fma NEW: ISO C 9x
4413 fmaf NEW: ISO C 9x
4414 fmal NEW: ISO C 9x
4415 fmax NEW: ISO C 9x
4416 fmaxf NEW: ISO C 9x
4417 fmaxl NEW: ISO C 9x
4418 fmin NEW: ISO C 9x
4419 fminf NEW: ISO C 9x
4420 fminl NEW: ISO C 9x
4421 fmtmsg NEW: Unix98
4422 fopen64 NEW: LFS
4423 fputc_locked REMOVED
4424 fputs_unlocked NEW: GNU ext.
4425 fread_unlocked NEW: GNU ext.
4426 freopen64 NEW: LFS
4427 fseeko NEW: Unix98
4428 fsetpos64 NEW: LFS
4429 fstatfs64 NEW: LFS
4430 fstatvfs NEW: Unix98
4431 fstatvfs64 NEW: LFS
4432 ftello NEW: Unix98
4433 ftello64 NEW: LFS
4434 ftruncate64 NEW: LFS
4435 ftw64 NEW: LFS
4436 fwrite_unlocked NEW: GNU ext.
4437 gai_strerror NEW: IPv6
4438 gamma_r REMOVED
4439 gammaf_r REMOVED
4440 gammal_r REMOVED
4441 getchar_locked REMOVED
4442 getdate NEW: Unix98
4443 getdate_err NEW: Unix98
4444 getdate_r NEW: GNU ext.
4445 getmsg NEW: STREAMS
4446 getnameinfo NEW: IPv6
4447 getnetname NEW: Secure RPC
4448 getpmsg NEW: STREAMS
4449 getpt NEW: Unix98 PTY
4450 getrlimit64 NEW: LFS
4451 getutxent NEW: Unix98
4452 getutxid NEW: Unix98
4453 getutxline NEW: Unix98
4454 glob64 NEW: GNU ext.
4455 globfree64 NEW: GNU ext.
4456 gnu_get_libc_release NEW: GNU ext.
4457 gnu_get_libc_version NEW: GNU ext.
4458 grantpt NEW: Unix98 PTY
4459 host2netname NEW: Secure RPC
4460 iconv NEW: iconv
4461 iconv_close NEW: iconv
4462 iconv_open NEW: iconv
4463 if_freenameindex NEW: IPv6
4464 if_indextoname NEW: IPv6
4465 if_nameindex NEW: IPv6
4466 if_nametoindex NEW: IPv6
4467 in6addr_any NEW: IPv6
4468 in6addr_loopback NEW: IPv6
4469 inet6_isipv4mapped NEW: IPv6
4470 isastream NEW: STREAMS
4471 iswblank NEW: GNU ext.
4472 key_decryptsession NEW: Secure RPC
4473 key_decryptsession_pk NEW: Secure RPC
4474 key_encryptsession NEW: Secure RPC
4475 key_encryptsession_pk NEW: Secure RPC
4476 key_gendes NEW: Secure RPC
4477 key_get_conv NEW: Secure RPC
4478 key_secretkey_is_set NEW: Secure RPC
4479 key_setnet NEW: Secure RPC
4480 key_setsecret NEW: Secure RPC
4481 llrint NEW: ISO C 9x
4482 llrintf NEW: ISO C 9x
4483 llrintl NEW: ISO C 9x
4484 llround NEW: ISO C 9x
4485 llroundf NEW: ISO C 9x
4486 llroundl NEW: ISO C 9x
4487 log2 NEW: ISO C 9x
4488 log2f NEW: ISO C 9x
4489 log2l NEW: ISO C 9x
4490 lrint NEW: ISO C 9x
4491 lrintf NEW: ISO C 9x
4492 lrintl NEW: ISO C 9x
4493 lround NEW: ISO C 9x
4494 lroundf NEW: ISO C 9x
4495 lroundl NEW: ISO C 9x
4496 lseek64 NEW: LFS
4497 makecontext NEW: Unix98
4498 mempcpy NEW: GNU ext.
4499 mmap64 NEW: LFS
4500 moncontrol REMOVED
4501 modify_ldt NEW: kernel
4502 nan NEW: ISO C 9x
4503 nanf NEW: ISO C 9x
4504 nanl NEW: ISO C 9x
4505 nearbyint NEW: ISO C 9x
4506 nearbyintf NEW: ISO C 9x
4507 nearbyintl NEW: ISO C 9x
4508 netname2host NEW: Secure RPC
4509 netname2user NEW: Secure RPC
4510 nexttoward NEW: ISO C 9x
4511 nexttowardf NEW: ISO C 9x
4512 nexttowardl NEW: ISO C 9x
4513 nftw NEW: Unix98
4514 nftw64 NEW: LFS
4515 open64 NEW: LFS
4516 passwd2des NEW: Secure RPC
4517 pow10 NEW: GNU ext.
4518 pow10f NEW: GNU ext.
4519 pow10l NEW: GNU ext.
4520 pread NEW: Unix98
4521 pread64 NEW: LFS
4522 printf_size NEW: GNU ext.
4523 printf_size_info NEW: GNU ext.
4524 profil_counter REMOVED
4525 pthread_mutexattr_getkind_np REPLACED
4526 pthread_mutexattr_setkind_np REPLACED
4527 ptsname NEW: Unix98 PTY
4528 ptsname_r NEW: Unix98 PTY
4529 putc_locked REMOVED
4530 putchar_locked REMOVED
4531 putgrent NEW: GNU ext.
4532 putmsg NEW: STREAMS
4533 putpmsg NEW: STREAMS
4534 pututxline NEW: Unix98
4535 pwrite NEW: Unix98
4536 pwrite64 NEW: LFS
4537 readdir64 NEW: LFS
4538 readdir64_r NEW: LFS
4539 remquo NEW: ISO C 9x
4540 remquof NEW: ISO C 9x
4541 remquol NEW: ISO C 9x
4542 round NEW: ISO C 9x
4543 roundf NEW: ISO C 9x
4544 roundl NEW: ISO C 9x
4545 rtime NEW: GNU ext.
4546 scalbln NEW: ISO C 9x
4547 scalblnf NEW: ISO C 9x
4548 scalblnl NEW: ISO C 9x
4549 scandir64 NEW: LFS
4550 sendfile NEW: kernel
4551 setcontext NEW: Unix98
4552 setrlimit64 NEW: LFS
4553 setutxent NEW: Unix98
4554 sighold NEW: Unix98
4555 sigignore NEW: Unix98
4556 sigqueue NEW: POSIX.1b
4557 sigrelse NEW: Unix98
4558 sigset NEW: POSIX.1b
4559 sigtimedwait NEW: POSIX.1b
4560 sigwaitinfo NEW: POSIX.1b
4561 sincos NEW: GNU ext.
4562 sincosf NEW: GNU ext.
4563 sincosl NEW: GNU ext.
4564 statfs64 NEW: LFS
4565 statvfs NEW: Unix98
4566 statvfs64 NEW: LFS
4567 strcasestr NEW: GNU ext.
4568 strtoimax NEW: ISO C 9x
4569 strtoumax NEW: ISO C 9x
4570 strverscmp NEW: GNU ext.
4571 svcauthdes_stats NEW: Secure RPC
4572 svcunix_create NEW: sunrpc etx
4573 svcunixfd_create NEW: sunrpc ext
4574 swapcontext NEW: Unix98
4575 tcgetsid NEW: Unix98 PTY
4576 tdestroy NEW: GNU ext.
4577 tgamma NEW: ISO C 9x
4578 tgammaf NEW: ISO C 9x
4579 tgammal NEW: ISO C 9x
4580 tmpfile64 NEW: LFS
4581 trunc NEW: ISO C 9x
4582 truncate64 NEW: LFS
4583 truncf NEW: ISO C 9x
4584 truncl NEW: ISO C 9x
4585 umount2 NEW: kernel
4586 unlockpt NEW: Unix98 PTY
4587 updwtmpx NEW: Unix98
4588 user2netname NEW: Secure RPC
4589 utmpxname NEW: Unix98
4590 versionsort NEW: GNU ext.
4591 versionsort64 NEW: GNU ext.
4592 waitid NEW: Unix98
4593 wcscasecmp NEW: GNU ext.
4594 wcsncasecmp NEW: GNU ext.
4595 wcsnlen NEW: GNU ext.
4596 wcstoimax NEW: ISO C 9x
4597 wcstoll NEW: ISO C 9x
4598 wcstoull NEW: ISO C 9x
4599 wcstoumax NEW: ISO C 9x
4600 wcswcs NEW: Unix98
4601 wordexp NEW: POSIX.2
4602 wordfree NEW: POSIX.2
4603 write_profiling REMOVED
4604 xdecrypt NEW: Secure RPC
4605 xdr_authdes_cred NEW: Secure RPC
4606 xdr_authdes_verf NEW: Secure RPC
4607 xdr_cryptkeyarg NEW: Secure RPC
4608 xdr_cryptkeyarg2 NEW: Secure RPC
4609 xdr_cryptkeyres NEW: Secure RPC
4610 xdr_getcredres NEW: Secure RPC
4611 xdr_key_netstarg NEW: Secure RPC
4612 xdr_key_netstres NEW: Secure RPC
4613 xdr_keybuf NEW: Secure RPC
4614 xdr_keystatus NEW: Secure RPC
4615 xdr_netnamestr NEW: Secure RPC
4616 xdr_sizeof NEW: Secure RPC
4617 xdr_unixcred NEW: sunrpc ext
4618 xencrypt NEW: Secure RPC
4619 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4620 \f
4621 Version 2.0.6
4622
4623 * more bug fixes
4624
4625 \f
4626 Version 2.0.5
4627
4628 * more bug fixes
4629
4630 * inet_ntoa is thread-safe
4631
4632 * updwtmp is moved from libutil to libc
4633
4634 * rewrite of cbrt function
4635
4636 * update of timezone data
4637 \f
4638 Version 2.0.4
4639
4640 * more bug fixes
4641 \f
4642 Version 2.0.3
4643
4644 * more bug fixes
4645 \f
4646 Version 2.0.2
4647
4648 * more bug fixes
4649
4650 * add atoll function
4651
4652 * fix complex problems in Berkeley DB code
4653
4654 * fix math functions
4655 \f
4656 Version 2.0.1
4657
4658 * fixed lots of header problems (especially Linux/GNU specific)
4659
4660 * dynamic loader preserves all registers
4661
4662 * Roland McGrath provided support for handling of auxiliary objects in
4663 the ELF dynamic loader.
4664
4665 * support for parallel builds is improved
4666 \f
4667 Version 2.0
4668
4669 * GNU extensions are no longer declared by default. To enable them you
4670 must define the macro `_GNU_SOURCE' in your program or compile with
4671 `-D_GNU_SOURCE'.
4672
4673 * The library has changed from using GNU ld symbol aliases to using weak
4674 symbols where available. The ELF object file format supports weak
4675 symbols; GNU ld also supports weak symbols in the a.out format. (There
4676 is also now support for other GNU ld extensions in ELF. Use the
4677 `--with-elf' option to configure to indicate you have ELF, and
4678 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4679 of many files which contained only symbol aliases, reducing the size of
4680 the source and the compiled library; many other files were renamed to
4681 less cryptic names previously occupied by the symbol alias files.
4682 There is a new header file <elf.h> for programs which operate on
4683 files in the ELF format.
4684
4685 * Converted to Autoconf version 2, so `configure' has more options.
4686 Run `configure --help' to see the details.
4687
4688 * The library can now be configured to build profiling, highly-optimized
4689 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4690 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4691 `configure' enable building these extra libraries. The shared library is
4692 built by default when using both ELF and GNU ld. When shared libraries
4693 are enabled, the new library `-ldl' is available for arbitrary run-time
4694 loading of shared objects; its interface is defined in <dlfcn.h>. The
4695 new header file <link.h> gives access to the internals of the run-time
4696 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4697 application of same name on other systems and it provides information
4698 about dynamically linked binaries.
4699
4700 * The C library now provides the run-time support code for profiling
4701 executables compiled with `-pg'. Programs can control the profiling code
4702 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4703 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4704 the support for this file format was contributed by David Mosberger-Tang.
4705
4706 * The math code has been replaced with a math library based on fdlibm from
4707 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4708 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4709 functions. The math functions now reside in a separate library, so
4710 programs using them will need to use `-lm' their linking commands.
4711
4712 * John C. Bowman contributed optimized ix87 assembler inline functions.
4713
4714 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4715 mechanism similar to that found in Solaris 2. This is now used for the
4716 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4717 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4718 file controls what services are used for each individual database. This
4719 works by loading shared libraries with names specified in `nsswitch.conf',
4720 so service modules can be changed or added at any time without even
4721 relinking any program. Currently there are the file, db, and NIS based
4722 NSS services available.
4723
4724 * The new functions `strtoq' and `strtouq' parse integer values from
4725 strings, like `strtol' and `strtoul', but they return `long long int' and
4726 `unsigned long long int' values, respectively (64-bit quantities).
4727
4728 * The new functions `strtof' and `strtold' parse floating-point values from
4729 strings, like `strtod', but they return `float' and `long double' values,
4730 respectively (on some machines `double' and `long double' are the same).
4731
4732 * Ulrich Drepper has contributed new implementations of the floating-point
4733 printing and reading code used in the `printf' family of functions and
4734 `strtod', `strtof', and `strtold'. These new functions are perfectly
4735 accurate, and much faster than the old ones.
4736
4737 * The implementation of the POSIX locale model was completely rewritten by
4738 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4739 to compile the POSIX locale definition.
4740
4741 * The former dummy implementations of the strcoll and strxfrm function are
4742 now replaced by fully functional code contributed by Ulrich Drepper. The
4743 collation information comes from the POSIX locale definitions.
4744
4745 * The new header <langinfo.h> defines an interface for accessing
4746 various locale-dependent data (using the locale chosen with `setlocale').
4747
4748 * Ulrich Drepper has contributed a new suite of functions for operation on
4749 wide-character and multibyte-character strings, in <wchar.h>;
4750 and classification and case conversion of wide characters, in <wctype.h>.
4751 These new functions are conforming to the ISO C, Amendement 1 specification.
4752
4753 * There is now a second implementation of the standard I/O library available.
4754 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4755 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4756
4757 * You can now use positional parameter specifications in format strings
4758 for the `printf' and `scanf' families of functions. For example,
4759 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4760 ``Number 6, Mr Jones''. This is mainly useful when providing different
4761 format strings for different languages, whose grammars may dictate
4762 different orderings of the values being printed. To support this
4763 feature, the interface for `register_printf_handler' has changed; see
4764 the header file <printf.h> for details.
4765
4766 * The `printf' and `scanf' families of functions now understand a new
4767 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4768 to group numbers as indicated by the locale; for `scanf' and friends, this
4769 says to accept as valid only a number with all the proper grouping
4770 separators in the right places. In the default "C" locale, numbers are
4771 not grouped; but locales for specific countries will define the usual
4772 conventions (i.e. separate thousands with `,' in the US locale).
4773
4774 * The pgrp functions have been regularized, slightly incompatibly but much
4775 less confusingly. The core functions are now `getpgid' and `setpgid',
4776 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4777 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4778 provided for compatibility. There is no longer an incompatible `getpgrp'
4779 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4780
4781 * The new header file <fts.h> and suite of functions simplify programs that
4782 operate on directory trees. This code comes from 4.4 BSD.
4783
4784 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4785 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4786 NSS scheme used in glibc.
4787
4788 * There is a new malloc debugging hook `__memalign_hook'.
4789
4790 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4791 `unsigned int' in <sys/types.h>. These are for compatibility only and
4792 their use is discouraged.
4793
4794 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4795 done differently, so that it works even without GNU ld.
4796
4797 * New function `euidaccess' checks allowed access to a file like `access',
4798 but using the effective IDs instead of the real IDs.
4799
4800 * The time zone data files have been updated for the latest and greatest
4801 local time conventions of the countries of the world.
4802
4803 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4804 see <dirent.h>.
4805
4806 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4807 for formatting floating-point numbers. They are provided only for
4808 compatibility; new programs should use `sprintf' instead. There are
4809 also equivalent function for the `long double' floating-point type and
4810 all functions also exist in a reentrant form.
4811
4812 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4813 functions for maintaining the login-record files (primarily of use to
4814 system programs such as `login'), and convenient functions for
4815 allocating and initializing a pseudo-terminal (pty) device.
4816
4817 * Ulrich Drepper has contributed new support for System V style
4818 shared memory and IPC on systems that support it.
4819
4820 * Ulrich Drepper has contributed several miscellaneous new functions found
4821 in System V: The `hsearch' family of functions provide an effective
4822 implementation of hash tables; `a64l' and `l64a' provide a very simple
4823 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4824 number generator.
4825
4826 * Ulrich Drepper has contributed new reentrant counterparts for the
4827 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4828
4829 * Ulrich Drepper has contributed new, highly-optimized versions of several
4830 string functions for the i486/Pentium family of processors.
4831
4832 * Ulrich Drepper has updated the Linux-specific code, based largely
4833 on work done in Hongjiu Lu's version of GNU libc for Linux.
4834 The GNU library now supports Linux versions 2.0.10 and later,
4835 using the ELF object file format (i[3456]86-*-linux).
4836
4837 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4838
4839 * David Mosberger-Tang and Richard Henderson have ported the C library
4840 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4841 dynamic linking support for ELF/Alpha.
4842
4843 * Richard Henderson contributed several Alpha optimized assembler function
4844 for arithmetic and string handling.
4845
4846 * Ulrich Drepper has contributed a new set of message catalog functions to
4847 support multiple languages using the <libintl.h> interface, for use with
4848 his new package GNU gettext. Translation volunteers have contributed
4849 catalogs of the library's messages in Spanish, German, and Korean.
4850
4851 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4852 program and the `catgets' function for reading the catalog files it
4853 creates. (The <libintl.h> interface is preferred; we include the
4854 <nl_types.h> interface using `catgets' only for source compatibility with
4855 programs already written to use it.)
4856
4857 * New header file <values.h> gives SVID-compatible names for <limits.h>
4858 constants.
4859
4860 * Various new macros, declarations, and small header files for compatibility
4861 with 4.4 BSD.
4862
4863 * New function `group_member' is a convenient way to check if a process has
4864 a given effective group ID.
4865
4866 * When using GCC 2.7 and later, the socket functions are now declared in a
4867 special way so that passing an argument of type `struct sockaddr_in *',
4868 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4869 `struct sockaddr *' type, does not generate a type-clash warning.
4870
4871 * New function `error' declared in header file <error.h> is a convenient
4872 function for printing error messages and optionally exiting; this is the
4873 canonical function used in GNU programs. The new functions `err', `warn',
4874 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4875 doing the same thing.
4876
4877 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4878 POSIX.2 `glob' function to do ~ and {...} expansion.
4879
4880 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4881 `clearenv' which is used in POSIX.9 is also available.
4882
4883 * New function `getsid' returns session ID number on systems that support it.
4884
4885 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4886 files <db.h> and <mpool.h> provide a rich set of functions for several
4887 types of simple databases stored in memory and in files, and <ndbm.h> is
4888 an old `ndbm'-compatible interface using the `db' functions. Link with
4889 `-ldb' to get these functions.
4890
4891 * New macro `strdupa' copies a string like `strdup', but uses local stack
4892 space from `alloca' instead of dynamic heap space from `malloc'.
4893
4894 * New function `strnlen' is like `strlen' but searches only a given maximum
4895 number of characters for the null terminator. `stpncpy', `strndup' and
4896 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4897 function.
4898
4899 * New function `statfs' in header <sys/statfs.h>.
4900
4901 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4902 provide convenient functions for operating on blocks of null-terminated
4903 strings.
4904
4905 * A new suite of functions in <utmp.h> handle all the details of reading
4906 and writing the utmp file.
4907
4908 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4909 Thorsten Kukuk.
4910
4911 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4912 completely POSIX compliant and also implemented the extended functionality
4913 to handle alternate digit representation and alternate era date formats.
4914
4915 * Ulrich Drepper provided an implementation of the `strptime' function
4916 defined in XPG4.2 which transforms a string into a `struct tm' value.
4917
4918 * Paul Eggert provided the tzselect shell script as part of the timezone
4919 code. The shell script makes it easy to select the correct timezone
4920 specification.
4921
4922 * The implementation of the malloc family of functions is completely replaced
4923 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4924 The implementation uses the mmap function (if available) and it is
4925 optimized for the use in multi threaded programs.
4926
4927 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4928 functions. This new functionality is usable by specifying a special
4929 salt string and it is compatible with implementation on *BSD systems.
4930
4931 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4932 `getsubopt' to handle second level command line options, `bsd_signal'
4933 to access BSD style `signal' functionality, the obsolete `regexp' style
4934 expression matcher.
4935
4936 * the `lchown' function is available on system which support this
4937 functionality.
4938
4939 * The implementation of the shadow password handling function was contributed
4940 by Ulrich Drepper.
4941
4942 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4943
4944 * POSIX.1g support was added. The <sys/select.h> header is available,
4945 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4946 implementation of `getaddrinfo'.
4947 \f
4948 Version 1.09
4949
4950 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4951
4952 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4953 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4954
4955 * The new function `daemon' from 4.4 BSD is useful for server programs that
4956 want to put themselves in the background.
4957
4958 * Joel Sherrill has contributed support for several standalone boards that
4959 run without an operating system.
4960
4961 * `printf', `scanf' and friends now accept a `q' type modifier for long
4962 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4963
4964 * All of the code taken from BSD (notably most of the math and networking
4965 routines) has been updated from the BSD 4.4-Lite release.
4966
4967 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4968
4969 * The new functions `getdomainname' and `setdomainname' fetch or change the
4970 YP/NIS domain name. These are system calls which exist on systems which
4971 have YP (aka NIS).
4972
4973 * The time zone data files have been updated for the latest international
4974 conventions.
4975
4976 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4977 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4978 \f
4979 Version 1.08
4980
4981 * The C library now includes support for Sun RPC, from Sun's free
4982 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4983 are included. (There is still no support for YP.)
4984
4985 * Tom Quinn has contributed a port of the C library to SGI machines running
4986 Irix 4 (mips-sgi-irix4).
4987
4988 * The new `lockf' function is a simplified interface to the locking
4989 facilities of `fcntl', included for compatibility.
4990
4991 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4992
4993 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4994
4995 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4996 compatibility.
4997
4998 * The error message printed by `assert' for a failed assertion now includes
4999 the name of the program (if using GNU ld) and the name of the calling
5000 function (with versions of GCC that support this).
5001
5002 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5003
5004 * The library now includes the <sys/mman.h> header file and memory
5005 management functions `mmap', `munmap', `mprotect', `msync', and
5006 `madvise', on systems that support those facilities.
5007
5008 * The interface for `mcheck' has changed slightly: the function called to
5009 abort the program when an allocation inconsistency is detected now takes
5010 an argument that indicates the type of failure. The new function
5011 `mprobe' lets you request a consistency check for a particular block at
5012 any time (checks are normally done only when you call `free' or `realloc'
5013 on a block).
5014
5015 * It is now possible to easily cross-compile the C library, building on one
5016 system a library to run on another machine and/or operating system. All
5017 you need to do is set the variable `HOST_CC' in `configparms' to the
5018 native compiler for programs to run on the machine you are building on (a
5019 few generator programs are used on Unix systems); set `CC' to the
5020 cross-compiler.
5021
5022 * The new function `fexecve' (only implemented on the GNU system) executes
5023 a program file given a file descriptor already open on the file.
5024 \f
5025 Version 1.07
5026
5027 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5028 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5029
5030 * You can set the variable `libprefix' in `configparms' to specify a prefix
5031 to be prepended to installed library files; this makes it easy to install
5032 the GNU C library to be linked as `-lgnuc' or whatever.
5033
5034 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5035 copies a limited number of characters from a string, and returns the
5036 address of the last character written.
5037
5038 * You no longer need to check for whether the installed `stddef.h' is
5039 compatible with the GNU C library. configure now checks for you.
5040
5041 * You can now define a per-stream `fileno' function to convert the
5042 stream's cookie into an integral file descriptor.
5043
5044 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5045 allocates zero bytes of storage, and returns a unique pointer which
5046 you can pass to `realloc' or `free'. The behavior is undefined if
5047 you dereference this pointer.
5048
5049 * The C library now runs on Sony NEWS m68k machines running either
5050 NewsOS 3 or NewsOS 4.
5051
5052 * The new `syscall' function is a system-dependent primitive function
5053 for invoking system calls. It has the canonical behavior on Unix
5054 systems, including unreliable return values for some calls (such as
5055 `pipe', `fork' and `getppid').
5056
5057 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5058 to `EAGAIN', which is the preferred name. On systems whose kernels
5059 use two distinct codes, the C library now translates EWOULDBLOCK to
5060 EAGAIN in every system call function.
5061 \f
5062 Version 1.06
5063
5064 * The GNU C Library Reference Manual is now distributed with the library.
5065 `make dvi' will produce a DVI file of the printed manual.
5066 `make info' will produce Info files that you can read on line using C-h i
5067 in Emacs or the `info' program.
5068 Please send comments on the manual to bug-glibc-manual@gnu.org.
5069
5070 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5071
5072 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5073
5074 * Jason Merrill has contributed a port to the Sequent Symmetry running
5075 Dynix version 3 (i386-sequent-dynix).
5076
5077 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5078 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5079
5080 * New function `memory_warnings' lets you arrange to get warnings when
5081 malloc is running out of memory to allocate, like Emacs gives you.
5082
5083 * The C library now contains the relocating allocator used in Emacs 19 for
5084 its editing buffers. This allocator (ralloc) minimizes allocation
5085 overhead and fragmentation by moving allocated regions around whenever it
5086 needs to. You always refer to a ralloc'd region with a "handle" (a
5087 pointer to a pointer--an object of type `void **').
5088
5089 * There is a new `printf' format: `%m' gives you the string corresponding
5090 to the error code in `errno'.
5091
5092 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5093 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5094 pass, the `a' modifier says to fill in a `char **' you pass with a
5095 malloc'd string.
5096
5097 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5098 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5099 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5100
5101 * `mkstemp' is a traditional Unix function to atomically create and open a
5102 uniquely-named temporary file.
5103 \f
5104 Version 1.05
5105
5106 * The standard location for the file that says what the local timezone is
5107 has changed again. It is now `/usr/local/etc/localtime' (or more
5108 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5109
5110 * The distribution no longer contains any files with names longer than 14
5111 characters.
5112
5113 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5114 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5115
5116 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5117 \f
5118 Version 1.04
5119
5120 * The configuration process has changed quite a bit. The `configure'
5121 script is now used just like the configuration scripts for other GNU
5122 packages. The `sysdeps' directory hierarchy is much rearranged.
5123 The file `INSTALL' explains the new scheme in detail.
5124
5125 * The header files no longer need to be processed into ANSI C and
5126 traditional C versions. There is just one set of files to install, and
5127 it will work with ANSI or old C compilers (including `gcc -traditional').
5128
5129 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5130 MIPS DECStation running Ultrix 4.
5131
5132 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5133 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5134 made itself into a shared library.
5135
5136 * Yet further improved support for the i386, running 4.3 BSD-like systems
5137 (such as Mach 3 with the Unix single-server), or System V.
5138
5139 * New function `strncasecmp' to do case-insensitive string comparison
5140 with limited length.
5141
5142 * New function `strsep' is a reentrant alternative to `strtok'.
5143
5144 * New functions `scandir' and `alphasort' for searching directories.
5145
5146 * New function `setenv' is a better interface to `putenv'.
5147
5148 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5149 function for traversing a directory tree.
5150
5151 * The GNU obstack package is now also part of the C library.
5152 The new function `open_obstack_stream' creates a stdio stream that
5153 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5154 formatted output directly to an obstack.
5155
5156 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5157 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5158
5159 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5160
5161 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5162 things to your strings.
5163
5164 * There are some new test programs: test-fseek, testmb, and testrand.
5165
5166 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5167 These ports are not finished, but are a good starting place for really
5168 supporting those systems.
5169
5170 * `/etc/localtime' is now the standard location for the file that says what
5171 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5172 This follows the general principle that `/etc' is the place for all local
5173 configuration files.
5174
5175 * The C library header files now use `extern "C"' when used by the C++
5176 compiler, so the C library should now work with C++ code.
5177
5178 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5179 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5180 in <strings.h>.)
5181
5182 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5183 function which uses the `merge sort' algorithm, and is said to be
5184 significantly faster than the old GNU `qsort' function. Merge sort is
5185 now the standard `qsort' function. The new algorithm can require a lot
5186 of temporary storage; so, the old sorting function is called when the
5187 required storage is not available.
5188
5189 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5190 provides the Unix `crypt' function, plus some other entry points.
5191
5192 * The code and header files taken from 4.4 BSD have been updated with the
5193 latest files released from Berkeley.
5194 \f
5195 ----------------------------------------------------------------------
5196 Copyright information:
5197
5198 Copyright (C) 1992-2018 Free Software Foundation, Inc.
5199
5200 Permission is granted to anyone to make or distribute verbatim copies
5201 of this document as received, in any medium, provided that the
5202 copyright notice and this permission notice are preserved,
5203 thus giving the recipient permission to redistribute in turn.
5204
5205 Permission is granted to distribute modified versions
5206 of this document, or of portions of it,
5207 under the above conditions, provided also that they
5208 carry prominent notices stating who last changed them.
5209 \f
5210 Local variables:
5211 version-control: never
5212 fill-column: 76
5213 End: