]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
f054dc0433a8c67c60d5752979017ac71c94dccd
[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 2.29
9
10 Major new features:
11
12 * A new convenience target has been added for distribution maintainers
13 to build and install all locales as directories with files. The new
14 target is run by issuing the following command in your build tree:
15 'make localedata/install-locale-files', with an optional DESTDIR
16 to set the install root if you wish to install into a non-default
17 configured location.
18
19 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
20
21 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
22 for _GNU_SOURCE, to match BSD environments.
23
24 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
25 indicates that it will abort the transaction prior to entering the kernel
26 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
27 suspended, and this caused some undefined side-effects issues by aborting
28 transactions manually. Glibc avoided it by abort transactions manually on
29 each syscall, but it lead to performance issues on newer kernels where the
30 HTM state is saved and restore lazily (the state being saved even when the
31 process actually does not use HTM).
32
33 Deprecated and removed features, and other changes affecting compatibility:
34
35 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
36 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
37
38 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
39 in <sys/procfs.h>, has been corrected to match the type actually used by
40 the Linux kernel. This affects the size and layout of that structure on
41 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
42
43 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
44 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
45 defined in <sys/procfs.h>, has been corrected to match the type actually
46 used by the Linux kernel. This affects the size and layout of those
47 structures.
48
49 Changes to build and runtime requirements:
50
51 [Add changes to build and runtime requirements here]
52
53 Security related changes:
54
55 [Add security related changes here]
56
57 The following bugs are resolved with this release:
58
59 [The release manager will add the list generated by
60 scripts/list-fixed-bugs.py just before the release.]
61
62 \f
63 Version 2.28
64
65 Major new features:
66
67 * The localization data for ISO 14651 is updated to match the 2016
68 Edition 4 release of the standard, this matches data provided by
69 Unicode 9.0.0. This update introduces significant improvements to the
70 collation of Unicode characters. This release deviates slightly from
71 the standard in that the collation element ordering for lowercase and
72 uppercase LATIN script characters is adjusted to ensure that regular
73 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
74 is not matched by [a-z]. With the update many locales have been
75 updated to take advantage of the new collation information. The new
76 collation information has increased the size of the compiled locale
77 archive or binary locales.
78
79 * The GNU C Library can now be compiled with support for Intel CET, AKA
80 Intel Control-flow Enforcement Technology. When the library is built
81 with --enable-cet, the resulting glibc is protected with indirect
82 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
83 compatible with all existing executables and shared libraries. This
84 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
85 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
86 capable of multi-byte NOPs, like x86-64 processors as well as Intel
87 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
88 x86_64 and x32 on non-CET processors. --enable-cet has been tested
89 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
90 validated for i686.
91
92 * The GNU C Library now has correct support for ABSOLUTE symbols
93 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
94 relocated incorrectly or in some cases discarded. The GNU linker can
95 make use of the newer semantics, but it must communicate it to the
96 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
97 field) to indicate such support is required.
98
99 * Unicode 11.0.0 Support: Character encoding, character type info, and
100 transliteration tables are all updated to Unicode 11.0.0, using
101 generator scripts contributed by Mike FABIAN (Red Hat).
102
103 * <math.h> functions that round their results to a narrower type are added
104 from TS 18661-1:2014 and TS 18661-3:2015:
105
106 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
107 fMxaddfNx functions.
108
109 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
110 fMxsubfNx functions.
111
112 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
113 fMxmulfNx functions.
114
115 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
116 fMxdivfNx functions.
117
118 * Two grammatical forms of month names are now supported for the following
119 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
120 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
121 support two grammatical forms in abbreviated month names: Catalan, Greek,
122 and Kashubian.
123
124 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
125 include the support for two grammatical forms of month names.
126
127 * Building and running on GNU/Hurd systems now works without out-of-tree
128 patches.
129
130 * The renameat2 function has been added, a variant of the renameat function
131 which has a flags argument. If the flags are zero, the renameat2 function
132 acts like renameat. If the flag is not zero and there is no kernel
133 support for renameat2, the function will fail with an errno value of
134 EINVAL. This is different from the existing gnulib function renameatu,
135 which performs a plain rename operation in case of a RENAME_NOREPLACE
136 flags and a non-existing destination (and therefore has a race condition
137 that can clobber the destination inadvertently).
138
139 * The statx function has been added, a variant of the fstatat64
140 function with an additional flags argument. If there is no direct
141 kernel support for statx, glibc provides basic stat support based on
142 the fstatat64 function.
143
144 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
145 library if installed. libidn2 version 2.0.5 or later is recommended. If
146 libidn2 is not available, internationalized domain names are not encoded
147 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
148 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
149 with an encoding error.) Flags which used to change the IDN encoding and
150 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
151 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
152 deprecated. They no longer have any effect.
153
154 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
155 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
156 range of ELF gABI expressions including such constructs as
157 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
158 have been further restricted, and where in the past a dynamic string
159 token sequence may have been interpreted as a literal string it will
160 now cause a load failure. These load failures were always considered
161 unspecified behaviour from the perspective of the dynamic loader, and
162 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
163 results in a load failure now.
164
165 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
166 implementation includes all the standard functions provided by
167 <threads.h>:
168
169 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
170 thrd_detach, thrd_exit, and thrd_join for thread management.
171
172 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
173 mtx_destroy for mutual exclusion.
174
175 - call_once for function call synchronization.
176
177 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
178 cnd_wait for conditional variables.
179
180 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
181
182 Application developers must link against libpthread to use ISO C threads.
183
184 Deprecated and removed features, and other changes affecting compatibility:
185
186 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
187 installed. Software that was using either header should be updated to
188 use standard <stdio.h> interfaces instead.
189
190 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
191 This was never required by the C standard, and the macros just expanded
192 to call alternative names for the same functions. If you hoped getc and
193 putc would provide performance improvements over fgetc and fputc, instead
194 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
195 necessary, flockfile and funlockfile.
196
197 * All stdio functions now treat end-of-file as a sticky condition. If you
198 read from a file until EOF, and then the file is enlarged by another
199 process, you must call clearerr or another function with the same effect
200 (e.g. fseek, rewind) before you can read the additional data. This
201 corrects a longstanding C99 conformance bug. It is most likely to affect
202 programs that use stdio to read interactive input from a terminal.
203 (Bug #1190.)
204
205 * The macros 'major', 'minor', and 'makedev' are now only available from
206 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
207 headers that happen to include <sys/types.h>. These macros are rarely
208 used, not part of POSIX nor XSI, and their names frequently collide with
209 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
210 further explanation.
211
212 <sys/sysmacros.h> is a GNU extension. Portable programs that require
213 these macros should first include <sys/types.h>, and then include
214 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
215
216 * The tilegx*-*-linux-gnu configurations are no longer supported.
217
218 * The obsolete function ustat is no longer available to newly linked
219 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
220 function has been deprecated in favor of fstatfs and statfs.
221
222 * The obsolete function nfsservctl is no longer available to newly linked
223 binaries. This function was specific to systems using the Linux kernel
224 and could not usefully be used with the GNU C Library on systems with
225 version 3.1 or later of the Linux kernel.
226
227 * The obsolete function name llseek is no longer available to newly linked
228 binaries. This function was specific to systems using the Linux kernel
229 and was not declared in a header. Programs should use the lseek64 name
230 for this function instead.
231
232 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
233 getaddrinfo and getnameinfo functions have been deprecated. The behavior
234 previously selected by them is now always enabled.
235
236 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
237 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
238 restriction (rejecting '_' in host names, among other things) has been
239 removed, for increased compatibility with non-IDN name resolution.
240
241 * The fcntl function now have a Long File Support variant named fcntl64. It
242 is added to fix some Linux Open File Description (OFD) locks usage on non
243 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
244 fcntl and LFS support is handled transparently. Also for Linux, the OFD
245 locks act as a cancellation entrypoint.
246
247 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
248 ecb_crypt, and des_setparity are no longer available to newly linked
249 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
250 longer installed. These functions encrypted and decrypted data with the
251 DES block cipher, which is no longer considered secure. Software that
252 still uses these functions should switch to a modern cryptography library,
253 such as libgcrypt.
254
255 * Reflecting the removal of the encrypt and setkey functions above, the
256 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
257 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
258 enabled.
259
260 * The obsolete function fcrypt is no longer available to newly linked
261 binaries. It was just another name for the standard function crypt,
262 and it has not appeared in any header file in many years.
263
264 * We have tentative plans to hand off maintenance of the passphrase-hashing
265 library, libcrypt, to a separate development project that will, we hope,
266 keep up better with new passphrase-hashing algorithms. We will continue
267 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
268 'crypt_r' should not need to change at all; however, distributions will
269 need to install <crypt.h> and libcrypt from a separate project.
270
271 In this release, if the configure option --disable-crypt is used, glibc
272 will not install <crypt.h> or libcrypt, making room for the separate
273 project's versions of these files. The plan is to make this the default
274 behavior in a future release.
275
276 Changes to build and runtime requirements:
277
278 GNU make 4.0 or later is now required to build glibc.
279
280 Security related changes:
281
282 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
283 been fixed by removing the glibc-internal IDNA implementation and using
284 the system-provided libidn2 library instead. Originally reported by Hanno
285 Böck and Christian Weisgerber.
286
287 CVE-2017-18269: An SSE2-based memmove implementation for the i386
288 architecture could corrupt memory. Reported by Max Horn.
289
290 CVE-2018-11236: Very long pathname arguments to realpath function could
291 result in an integer overflow and buffer overflow. Reported by Alexey
292 Izbyshev.
293
294 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
295 architecture could write beyond the target buffer, resulting in a buffer
296 overflow. Reported by Andreas Schwab.
297
298 The following bugs are resolved with this release:
299
300 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
301 [6889] manual: 'PWD' mentioned but not specified
302 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
303 when __WORDSIZE != 64
304 [13762] regex: re_search etc. should return -2 on memory exhaustion
305 [13888] build: /tmp usage during testing
306 [13932] math: dbl-64 pow unexpectedly slow for some inputs
307 [14092] nptl: Support C11 threads
308 [14095] localedata: Review / update collation data from Unicode / ISO
309 14651
310 [14508] libc: -Wformat warnings
311 [14553] libc: Namespace pollution loff_t in sys/types.h
312 [14890] libc: Make NT_PRFPREG canonical.
313 [15105] libc: Extra PLT references with -Os
314 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
315 conversion is given
316 [16335] manual: Feature test macro documentation incomplete and out of
317 date
318 [16552] libc: Unify umount implementations in terms of umount2
319 [17082] libc: htons et al.: statement-expressions prevent use on global
320 scope with -O1 and higher
321 [17343] libc: Signed integer overflow in /stdlib/random_r.c
322 [17438] localedata: pt_BR: wrong d_fmt delimiter
323 [17662] libc: please implement binding for the new renameat2 syscall
324 [17721] libc: __restrict defined as /* Ignore */ even in c11
325 [17979] libc: inconsistency between uchar.h and stdint.h
326 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
327 [18023] libc: extend_alloca is broken (questionable pointer comparison,
328 horrible machine code)
329 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
330 last constant.
331 [18471] libc: llseek should be a compat symbol
332 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
333 symbols
334 [18991] nss: nss_files skips large entry in database
335 [19239] libc: Including stdlib.h ends up with macros major and minor being
336 defined
337 [19463] libc: linknamespace failures when compiled with -Os
338 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
339 [19527] locale: Normalized charset name not recognized by setlocale
340 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
341 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
342 fpscr.c'
343 [19728] network: out of bounds stack read in libidn function
344 idna_to_ascii_4i (CVE-2016-6261)
345 [19729] network: out of bounds heap read on invalid utf-8 inputs in
346 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
347 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
348 the base address
349 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
350 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
351 [20419] dynamic-link: files with large allocated notes crash in
352 open_verify
353 [20530] libc: bswap_16 should use __builtin_bswap16() when available
354 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
355 [20980] manual: CFLAGS environment variable replaces vital options
356 [21163] regex: Assertion failure in pop_fail_stack when executing a
357 malformed regexp (CVE-2015-8985)
358 [21234] manual: use of CFLAGS makes glibc detect no optimization
359 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
360 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
361 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
362 [21508] locale: intl/tst-gettext failure with latest msgfmt
363 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
364 [21812] network: getifaddrs() returns entries with ifa_name == NULL
365 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
366 dlopen
367 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
368 AT_SECURE=1
369 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
370 [22247] network: Integer overflow in the decode_digit function in
371 puny_decode.c in libidn (CVE-2017-14062)
372 [22342] nscd: NSCD not properly caching netgroup
373 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
374 [22550] localedata: es_ES locale (and other es_* locales): collation
375 should treat ñ as a primary different character, sync the collation
376 for Spanish with CLDR
377 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
378 by gcc configured with --enable-default-pie
379 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
380 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
381 crossing 2GB threshold (CVE-2017-18269)
382 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
383 es_BO
384 [22735] time: Misleading typo in time.h source comment regarding
385 CLOCKS_PER_SECOND
386 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
387 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
388 `__assert_perror_fail ()' to `__assert_fail_base ()'
389 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
390 dlopen failures
391 [22786] libc: Stack buffer overflow in realpath() if input size is close
392 to SSIZE_MAX (CVE-2018-11236)
393 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
394 through an absolute DT_NEEDED path
395 [22792] build: tcb-offsets.h dependency dropped
396 [22797] libc: pkey_get() uses non-reserved name of argument
397 [22807] libc: PTRACE_* constants missing for powerpc
398 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
399 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
400 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
401 [22848] localedata: ca_ES: update date definitions from CLDR
402 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
403 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
404 [22896] localedata: Update locale data for an_ES
405 [22902] math: float128 test failures with GCC 8
406 [22918] libc: multiple common of `__nss_shadow_database'
407 [22919] libc: sparc32: backtrace yields infinite backtrace with
408 makecontext
409 [22926] libc: FTBFS on powerpcspe
410 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
411 required
412 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
413 [22947] libc: FAIL: misc/tst-preadvwritev2
414 [22963] localedata: cs_CZ: Add alternative month names
415 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
416 [22996] localedata: change LC_PAPER to en_US in es_BO locale
417 [22998] dynamic-link: execstack tests are disabled when SELinux is
418 disabled
419 [23005] network: Crash in __res_context_send after memory allocation
420 failure
421 [23007] math: strtod cannot handle -nan
422 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
423 [23036] regex: regex equivalence class regression
424 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
425 [23069] libc: sigaction broken on riscv64-linux-gnu
426 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
427 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
428 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
429 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
430 and libc build with -Os)
431 [23140] localedata: More languages need two forms of month names
432 [23145] libc: _init/_fini aren't marked as hidden
433 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
434 [23171] math: C++ iseqsig for long double converts arguments to double
435 [23178] nscd: sudo will fail when it is run in concurrent with commands
436 that changes /etc/passwd
437 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
438 (CVE-2018-11237)
439 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
440 [23208] localedata: New locale - Lower Sorbian (dsb)
441 [23233] regex: Memory leak in build_charclass_op function in file
442 posix/regcomp.c
443 [23236] stdio: Harden function pointers in _IO_str_fields
444 [23250] nptl: Offset of __private_ss differs from GCC
445 [23253] math: tgamma test suite failures on i686 with -march=x86-64
446 -mtune=generic -mfpmath=sse
447 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
448 AT_SECURE
449 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
450 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
451 parser.c
452 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
453 [23277] math: nan function should not have const attribute
454 [23279] math: scanf and strtod wrong for some hex floating-point
455 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
456 directed rounding
457 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
458 [23303] build: undefined reference to symbol
459 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
460 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
461 lookup
462 [23313] stdio: libio vtables validation and standard file object
463 interposition
464 [23329] libc: The __libc_freeres infrastructure is not properly run across
465 DSO boundaries.
466 [23349] libc: Various glibc headers no longer compatible with
467 <linux/time.h>
468 [23351] malloc: Remove unused code related to heap dumps and malloc
469 checking
470 [23363] stdio: stdio-common/tst-printf.c has non-free license
471 [23396] regex: Regex equivalence regression in single-byte locales
472 [23422] localedata: oc_FR: More updates of locale data
473 [23442] build: New warning with GCC 8
474 [23448] libc: Out of bounds access in IBM-1390 converter
475 [23456] libc: Wrong index_cpu_LZCNT
476 [23458] build: tst-get-cpu-features-static isn't added to tests
477 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
478 processors
479 [23467] dynamic-link: x86/CET: A property note parser bug
480
481 \f
482 Version 2.27
483
484 Major new features:
485
486 * The GNU C Library can now be compiled with support for building static
487 PIE executables (See --enable-static-pie in INSTALL). These static PIE
488 executables are like static executables but can be loaded at any address
489 and provide additional security hardening benefits at the cost of some
490 memory and performance. When the library is built with --enable-static-pie
491 the resulting libc.a is usable with GCC 8 and above to create static PIE
492 executables using the GCC option '-static-pie'. This feature is currently
493 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
494 aarch64 with binutils 2.30 or later.
495
496 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
497 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
498 H.J. Lu from Intel.
499
500 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
501
502 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
503
504 * In order to support faster and safer process termination the malloc API
505 family of functions will no longer print a failure address and stack
506 backtrace after detecting heap corruption. The goal is to minimize the
507 amount of work done after corruption is detected and to avoid potential
508 security issues in continued process execution. Reducing shutdown time
509 leads to lower overall process restart latency, so there is benefit both
510 from a security and performance perspective.
511
512 * The abort function terminates the process immediately, without flushing
513 stdio streams. Previous glibc versions used to flush streams, resulting
514 in deadlocks and further data corruption. This change also affects
515 process aborts as the result of assertion failures.
516
517 * On platforms where long double has the IEEE binary128 format (aarch64,
518 alpha, mips64, riscv, s390 and sparc), the math library now implements
519 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
520 These are the same interfaces added in version 2.26 for some platforms where
521 this format is supported but is not the format of long double.
522
523 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
524 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
525 implements interfaces for that type, as defined by ISO/IEC TS
526 18661-3:2015. These are corresponding interfaces to those supported for
527 _Float128.
528
529 * The math library now implements interfaces for the _Float32, _Float64 and
530 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
531 corresponding interfaces to those supported for _Float128.
532
533 * glibc now implements the memfd_create and mlock2 functions on Linux.
534
535 * Support for memory protection keys was added. The <sys/mman.h> header now
536 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
537 pkey_get.
538
539 * The copy_file_range function was added.
540
541 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
542
543 * The ldconfig utility now processes `include' directives using the C/POSIX
544 collation ordering. Previous glibc versions used locale-specific
545 ordering, the change might break systems that relied on that.
546
547 * Support for two grammatical forms of month names has been added.
548 In a call to strftime, the "%B" and "%b" format specifiers will now
549 produce the grammatical form required when the month is used as part
550 of a complete date. New "%OB" and "%Ob" specifiers produce the form
551 required when the month is named by itself. For instance, in Greek
552 and in many Slavic and Baltic languages, "%B" will produce the month
553 in genitive case, and "%OB" will produce the month in nominative case.
554
555 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
556 are all valid and will all accept any known form of month
557 name---standalone or complete, abbreviated or full. In a call to
558 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
559 the strings used by "%B" and "%b", respectively. New query
560 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
561 used by "%OB" and "%Ob", respectively.
562
563 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
564 define the strings for %OB and %Ob, respectively; these have the
565 same syntax as "mon" and "abmon". These arrays are optional; if they
566 are not provided then they have the same content as "mon" and "abmon",
567 respectively.
568
569 These features are provided for locales which define "alt_mon" and/or
570 "ab_alt_mon" in their locale source data. This release includes such
571 alternative month name data for the following languages: Belarusian,
572 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
573
574 This feature is currently a GNU extension, but it is expected to
575 be added to the next revision of POSIX, and it is also already
576 available on some BSD-derived operating systems.
577
578 This feature will cause existing statically compiled applications
579 to fail to load locales and fall back to the builtin C/POSIX locales.
580 See notes below for other changes affecting compatibility.
581
582 * Support for the RISC-V ISA running on Linux has been added. This port
583 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
584 for the following ISA and ABI pairs:
585
586 - rv64imac lp64
587 - rv64imafdc lp64
588 - rv64imafdc lp64d
589
590 Deprecated and removed features, and other changes affecting compatibility:
591
592 * Statically compiled applications attempting to load locales compiled for the
593 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
594 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
595 support for two grammatical forms of the month names, also extends the locale
596 data binary format. Static applications needing locale support must be
597 recompiled to match the runtime and data they are deployed with. In some
598 distributions there is an upgrade window where dynamically linked applications
599 may use a new library but the old locale data and also fall back to the
600 builtin C/POSIX locales; restarting the application process is sufficient to
601 fix this.
602
603 * Support for statically linked applications which call dlopen is deprecated
604 and will be removed in a future version of glibc. Applications which call
605 dlopen need to be linked dynamically instead.
606
607 * Support for old programs which use internal stdio data structures and
608 functions is deprecated. This includes programs which use the C++ streams
609 provided by libstdc++ in GCC 2.95. Programs which use the internal
610 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
611 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
612 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
613 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
614 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
615 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
616 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
617 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
618 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
619 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
620 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
621 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
622 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
623 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
624 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
625 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
626 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
627 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
628 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
629 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
630 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
631 future version of glibc. Unlike other symbol removals, these old
632 applications will not be supported using compatibility symbols.
633
634 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
635 defined by <sys/ptrace.h>.
636
637 * libm no longer supports SVID error handling (calling a user-provided
638 matherr function on error) or the _LIB_VERSION variable to control error
639 handling. (SVID error handling and the _LIB_VERSION variable still work
640 for binaries linked against older versions of the GNU C Library.) The
641 libieee.a library is no longer provided. math.h no longer defines struct
642 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
643 TLOSS, PLOSS and HUGE.
644
645 * The libm functions pow10, pow10f and pow10l are no longer supported for
646 new programs. Programs should use the standard names exp10, exp10f and
647 exp10l for these functions instead.
648
649 * The mcontext_t type is no longer the same as struct sigcontext. On
650 platforms where it was previously the same, this changes the C++ name
651 mangling for interfaces involving this type.
652
653 * The add-ons mechanism for building additional packages at the same time as
654 glibc has been removed. The --enable-add-ons configure option is now
655 ignored.
656
657 * The --without-fp configure option is now ignored. Whether hardware
658 floating-point instructions are used is now configured based on whether
659 the compiler used at configure time (without any options implied by a
660 --with-cpu= configure option) uses such instructions.
661
662 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
663 the specified string can be parsed as a domain name.
664
665 * In the malloc_info output, the <heap> element may contain another <aspace>
666 element, "subheaps", which contains the number of sub-heaps.
667
668 * The libresolv function p_secstodate is no longer supported for new
669 programs.
670
671 * The tilepro-*-linux-gnu configuration is no longer supported.
672
673 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
674 and will be removed in a future release. Software that is still using
675 either header should be updated to use standard <stdio.h> interfaces
676 instead.
677
678 libio.h was originally the header for a set of supported GNU extensions,
679 but they have not been maintained as such in many years, they are now
680 standing in the way of improvements to stdio, and we don't think there are
681 any remaining external users. _G_config.h was never intended for public
682 use, but predates the bits convention.
683
684 Changes to build and runtime requirements:
685
686 * bison version 2.7 or later is required to generate code in the 'intl'
687 subdirectory.
688
689 Security related changes:
690
691 CVE-2009-5064: The ldd script would sometimes run the program under
692 examination directly, without preventing code execution through the
693 dynamic linker. (The glibc project disputes that this is a security
694 vulnerability; only trusted binaries must be examined using the ldd
695 script.)
696
697 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
698 suffered from a one-byte overflow during ~ operator processing (either
699 on the stack or the heap, depending on the length of the user name).
700 Reported by Tim Rühsen.
701
702 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
703 would sometimes fail to free memory allocated during ~ operator
704 processing, leading to a memory leak and, potentially, to a denial
705 of service.
706
707 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
708 without GLOB_NOESCAPE, could write past the end of a buffer while
709 unescaping user names. Reported by Tim Rühsen.
710
711 CVE-2017-17426: The malloc function, when called with an object size near
712 the value SIZE_MAX, would return a pointer to a buffer which is too small,
713 instead of NULL. This was a regression introduced with the new malloc
714 thread cache in glibc 2.26. Reported by Iain Buclaw.
715
716 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
717 to the allocation of too much memory. (This is not a security bug per se,
718 it is mentioned here only because of the CVE assignment.) Reported by
719 Qualys.
720
721 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
722 of the number of search path components. (This is not a security
723 vulnerability per se because no trust boundary is crossed if the fix for
724 CVE-2017-1000366 has been applied, but it is mentioned here only because
725 of the CVE assignment.) Reported by Qualys.
726
727 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
728 for AT_SECURE or SUID binaries could be used to load libraries from the
729 current directory.
730
731 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
732 succeeds without returning an absolute path due to unexpected behaviour
733 of the Linux kernel getcwd syscall. Reported by halfdog.
734
735 CVE-2018-6485: The posix_memalign and memalign functions, when called with
736 an object size near the value of SIZE_MAX, would return a pointer to a
737 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
738
739 CVE-2018-6551: The malloc function, when called with an object size near
740 the value of SIZE_MAX, would return a pointer to a buffer which is too
741 small, instead of NULL.
742
743 The following bugs are resolved with this release:
744
745 [866] glob: glob should match dangling symlinks
746 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
747 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
748 locale
749 [5997] math: Very slow execution of sinf function
750 [10580] localedata: hr_HR: updated locale
751 [10871] locale: 'mon' array should contain both nominative and genitive
752 cases
753 [12349] localedata: eu_ES: incorrect thousands separator
754 [13605] localedata: shn_MM: new Shan locale
755 [13805] localedata: ru_RU: currency should use ',' as radix point
756 [13953] localedata: km_KH: locale update
757 [13994] localedata: mjw_IN: new locale
758 [14121] build: make writes .mo files in po directory
759 [14333] libc: Fix the race between atexit() and exit()
760 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
761 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
762 "Bangla"
763 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
764 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
765 width Latin characters
766 [15332] localedata: es_CU: locale update
767 [15436] stdio: Don't close or flush stdio streams on abort
768 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
769 letters
770 [16148] localedata: ca_ES: incorrect thousands separator
771 [16750] dynamic-link: ldd should not try to execute the binaries
772 (CVE-2009-5064)
773 [16777] localedata: pl_PL: incorrect thousands separator in locale
774 [16905] localedata: hanzi: new collation
775 [17563] localedata: cmn_TW: add hanzi collation
776 [17750] localedata: wrong collation order of diacritics in most locales
777 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
778 if malloc succeeded
779 [17956] build: Build fails on missing definitions from header file
780 nss/nss.h when Mozilla NSS is used for cryptography
781 [18203] libc: realpath() does not handle unreachable paths correctly
782 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
783 [18812] localedata: kab_DZ: new Kabyle Algeria locale
784 [18822] libc: Internal functions are called via PLT
785 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
786 [19170] libc: __gmon_start__ defined in hppa in crtn.S
787 [19574] libc: glibc should support building static PIE binaries
788 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
789 U+309A
790 [19971] glob: glob: Do not skip entries with zero d_ino values
791 [19982] localedata: fr.po: spelling mistake for error code EXDEV
792 [20008] localedata: km_KH: convert to translit_neutral
793 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
794 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
795 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
796 [20482] localedata: de_CH: abbreviated weekdays should be two letters
797 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
798 for Nicaragua
799 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
800 they're set, wrong errors returned by gaih_inet when lookup functions
801 are not found.
802 [20756] localedata: [PATCH] Use Unicode wise thousands separator
803 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
804 access
805 [20952] localedata: yuw_PG: new locale
806 [21084] localedata: charmaps/IBM858: new codepage
807 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
808 [21242] libc: assert gives pedantic warning in old gcc versions
809 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
810 __regcall calling convention
811 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
812 [21326] libc: C99 functions are not declared for C++11 and later if
813 _GNU_SOURCE is not predefined by g++
814 [21457] libc: sys/ucontext.h namespace
815 [21530] libc: tmpfile() should be implemented using O_TMPFILE
816 [21660] math: GCC fails to compile a formula with tgmath.h
817 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
818 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
819 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
820 [21684] math: tgmath.h handling of complex integers
821 [21685] math: tgmath.h handling of bit-fields
822 [21686] math: tgmath.h handling of __int128
823 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
824 locale
825 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
826 protector=all
827 [21750] localedata: column width of characters incompatible with classical
828 wcwidth
829 [21754] malloc: malloc: Perform as little work as possible after heap
830 consistency check failures
831 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
832 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
833 [21791] string: Unused XXX_chk_XXX functions in libc.a
834 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
835 PIE
836 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
837 locales
838 [21845] localedata: Added new Locale bho_NP
839 [21853] localedata: Fix abday Which looks same as day in zh_SG
840 [21854] localedata: Added New Locale en_SC
841 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
842 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
843 _dl_runtime_resolve_avx_slow
844 [21885] network: getaddrinfo: gethosts does not release resolver context
845 on memory allocation failure
846 [21899] libc: XPG4.2 sigaction namespace
847 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
848 the suspect)
849 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
850 default (i386)
851 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
852 errno value, causing getaddrinfo to fail
853 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
854 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
855 instead of EAI_NODATA
856 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
857 constant PTRACE_SEIZE_DEVEL
858 [21930] math: C-only gcc builtins used in <math.h> isinf
859 [21932] network: Unpaired __resolv_context_get in generic get*_r
860 implementation
861 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
862 [21944] libc: sigval namespace
863 [21951] localedata: Update hanzi collation by stroke
864 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
865 sysdeps/x86_64/fpu/e_expf.S
866 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
867 allocation in syscall loops)
868 [21959] localedata: Fix Country name for xh_ZA
869 [21960] localedata: Fix abmon for bem_ZM
870 [21966] math: AVX2 mathvec functions use FMA without checking
871 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
872 [21971] localedata: Added New Locale for mfe_MU
873 [21972] libc: assert macro requires operator== (int) for its argument type
874 [21973] math: [sparc] libm missing sqrtl compat symbol
875 [21974] libc: Remove __bb_init_func and __bb_exit_func
876 [21982] string: stratcliff.c: error: assuming signed overflow does not
877 occur with -O3
878 [21986] stdio: __guess_grouping is called incorrectly
879 [21987] math: [sparc32] wrong bits/long-double.h installed
880 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
881 amounts)
882 [22022] localedata: Missing country_name for mni_IN
883 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
884 niu_NZ
885 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
886 checks
887 [22026] locale: iconv_open: heap overflow on gconv_init failure
888 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
889 [22035] math: [m68k] bits/math-inline.h macro namespace
890 [22038] localedata: Fix abbreviated weeks and months for Somali
891 [22044] localedata: Remove redundant data for Limburgish Language
892 [22050] malloc: Linking with -lmcheck does not hook
893 __malloc_initialize_hook correctly
894 [22051] libc: zero terminator in the middle of glibc's .eh_frame
895 [22052] malloc: malloc failed to compile with GCC 7 and -O3
896 [22070] localedata: charmaps/UTF-8: wcwidth for
897 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
898 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
899 Jungseong and Jongseong) should be 0
900 [22078] nss: nss_files performance issue in multi mode
901 [22082] math: bits/math-finite.h exp10 condition
902 [22086] libc: pcprofiledump incorrect cross-endian condition
903 [22093] dynamic-link: ld.so no longer searches in .../x86_64
904 [22095] network: Name server address allocation memory leak in resolv.conf
905 parsing after OOM
906 [22096] network: __resolv_conf_attach can incorrectly free passed conf
907 object
908 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
909 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
910 e.g. ET_GNU_DEBUG_*
911 [22111] malloc: malloc: per thread cache is not returned when thread exits
912 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
913 [22134] libc: [linux] implement fexecve with execveat
914 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
915 ppc64le
916 [22145] libc: ttyname() gives up too early in the face of namespaces
917 [22146] math: C++ build issue with float128 on x86_64
918 [22153] nptl: nptl: save error code before process termination
919 [22156] libc: [hppa,ia64,microblaze] Executable stack default
920 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
921 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
922 [22165] libc: [hppa] Text relocations in libc.so
923 [22180] libc: destructor registered via __cxa_atexit is called twice
924 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
925 symlinks") cause "make" segfaults
926 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
927 math_force_eval
928 [22207] libc: FAIL: stdlib/test-atexit-race
929 [22225] math: nearbyint arithmetic moved before feholdexcept
930 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
931 [22235] math: iscanonical in C++ and float128
932 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
933 the svid compat wrapper
934 [22244] math: ynf and yn are wrong without the svid compat wrapper
935 [22273] libc: Improper assert in Linux posix_spawn implementation
936 [22284] libc: -pg -pie doesn't work
937 [22292] locale: localedef exits with error 4 when it should be error 1
938 [22294] locale: Allow "" for int_currency_symbol definition in locales.
939 [22295] locale: Don't warn on non-symbolic characters in locale sources in
940 --verbose.
941 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
942 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
943 to 2.26
944 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
945 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
946 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
947 [22322] libc: [mips64] wrong bits/long-double.h installed
948 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
949 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
950 (CVE-2017-15804)
951 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
952 [22343] malloc: Integer overflow in posix_memalign
953 [22347] libc: getrandom() returns the number of bytes that were copied to
954 the buffer even though the comments say "Return 0 on success and -1 on
955 failure."
956 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
957 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
958 -m32
959 [22370] dynamic-link: Incorrect note padding check
960 [22375] libc: malloc returns pointer from tcache_get when should return
961 NULL (CVE-2017-17426)
962 [22377] math: iseqsig, float128 and C++
963 [22382] localedata: Error in tpi_PG locale
964 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
965 inside the ASCII printable range
966 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
967 -mlong-double-64
968 [22403] localedata: Slash needs escaping in some locales
969 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
970 [22409] network: res_hnok does not accept some host names used on the
971 Internet
972 [22412] network: res_dnok, res_hnok should perform syntax checks
973 [22413] network: ns_name_pton ignores syntactically invalid trailing
974 backslash
975 [22415] stdio: setvbuf can lead to invalid free/segfault
976 [22432] build: Non-deterministic build
977 [22439] malloc: malloc_info should compute summary statistics for all sub-
978 heaps in an arena
979 [22442] network: if_nametoindex could report index for the wrong
980 networking interface
981 [22446] build: aliasing violation calling readlink in handle_request
982 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
983 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
984 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
985 __nscd_hash/__nss_hash
986 [22463] network: p_secstodate overflow handling
987 [22469] localedata: pl_PL LC_COLLATE does not use i18n
988 [22478] libc: sigwait can fail with EINTR
989 [22505] libc: ldconfig processes include directive in locale-specific
990 order
991 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
992 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
993 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
994 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
995 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
996 [22534] localedata: Collation rules for Serbian and Bosnian should be the
997 same as for Croatian
998 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
999 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1000 [22577] libc: missing newline after "cannot allocate TLS data structures
1001 for initial thread"
1002 [22588] manual: manual/conf.texi: missing underscore in front of
1003 SC_SSIZE_MAX
1004 [22593] math: nextafter and nexttoward are declared with const attribute
1005 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1006 [22603] string: ia64 memchr overflows internal pointer check
1007 [22605] libc: SH clone does not set the exit code correctly
1008 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1009 (CVE-2017-1000408)
1010 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1011 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1012 modified in case of success
1013 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1014 [22615] manual: manual: ambiguous wording about errno value in case of
1015 success
1016 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1017 stack-protector=all
1018 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1019 binaries or if /proc is not mounted (CVE-2017-16997)
1020 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1021 [22630] build: $(no-pie-ldflag) is no longer effective
1022 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1023 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1024 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1025 [22637] nptl: guard size is subtracted from thread stack size instead of
1026 adding it on top
1027 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1028 [22657] localedata: hu_HU: Avoid double space in date
1029 [22660] math: fmax, fmin sNaN handling on alpha
1030 [22664] libc: New warning of GCC8
1031 [22665] math: alpha: ceil and floor raise inexact exceptions
1032 [22666] math: alpha: trunc raise inexact exceptions
1033 [22667] libc: makecontext lacks stack alignment on i386
1034 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1035 [22679] libc: getcwd(3) can succeed without returning an absolute path
1036 (CVE-2018-1000001)
1037 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1038 elision and tunables
1039 [22687] math: [powerpc-nofpu] complex long double functions spurious
1040 "invalid" exception
1041 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1042 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1043 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1044 exception
1045 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1046 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1047 power4
1048 [22701] nis: Incomplete removal of libnsl
1049 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1050 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1051 [22715] dynamic-link: FAIL: elf/tst-audit10
1052 [22719] libc: Backtrace tests fail on hppa
1053 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1054 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1055 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1056 before the first call to crypt_r () in crypt/badsalttest.c
1057
1058 \f
1059 Version 2.26
1060
1061 Major new features:
1062
1063 * A per-thread cache has been added to malloc. Access to the cache requires
1064 no locks and therefore significantly accelerates the fast path to allocate
1065 and free small amounts of memory. Refilling an empty cache requires locking
1066 the underlying arena. Performance measurements show significant gains in a
1067 wide variety of user workloads. Workloads were captured using a special
1068 instrumented malloc and analyzed with a malloc simulator. Contributed by
1069 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1070
1071 * Unicode 10.0.0 Support: Character encoding, character type info, and
1072 transliteration tables are all updated to Unicode 10.0.0, using
1073 generator scripts contributed by Mike FABIAN (Red Hat).
1074 These updates cause user visible changes, especially the changes in
1075 wcwidth for many emoji characters cause problems when emoji sequences
1076 are rendered with pango, see for example:
1077 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1078
1079 * Collation of Hungarian has been overhauled and is now consistent with "The
1080 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1081 Egmont Koblinger.
1082
1083 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1084
1085 - The GNU C Library will now detect when /etc/resolv.conf has been
1086 modified and reload the changed configuration. The new resolver option
1087 “no-reload” (RES_NORELOAD) disables this behavior.
1088
1089 - The GNU C Library now supports an arbitrary number of search domains
1090 (configured using the “search” directive in /etc/resolv.conf);
1091 previously, there was a hard limit of six domains. For backward
1092 compatibility, applications that directly modify the ‘_res’ global
1093 object are still limited to six search domains.
1094
1095 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1096 Library will now randomly pick a name server from the configuration as a
1097 starting point. (Previously, the second name server was always used.)
1098
1099 * The tunables feature is now enabled by default. This allows users to tweak
1100 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1101
1102 * New function reallocarray, which resizes an allocated block (like realloc)
1103 to the product of two sizes, with a guaranteed clean failure upon integer
1104 overflow in the multiplication. Originally from OpenBSD, contributed by
1105 Dennis Wölfing and Rüdiger Sonderfeld.
1106
1107 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1108 These are extended versions of preadv and pwritev, respectively, taking an
1109 additional flags argument. The set of supported flags depends on the
1110 running kernel; full support currently requires kernel 4.7 or later.
1111
1112 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1113 create a new session ID for the spawned process. This feature is
1114 scheduled to be added to the next major revision of POSIX; for the time
1115 being, it is available under _GNU_SOURCE.
1116
1117 * errno.h is now safe to use from C-preprocessed assembly language on all
1118 supported operating systems. In this context, it will only define the
1119 Exxxx constants, as preprocessor macros expanding to integer literals.
1120
1121 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1122 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1123 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1124 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1125
1126 To compile programs that use this feature, the compiler must support
1127 128-bit floating point with the type name _Float128 (as defined by TS
1128 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1129 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1130 must be defined to make the new interfaces visible.
1131
1132 The new functions and macros correspond to those present for other
1133 floating-point types (except for a few obsolescent interfaces not
1134 supported for the new type), with F128 or f128 suffixes; for example,
1135 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1136 printf or scanf formats for the new type; the strfromf128 and strtof128
1137 interfaces should be used instead.
1138
1139 Deprecated and removed features, and other changes affecting compatibility:
1140
1141 * The synchronization that pthread_spin_unlock performs has been changed to
1142 now be equivalent to a C11 atomic store with release memory order to the
1143 spin lock's memory location. Previously, several (but not all)
1144 architectures used stronger synchronization (e.g., containing what is
1145 often called a full barrier). This change can improve performance, but
1146 may affect odd fringe uses of spin locks that depend on the previous
1147 behavior (e.g., using spin locks as atomic variables to try to implement
1148 Dekker's mutual exclusion algorithm).
1149
1150 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1151 removed.
1152
1153 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1154 will only be built and installed when the GNU C Library is configured with
1155 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1156 as TIRPC or rpcsvc-proto, to be used.
1157
1158 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1159 libnss_compat, are deprecated, and will not be built or installed by
1160 default.
1161
1162 The NIS(+) support library, libnsl, is also deprecated. By default, a
1163 compatibility shared library will be built and installed, but not headers
1164 or development libraries. Only a few NIS-related programs require this
1165 library. (In particular, the GNU C Library has never required programs
1166 that use 'gethostbyname' to be linked with libnsl.)
1167
1168 Replacement implementations based on TIRPC, which additionally support
1169 IPv6, are available from <https://github.com/thkukuk/>. The configure
1170 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1171 name service modules, to be built and installed.
1172
1173 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1174 support is enabled, the configured recursive resolver must support EDNS.
1175 (Responding to EDNS-enabled queries with responses which are not
1176 EDNS-enabled is fine, but FORMERR responses are not.)
1177
1178 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1179 servers have not supported this opcode for a long time.
1180
1181 * The _res_opcodes variable has been removed from libresolv. It had been
1182 exported by accident.
1183
1184 * <string.h> no longer includes inline versions of any string functions,
1185 as this kind of optimization is better done by the compiler. The macros
1186 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1187
1188 * The nonstandard header <xlocale.h> has been removed. Most programs should
1189 use <locale.h> instead. If you have a specific need for the definition of
1190 locale_t with no other declarations, please contact
1191 libc-alpha@sourceware.org and explain.
1192
1193 * The obsolete header <sys/ultrasound.h> has been removed.
1194
1195 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1196
1197 * The obsolete function cfree has been removed. Applications should use
1198 free instead.
1199
1200 * The stack_t type no longer has the name struct sigaltstack. This changes
1201 the C++ name mangling for interfaces involving this type.
1202
1203 * The ucontext_t type no longer has the name struct ucontext. This changes
1204 the C++ name mangling for interfaces involving this type.
1205
1206 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1207 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1208 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1209 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1210 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1211 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1212 fpregset_t type no longer has the name struct fpu. This changes the C++
1213 name mangling for interfaces involving those types.
1214
1215 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1216 synced with the kernel:
1217
1218 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1219 are not supported on this architecture and have been removed.
1220
1221 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1222 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1223 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1224
1225 Programs that assume the GET/SETREGS ptrace requests are universally
1226 available will now fail to build, instead of malfunctioning at runtime.
1227
1228 Changes to build and runtime requirements:
1229
1230 * Linux kernel 3.2 or later is required at runtime, on all architectures
1231 supported by that kernel. (This is a change from version 2.25 only for
1232 x86-32 and x86-64.)
1233
1234 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1235
1236 * On most architectures, GCC 4.9 or later is required to build the GNU C
1237 Library. On powerpc64le, GCC 6.2 or later is required.
1238
1239 Older GCC versions and non-GNU compilers are still supported when
1240 compiling programs that use the GNU C Library. (We do not know exactly
1241 how old, and some GNU extensions to C may be _de facto_ required. If you
1242 are interested in helping us make this statement less vague, please
1243 contact libc-alpha@sourceware.org.)
1244
1245 Security related changes:
1246
1247 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1248 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1249
1250 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1251 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1252
1253 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1254 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1255 (CVE-2010-3192).
1256
1257 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1258 fixed (CVE-2017-12133).
1259
1260 The following bugs are resolved with this release:
1261
1262 [984] network: Respond to changed resolv.conf in gethostbyname
1263 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1264 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1265 fields
1266 [12189] libc: __stack_chk_fail should not attempt a backtrace
1267 (CVE-2010-3192)
1268 [14096] time: Race condition on timezone/tst-timezone.out
1269 [14172] localedata: az_IR: new locale
1270 [14995] build: glibc fails to build if gold is the default linker, even if
1271 ld.bfd is available
1272 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1273 [16637] network: inet_pton function is accepting IPv6 with bad format
1274 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1275 x86 and other generic code
1276 [16875] localedata: ko_KR: fix lang_name
1277 [17225] localedata: ar_SY: localized month names for May and June are
1278 incorrect
1279 [17297] localedata: da_DK: wrong date_fmt string
1280 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1281 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1282 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1283 [19066] localedata: ar_SA abbreviated day and month names are in English
1284 [19569] network: resolv: Support an arbitrary number of search domains
1285 [19570] network: Implement random DNS server selection in the stub
1286 resolver
1287 [19838] locale: localedef fails on PA-RISC
1288 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1289 order of 0D36 and 0D37
1290 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1291 chillu characters
1292 [20098] libc: FAIL: debug/backtrace-tst on hppa
1293 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1294 receiving data
1295 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1296 whitespace
1297 [20313] localedata: Update locale data to Unicode 9.0
1298 [20424] manual: Document how to provide a malloc replacement
1299 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1300 for Peru
1301 [20686] locale: Add el_GR@euro to SUPPORTED.
1302 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1303 failures consistently
1304 [21015] dynamic-link: Document and fix --enable-bind-now
1305 [21016] nptl: pthread_cond support is broken on hppa
1306 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1307 frame-pointer on i386
1308 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1309 register
1310 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1311 [21088] libc: Build fails with --enable-static-nss
1312 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1313 generic c code is used
1314 [21109] libc: Tunables broken on big-endian
1315 [21112] math: powf has large ulp errors with base close to 1 and exponent
1316 around 4000
1317 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1318 (CVE-2017-12133)
1319 [21120] malloc: glibc malloc is incompatible with GCC 7
1320 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1321 with -lieee
1322 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1323 (-0.0) when linking with -lieee
1324 [21171] math: log10, log2 and lgamma return incorrect results
1325 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1326 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1327 [21207] localedata: ce_RU: update weekdays from CLDR
1328 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1329 [21217] localedata: Update months from CLDR-31
1330 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1331 [21243] libc: support_delete_temp_file should issue warning for failed
1332 remove()
1333 [21244] libc: support resolv_test_start() socket fd close should be
1334 checked for errors.
1335 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1336 new posix_spawn implementation
1337 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1338 leads to lower CPU frequency
1339 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1340 [21261] libc: [sparc64] bits/setjmp.h namespace
1341 [21267] network: [mips] bits/socket.h IOC* namespace
1342 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1343 [21270] libc: mmap64 silently truncates large offset values
1344 [21275] libc: posix_spawn always crashes on ia64 now
1345 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1346 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1347 [21289] libc: Incorrect declaration for 32-bit platforms with
1348 _FILE_OFFSET_BITS=64 causes build error
1349 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1350 not support gethostbyname4_r
1351 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1352 switching
1353 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1354 arenas
1355 [21340] libc: Support POSIX_SPAWN_SETSID
1356 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1357 [21359] network: ns_name_pack needs additional byte in destination buffer
1358 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1359 against fragmentation attacks (CVE-2017-12132)
1360 [21369] network: resolv: Remove EDNS fallback
1361 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1362 and _POSIX_C_SOURCE
1363 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1364 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1365 [21393] stdio: Missing dup3 error check in freopen, freopen64
1366 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1367 [21399] localedata: Bad description for U00EC in
1368 localedata/charmaps/CP1254
1369 [21411] malloc: realloc documentation error
1370 [21426] network: sys/socket.h uio.h namespace
1371 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1372 [21445] libc: signal.h bsd_signal namespace
1373 [21455] network: Network headers stdint.h namespace
1374 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1375 retry value
1376 [21475] network: resolv: Overlong search path is truncated mid-label
1377 [21511] libc: sigstack namespace
1378 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1379 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1380 bits/syscall.h: No such file or directory
1381 [21517] libc: struct sigaltstack namespace
1382 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1383 [21533] localedata: Update locale data to Unicode 10.0
1384 [21537] libc:
1385 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1386 junk at end of line, first unrecognized character is `@'
1387 [21538] libc: SIG_HOLD missing for XPG4
1388 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1389 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1390 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1391 [21543] libc: sigevent namespace
1392 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1393 size for general purpose registers in mcontext_t structure
1394 [21550] libc: sigwait namespace
1395 [21552] libc: XPG4 bsd_signal namespace
1396 [21554] libc: sigpause namespace
1397 [21560] libc: sys/wait.h signal.h namespace
1398 [21561] libc: waitid namespace
1399 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1400 '/usr/include/stdlib.h': Permission denied
1401 [21575] libc: sys/wait.h missing struct rusage definition
1402 [21584] libc: sigaltstack etc namespace
1403 [21597] libc: siginterrupt namespace
1404 [21607] math: hppa: FAIL: math/test-tgmath
1405 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1406 issue on x86-64
1407 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1408 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1409 stack and heap (CVE-2017-1000366)
1410 [21625] libc: wait3 namespace
1411 [21654] nss: Incorrect pointer alignment in NSS group merge result
1412 construction
1413 [21657] network: Parse interface zone id for node-local multicast
1414 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1415 between_2_3
1416 [21666] libc: .symver is used on common symbol
1417 [21668] network: resolv: res_init cross-thread broadcast introduces race
1418 conditions
1419 [21687] math: tgmath.h totalorder, totalordermag return type
1420 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1421 Hindi Locale
1422 [21696] libc: Incorrect assumption of of __cpu_mask in
1423 posix/sched_cpucount.c
1424 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1425 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1426 locale
1427 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1428 leading to relocation crash
1429 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1430 `resolv_conf_matches (resp, conf)' failed.
1431 [21710] localedata: Added Samoan language locale for Samoa
1432 [21711] localedata: Pashto yesstr/nostr locale are missing
1433 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1434 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1435 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1436 [21723] localedata: yesstr/nostr missing for Chinese language locale
1437 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1438 locale
1439 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1440 locale
1441 [21728] localedata: New Locale for Tongan language
1442 [21729] localedata: incorrect LC_NAME fields for hi_IN
1443 [21733] localedata: yesstr and nostr are missing for zh_HK
1444 [21734] localedata: Missing yesstr and nostr are for kw_GB
1445 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1446 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1447 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1448 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1449 strings
1450 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1451 protector=all
1452 [21749] localedata: Wrong abbreviated day name (“abday”) for
1453 ar_JO/ar_LB/ar_SY
1454 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1455 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1456 [21759] localedata: missing yesstr and nostr for Tigrinya
1457 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1458 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1459 [21767] localedata: Missing Bislama locales
1460 [21768] localedata: Missing yesstr and nostr for aa_ET
1461 [21770] localedata: Missing Field in li_NL
1462 [21778] nptl: Robust mutex may deadlock
1463 [21779] libc: MicroBlaze segfaults when loading libpthread
1464 [21783] localedata: Fix int_select international_call_prefixes
1465 [21784] localedata: Inconsistency in country_isbn
1466 [21788] localedata: Missing Country Postal Abbreviations
1467 [21794] localedata: Added-country_isbn-for-Italy
1468 [21795] localedata: Add/Fix country_isbn for France
1469 [21796] localedata: Added country_isbn for Republic of Korea
1470 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1471 [21799] localedata: Added int_select international_call_prefixes
1472 [21801] localedata: Added int_select international_call_prefixes
1473 [21804] nptl: Double semicolon in thread-shared-types.h
1474 [21807] localedata: LC_ADDRESS fix for pap_CW
1475 [21808] localedata: Fix LC_ADDRESS for pap_AW
1476 [21821] localedata: Added country_name in mai_IN
1477 [21822] localedata: Fix LC_TIME for mai_IN
1478 [21823] localedata: missing yesstr, nostr for sa_IN
1479 [21825] localedata: Fix name_mrs for mag_IN
1480 [21828] localedata: 2.26 changelog should mention user visible changes
1481 with unicode 9.0
1482 [21835] localedata: Added Maithili language locale for Nepal
1483 [21838] localedata: Removed redundant data for the_NP
1484 [21839] localedata: Fix LC_MONETARY for ta_LK
1485 [21844] localedata: Fix Latin characters and Months Sequence.
1486 [21848] localedata: Fix mai_NP Title Name
1487
1488 \f
1489 Version 2.25
1490
1491 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1492 24731-2:2010, is supported to enable declarations of functions from that
1493 TR. Note that not all functions from that TR are supported by the GNU C
1494 Library.
1495
1496 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1497 18661-1:2014, is supported to enable declarations of functions and macros
1498 from that TS. Note that not all features from that TS are supported by
1499 the GNU C Library.
1500
1501 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1502 18661-4:2015, is supported to enable declarations of functions and macros
1503 from that TS. Note that most features from that TS are not supported by
1504 the GNU C Library.
1505
1506 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1507 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1508 Since the GNU C Library defaults to a much newer revision of POSIX, this
1509 will only affect programs that specifically request an old conformance
1510 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1511 see a change in the visible declarations, but a program compiled with
1512 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1513 will not.
1514
1515 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1516 defined by all multithreaded code, but glibc has not required this for
1517 many years.
1518
1519 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1520 means that in a future release, the macros “major”, “minor”, and “makedev”
1521 will only be available from <sys/sysmacros.h>.
1522
1523 These macros are not part of POSIX nor XSI, and their names frequently
1524 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1525 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1526 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1527 problem.
1528
1529 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1530 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1531 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1532
1533 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1534 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1535 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1536 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1537 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1538 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1539 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1540 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1541 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1542 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1543 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1544 WINT_WIDTH.
1545
1546 * New <math.h> features are added from TS 18661-1:2014:
1547
1548 - Signaling NaN macros: SNANF, SNAN, SNANL.
1549
1550 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1551 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1552 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1553
1554 - llogb functions: the llogb, llogbf and llogbl functions, and the
1555 FP_LLOGB0 and FP_LLOGBNAN macros.
1556
1557 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1558 fminmagf, fminmagl.
1559
1560 - Comparison macros: iseqsig.
1561
1562 - Classification macros: iscanonical, issubnormal, iszero.
1563
1564 - Total order functions: totalorder, totalorderf, totalorderl,
1565 totalordermag, totalordermagf, totalordermagl.
1566
1567 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1568
1569 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1570 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1571
1572 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1573 are added to libc. They convert a floating-point number into string.
1574
1575 * Most of glibc can now be built with the stack smashing protector enabled.
1576 It is recommended to build glibc with --enable-stack-protector=strong.
1577 Implemented by Nick Alcock (Oracle).
1578
1579 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1580 intended to be used instead of memset() to erase sensitive data after use;
1581 the compiler will not optimize out calls to explicit_bzero even if they
1582 are "unnecessary" (in the sense that no _correct_ program can observe the
1583 effects of the memory clear).
1584
1585 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1586 to float instead of double. This does not affect the ABI of any libraries
1587 that are part of the GNU C Library, but may affect the ABI of other
1588 libraries that use this type in their interfaces.
1589
1590 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1591 float_t and double_t types are now defined to long double instead of float
1592 and double. These options are not the default, and this does not affect
1593 the ABI of any libraries that are part of the GNU C Library, but it may
1594 affect the ABI of other libraries that use this type in their interfaces,
1595 if they are compiled or used with those options.
1596
1597 * The getentropy and getrandom functions, and the <sys/random.h> header file
1598 have been added.
1599
1600 * The buffer size for byte-oriented stdio streams is now limited to 8192
1601 bytes by default. Previously, on Linux, the default buffer size on most
1602 file systems was 4096 bytes (and thus remains unchanged), except on
1603 network file systems, where the buffer size was unpredictable and could be
1604 as large as several megabytes.
1605
1606 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1607 for the Linux quota interface which predates kernel version 2.4.22 has
1608 been removed.
1609
1610 * The malloc_get_state and malloc_set_state functions have been removed.
1611 Already-existing binaries that dynamically link to these functions will
1612 get a hidden implementation in which malloc_get_state is a stub. As far
1613 as we know, these functions are used only by GNU Emacs and this change
1614 will not adversely affect already-built Emacs executables. Any undumped
1615 Emacs executables, which normally exist only during an Emacs build, should
1616 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1617
1618 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1619 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1620 “no-ip6-dotint” had already been the default, and support for the
1621 “ip6-dotint” option was removed from the Internet in 2006.
1622
1623 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1624 flag from <resolv.h> have been removed. The option relied on a
1625 backwards-incompatible DNS extension which was never deployed on the
1626 Internet.
1627
1628 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1629 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1630 They were already unimplemented.
1631
1632 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1633 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1634 removed again from the IETF name lookup interface specification in RFC
1635 2553. Applications should use getaddrinfo instead.
1636
1637 * DNSSEC-related declarations and definitions have been removed from the
1638 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1639 decode the data part of DNSSEC record types. Previous versions of glibc
1640 only implemented minimal support for the previous version of DNSSEC, which
1641 is incompatible with the currently deployed version.
1642
1643 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1644 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1645 <arpa/nameser.h> header file because the distinction between RR types and
1646 meta-RR types is not officially standardized, subject to revision, and
1647 thus not suitable for encoding in a macro.
1648
1649 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1650 and rhook members of the res_state type in <resolv.h> have been removed.
1651 The glibc stub resolver did not support these hooks, but the header file
1652 did not reflect that.
1653
1654 * For multi-arch support it is recommended to use a GCC which has
1655 been built with support for GNU indirect functions. This ensures
1656 that correct debugging information is generated for functions
1657 selected by IFUNC resolvers. This support can either be enabled by
1658 configuring GCC with '--enable-gnu-indirect-function', or by
1659 enabling it by default by setting 'default_gnu_indirect_function'
1660 variable for a particular architecture in the GCC source file
1661 'gcc/config.gcc'.
1662
1663 * GDB pretty printers have been added for mutex and condition variable
1664 structures in POSIX Threads. When installed and loaded in gdb these pretty
1665 printers show various pthread variables in human-readable form when read
1666 using the 'print' or 'display' commands in gdb.
1667
1668 * Tunables feature added to allow tweaking of the runtime for an application
1669 program. This feature can be enabled with the '--enable-tunables' configure
1670 flag. The GNU C Library manual has details on usage and README.tunables has
1671 instructions on adding new tunables to the library.
1672
1673 * A new version of condition variables functions have been implemented in
1674 the NPTL implementation of POSIX Threads to provide stronger ordering
1675 guarantees.
1676
1677 * A new version of pthread_rwlock functions have been implemented to use a more
1678 scalable algorithm primarily through not using a critical section anymore to
1679 make state changes.
1680
1681 Security related changes:
1682
1683 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1684 have been created with makecontext could fail to terminate due to a
1685 missing .cantunwind annotation. This has been observed to lead to a hang
1686 (denial of service) in some Go applications compiled with gccgo. Reported
1687 by Andreas Schwab. (CVE-2016-6323)
1688
1689 * The DNS stub resolver functions would crash due to a NULL pointer
1690 dereference when processing a query with a valid DNS question type which
1691 was used internally in the implementation. The stub resolver now uses a
1692 question type which is outside the range of valid question type values.
1693 (CVE-2015-5180)
1694
1695 The following bugs are resolved with this release:
1696
1697 [4099] stdio: Overly agressive caching by stream i/o functions.
1698 [7065] build: Support building glibc with -fstack-protector or -fstack-
1699 protector-all
1700 [9842] localedata: en_CA: incorrect date format
1701 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1702 before it started waiting
1703 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1704 [15765] nptl: sem_open is wrongly a cancellation point
1705 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1706 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1707 wrong type
1708 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1709 library linked with pthread
1710 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1711 "FIXME: Ingo" issue)
1712 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1713 [17252] libc: getrandom and getentropy syscall
1714 [17730] malloc: thread-local storage is sometimes improperly free()'d
1715 after being __libc_memalign()'d
1716 [18241] stdio: failed fseek on memstream does not set errno and can fail
1717 when it shouldnt
1718 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1719 not
1720 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1721 PTHREAD_PRIO_INHERIT mutex on ARM
1722 [18784] network: res_query and related function crash for special record
1723 type queries (CVE-2015-5180)
1724 [19380] math: strtod does not raise "inexact"
1725 [19387] string: Integer overflow in memchr
1726 [19390] string: Integer overflow in strncat
1727 [19398] build: linknamespace tests fail with massively parallel build
1728 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1729 termination
1730 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1731 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1732 [19514] libc: [PATCH] Fix spelling errors in spelling
1733 "implement"/"implementation" in several places
1734 [19582] network: Deprecate RES_USE_INET6
1735 [19673] manual: clog10 docs appear to be erroneous
1736 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1737 causes a segmentation fault
1738 [19826] libc: invalid pointer returned from __tls_get_addr with static
1739 linking
1740 [20016] network: resolv: Remove hooks support from the API
1741 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1742 uninitialized GOT
1743 [20033] math: [x86_64] vectorized math function don't call the __finite
1744 versions
1745 [20116] nptl: use after free in pthread_create
1746 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1747 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1748 is always true.
1749 [20311] nptl: please install proc_service.h
1750 [20366] libc: Compilation errors in installed headers in strict-compliance
1751 modes
1752 [20370] malloc: malloc: Arena free list management is still racy
1753 (incorrect fix in bug 19243)
1754 [20386] libc: assert (X = 0) does not result in GCC warning
1755 [20432] malloc: malloc: Minimize interface required for interposition
1756 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1757 in _Unwind_Backtrace (CVE-2016-6323)
1758 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1759 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1760 libpthread.a
1761 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1762 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1763 [20477] network: resolv: incorrect double-checked locking related to
1764 _res_hconf
1765 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1766 information.
1767 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1768 penalty
1769 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1770 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1771 cause transition penalty
1772 [20517] math: sparcv9 missing fdiml compat symbol
1773 [20524] manual: strverscmp is inconsistent
1774 [20525] libc: <sys/quota.h> should be based on kernel headers
1775 [20539] math: GCC 7 -static -lm fails to link at -O3
1776 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1777 mcount.oS)
1778 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1779 Checking
1780 [20591] network: Remove obsolete DNSSEC support
1781 [20592] network: DNS resource record type classification macros in
1782 <arpa/nameser.h> are incorrect
1783 [20593] network: Update DNS RR type definitions
1784 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1785 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1786 multi-arch
1787 [20629] network: libresolv: Remove support for bitstring labels
1788 (RES_USEBSTRING)
1789 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1790 assertion: "maxidx >= 2"
1791 [20660] math: [arm] Use VSQRT
1792 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1793 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1794 [20689] libc: Test for FMA should also check for AVX.
1795 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1796 [20715] math: iszero macro breaks existing code
1797 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1798 [20728] libc: powerpc: Missing TOC stub in clone
1799 [20729] build: glibc-2.24 fails to build for i486 with -Os
1800 [20750] build: Build fails with default PIE enabled gcc-6.x
1801 [20768] math: [alpha] sqrt fegetenv namespace
1802 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1803 glibc
1804 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1805 __FLT_EVAL_METHOD__ is 2
1806 [20790] network: rpcgen buffer overrun in get_prog_declaration
1807 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1808 [20829] libc: crypt snprintf namespace
1809 [20847] libc: tst-vfork3 failure
1810 [20855] math: Default bits/mathdef.h has inappropriate float_t
1811 [20859] math: [sh4] FP_ILOGB0 invalid
1812 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1813 U+20AC), not same as GBK
1814 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1815 [20916] math: pow handling of sNaN arguments
1816 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1817 [20919] math: Bad pow (qNaN, 0) result with -lieee
1818 [20940] math: hypot sNaN handling
1819 [20947] math: fmax, fmin sNaN handling
1820 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1821 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1822 RES_USE_INET6
1823 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1824 [20973] nptl: robust mutexes: Lost wake-ups
1825 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1826 [20978] nis: strlen on null pointer in nss_nisplus
1827 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1828 wrong condition
1829 [21014] string: i686 memchr overflows internal pointer check
1830 [21019] libc: [mips] n32 lseek incorrect on overflow
1831 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1832 [21026] libc: [MIPS] readahead syscall is broken on n64
1833 [21028] math: Fallback fesetexceptflag should always succeed
1834 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1835 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1836 clang
1837 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1838 [21061] librt: [microblaze] librt lost clock_* exports
1839 [21073] libc: tunables: insecure environment variables passed to
1840 subprocesses with AT_SECURE
1841 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1842 \f
1843 Version 2.24
1844
1845 * The minimum Linux kernel version that this version of the GNU C Library
1846 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1847 kernel version 2.6.32 or later suffices (on architectures that already
1848 required kernel versions more recent than 3.2, those requirements remain
1849 unchanged). Linux 3.2 or later kernel headers are required on all
1850 architectures.
1851
1852 * The pap_AN locale has been deleted. This has been deprecated for a long
1853 time. It has been replaced by pap_AW & pap_CW, both of which have long
1854 been included in previous releases.
1855
1856 * The readdir_r and readdir64_r functions have been deprecated. It is
1857 recommended to use readdir and readdir64 instead.
1858
1859 * The type “union wait” has been removed. It was deprecated in the early
1860 1990s and never part of POSIX. Application code should use the int type
1861 instead of “union wait”.
1862
1863 * A new NSS action is added to facilitate large distributed system
1864 administration. The action, MERGE, allows remote user stores like LDAP
1865 to be merged into local user stores like /etc/groups in order to provide
1866 easy to use, updated, and managed sets of merged credentials. The new
1867 action can be used by configuring it in /etc/nsswitch.conf:
1868 group: files [SUCCESS=merge] nis
1869 Implemented by Stephen Gallagher (Red Hat).
1870
1871 * The deprecated __malloc_initialize_hook variable has been removed from the
1872 API.
1873
1874 * The long unused localedef --old-style option has been removed. It hasn't
1875 done anything in over 16 years. Scripts using this option can safely
1876 drop it.
1877
1878 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1879 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1880 return the next representable value in the direction of positive infinity
1881 and the nextdown functions return the next representable value in the
1882 direction of negative infinity. These are currently enabled as GNU
1883 extensions.
1884
1885 Security related changes:
1886
1887 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1888 could result in a stack overflow when getnetbyname was called with an
1889 overly long name. (CVE-2016-3075)
1890
1891 * Previously, getaddrinfo copied large amounts of address data to the stack,
1892 even after the fix for CVE-2013-4458 has been applied, potentially
1893 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1894 instead. Reported by Michael Petlan. (CVE-2016-3706)
1895
1896 * The glob function suffered from a stack-based buffer overflow when it was
1897 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1898 Reported by Alexander Cherepanov. (CVE-2016-1234)
1899
1900 * The Sun RPC UDP client could exhaust all available stack space when
1901 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1902 alloca plugin for GCC. (CVE-2016-4429)
1903
1904 * The IPv6 name server management code in libresolv could result in a memory
1905 leak for each thread which is created, performs a failing naming lookup,
1906 and exits. Over time, this could result in a denial of service due to
1907 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1908
1909 The following bugs are resolved with this release:
1910
1911 [1170] localedata: ne_NP: update Nepali locale definition file
1912 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1913 of MS-DOS.
1914 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1915 [6796] math: fdim() does not set errno on overflow
1916 [10354] libc: posix_spawn should use vfork() in more cases than presently
1917 [11213] localedata: localedata: add copyright disclaimer to locale files
1918 [12143] localedata: chr_US: new Cherokee locale
1919 [12450] localedata: sgs_LT: new locale
1920 [12676] localedata: ln_CD: new locale
1921 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1922 CLDR data
1923 [13304] math: fma, fmaf, fmal produce wrong results
1924 [14259] build: --localedir arg to configure is ignored
1925 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1926 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1927 [14934] localedata: es_CL: wrong first weekday chilean locale
1928 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1929 romanisation
1930 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1931 and +/-
1932 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1933 [15368] nptl: raise() is not async-signal-safe
1934 [15479] math: ceil, floor, round and trunc raise inexact exception
1935 [15578] localedata: kk_KZ: various updates
1936 [16003] localedata: pap_AN: punt old locale
1937 [16137] localedata: iw_IL: punt old locale
1938 [16190] localedata: eo: new esperanto locale
1939 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1940 [16742] malloc: race condition: pthread_atfork() called before first
1941 malloc() results in unexpected locking behaviour/deadlocks
1942 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1943 all locales
1944 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1945 [17565] localedata: pt_PT: wrong (work-)week start
1946 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1947 [17950] build: Build fails with -msse
1948 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1949 [18433] libc: posix_spawn does not return correctly upon failure to
1950 execute
1951 [18453] localedata: charmaps/IBM875: incorrect codes
1952 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1953 -Wsystem-headers
1954 [18896] localedata: he_IL: improvements for currency
1955 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1956 Romanian locale data
1957 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1958 symbol
1959 [19056] libc: Deprecate readdir_r
1960 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1961 language
1962 [19198] localedata: nl_NL: small improvements for Dutch locales
1963 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1964 nameservers (CVE-2016-5417)
1965 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1966 machine
1967 [19400] locale: Language missing in "iso-639.def", trivial fix in
1968 description
1969 [19431] malloc: Deadlock between fflush, getdelim, and fork
1970 [19505] libc: Incorrect file descriptor validity checks in
1971 posix_spawn_file_actions_add{open,close,dup2}
1972 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1973 when using RTLD_NEXT
1974 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1975 `intl/{gettextP,loadinfo}.h'
1976 [19534] libc: execle, execlp may use malloc
1977 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1978 [19573] network: res_nclose and __res_maybe_init disagree about name
1979 server initialization, breaking Hesiod
1980 [19575] localedata: Status of GB18030 tables
1981 [19581] localedata: sr_* date_fmt string contains additional newline
1982 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1983 Excavator core
1984 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1985 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1986 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1987 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1988 double range
1989 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1990 part zero incorrect
1991 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1992 equality tests
1993 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1994 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1995 [19633] locale: strfmon_l applies global locale to number formatting
1996 [19642] network: Memory leak in getnameinfo
1997 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1998 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1999 glibc-2.22
2000 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2001 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2002 'tst-numeric.c'
2003 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2004 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2005 low part
2006 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2007 result
2008 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2009 unaligned stack
2010 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2011 pointers and lengths in error-case.
2012 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2013 report errors on UTF-16 surrogates values.
2014 [19755] nscd: nscd assertion failure in gc
2015 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2016 [19759] libc: mempcpy shouldn't be inlined
2017 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2018 [19765] libc: s390 needs an optimized mempcpy
2019 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2020 NAME_MAX limit assumption (CVE-2016-1234)
2021 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2022 [19787] network: Missing and incorrect truncation checks in getnameinfo
2023 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2024 modes
2025 [19791] network: Assertion failure in res_query.c with un-connectable name
2026 server addresses
2027 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2028 [19822] math: libm.so install clobbers old version
2029 [19825] network: resolv: send_vc can return uninitialized data in second
2030 response to getaddrinfo
2031 [19830] network: nss_dns: should check RDATA length against buffer length
2032 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2033 confronted with A/AAAA records of invalid size
2034 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2035 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2036 [19853] stdio: Printing IBM long double in decimal with high precision is
2037 sometimes incorrect
2038 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2039 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2040 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2041 record types
2042 [19865] network: Assertion failure or memory leak in
2043 _nss_dns_getcanonname_r
2044 [19868] network: nss_dns: netent code does not skip over non-PTR records
2045 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2046 (CVE-2016-3075)
2047 [19881] string: Improve x86-64 memset
2048 [19907] string: Incorrect memcpy tests
2049 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2050 resolving symbols
2051 [19925] libc: termios.h XCASE namespace
2052 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2053 [19929] libc: limits.h NL_NMAX namespace
2054 [19931] stdio: Memory leak in vfprintf
2055 [19957] libc: clone(CLONE_VM) access invalid parent memory
2056 [19963] localedata: en_IL: New locale
2057 [19989] stdio: stdio.h cuserid namespace
2058 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2059 gethosts
2060 [19996] locale: langinfo.h nl_langinfo_l namespace
2061 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2062 [20010] network: getaddrinfo: Stack overflow in hostent translation
2063 (CVE-2016-3706)
2064 [20012] stdio: libio: fmemopen append mode failure
2065 [20014] stdio: stdio.h namespace for pre-threads POSIX
2066 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2067 [20023] libc: fcntl.h timespec namespace
2068 [20024] math: [x86_64] vectorized sincos trashes the stack
2069 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2070 [20041] time: sys/time.h timespec namespace
2071 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2072 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2073 [20051] libc: ttyslot in wrong header under wrong conditions
2074 [20054] libc: gethostname not declared for XPG4
2075 [20055] libc: termios.h missing tcgetsid for XPG4
2076 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2077 executable
2078 [20073] libc: sys/stat.h fchmod namespace
2079 [20074] libc: stdlib.h rand_r namespace
2080 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2081 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2082 XPG3
2083 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2084 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2085 (CVE-2016-4429)
2086 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2087 [20119] libc: Wrong mask for processors level type from CPUID
2088 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2089 AS not supporting AVX512
2090 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2091 sNaN argument
2092 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2093 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2094 argument
2095 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2096 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2097 [20178] libc: posix_spawn{p} should not call exit
2098 [20191] stdio: libio: vtables hardening
2099 [20195] string: FMA4 detection requires CPUID execution with register
2100 eax=0x80000001
2101 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2102 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2103 subnormals
2104 [20212] math: acos (sNaN) returns sNaN
2105 [20213] math: asin (sNaN) returns sNaN
2106 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2107 [20218] math: [i386] asinhl (sNaN) returns sNaN
2108 [20219] math: [i386] atanhl (sNaN) returns sNaN
2109 [20222] stdio: fopencookie: Mangle function pointers
2110 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2111 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2112 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2113 input
2114 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2115 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2116 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2117 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2118 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2119 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2120 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2121 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2122 [20240] math: modf (sNaN) returns sNaN
2123 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2124 [20250] math: frexp (sNaN) returns sNaN
2125 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2126 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2127 double rounding
2128 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2129 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2130 subscript is below array bounds [-Werror=array-bounds]
2131 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2132 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2133 (Only arm/linux)
2134 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2135 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2136 fallbacks
2137 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2138 "invalid" exceptions
2139 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2140 [20316] localedata: id_ID: Februari instead of Pebruari
2141 [20327] string: POWER8 strcasecmp returns incorrect result
2142 [20347] math: Failure: Test: j0_downward (0xap+0)
2143 [20348] libc: FAIL: misc/tst-preadvwritev64
2144 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2145 [20350] libc: There is no test for p{read,write}64
2146 [20357] math: Incorrect cos result for 1.5174239687223976
2147 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2148 \f
2149 Version 2.23
2150
2151 * Unicode 8.0.0 Support: Character encoding, character type info, and
2152 transliteration tables are all updated to Unicode 8.0.0, using new
2153 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2154 These updates cause user visible changes, such as the fixes for bugs
2155 89, 16061, and 18568.
2156
2157 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2158 kernel-internal CPU set size. This means that requests that change the
2159 CPU affinity which failed before (for example, an all-ones CPU mask) will
2160 now succeed. Applications that need to determine the effective CPU
2161 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2162 setting it because the kernel can adjust it (and the previous size check
2163 would not detect this in the majority of cases).
2164
2165 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2166 the following new symbols are used: fts64_children, fts64_close,
2167 fts64_open, fts64_read and fts64_set.
2168
2169 * getaddrinfo now detects certain invalid responses on an internal netlink
2170 socket. If such responses are received, an affected process will
2171 terminate with an error message of "Unexpected error <number> on netlink
2172 descriptor <number>" or "Unexpected netlink response of size <number> on
2173 descriptor <number>". The most likely cause for these errors is a
2174 multi-threaded application which erroneously closes and reuses the netlink
2175 file descriptor while it is used by getaddrinfo.
2176
2177 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2178 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2179 unnecessary serialization of memory allocation requests across threads.
2180 The defect is now corrected. Users should see a substantial increase in
2181 the concurent throughput of allocation requests for applications which
2182 trigger this bug. Affected applications typically create and
2183 destroy threads frequently. (Bug 19048 was reported and analyzed by
2184 Ericsson.)
2185
2186 * There is now a --disable-timezone-tools configure option for disabling the
2187 building and installing of the timezone related utilities (zic, zdump, and
2188 tzselect). This is useful for people who build the timezone data and code
2189 independent of the GNU C Library.
2190
2191 * The obsolete header <regexp.h> has been removed. Programs that require
2192 this header must be updated to use <regex.h> instead.
2193
2194 * The obsolete functions bdflush, create_module, get_kernel_syms,
2195 query_module and uselib are no longer available to newly linked binaries;
2196 the header <sys/kdaemon.h> has been removed. These functions and header
2197 were specific to systems using the Linux kernel and could not usefully be
2198 used with the GNU C Library on systems with version 2.6 or later of the
2199 Linux kernel.
2200
2201 * Optimized string, wcsmbs and memory functions for IBM z13.
2202 Implemented by Stefan Liebler.
2203
2204 * Newly linked programs that define a variable called signgam will no longer
2205 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2206 require signgam to be set by those functions must ensure that they use the
2207 variable provided by the GNU C Library and declared in <math.h>, without
2208 defining their own copy.
2209
2210 * The minimum GCC version that can be used to build this version of the GNU
2211 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2212 still be used to compile programs using the GNU C Library.
2213
2214 Security related changes:
2215
2216 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2217 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2218
2219 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2220 the pointer guard feature. It is always enabled. Previously,
2221 LD_POINTER_GUARD could be used to disable security hardening in binaries
2222 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2223 (CVE-2015-8777)
2224
2225 * An integer overflow in hcreate and hcreate_r could lead to an
2226 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2227
2228 * The catopen function no longer has unbounded stack usage. Reported by
2229 Max. (CVE-2015-8779)
2230
2231 * The nan, nanf and nanl functions no longer have unbounded stack usage
2232 depending on the length of the string passed as an argument to the
2233 functions. Reported by Joseph Myers. (CVE-2014-9761)
2234
2235 * A stack-based buffer overflow was found in libresolv when invoked from
2236 libnss_dns, allowing specially crafted DNS responses to seize control
2237 of execution flow in the DNS client. The buffer overflow occurs in
2238 the functions send_dg (send datagram) and send_vc (send TCP) for the
2239 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2240 family. The use of AF_UNSPEC triggers the low-level resolver code to
2241 send out two parallel queries for A and AAAA. A mismanagement of the
2242 buffers used for those queries could result in the response of a query
2243 writing beyond the alloca allocated buffer created by
2244 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2245 the overflow. Thanks to the Google Security Team and Red Hat for
2246 reporting the security impact of this issue, and Robert Holiday of
2247 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2248
2249 The following bugs are resolved with this release:
2250
2251 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2252 [887] math: Math library function "logb" and "nextafter" inconsistent
2253 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2254 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2255 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2256 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2257 use `mkstemp'
2258 [4404] localedata: German translation of "Alarm clock" is misleading
2259 [6799] math: nextafter() and nexttoward() doen't set errno on
2260 overflow/underflow errors
2261 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2262 overflow/underflow
2263 [10432] nis: _nss_nis_setnetgrent assertion failure
2264 [11460] libc: fts has no LFS support
2265 [12926] network: getaddrinfo()/make_request() may spin forever
2266 [13065] nptl: Race condition in pthread barriers
2267 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2268 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2269 are not contiguous
2270 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2271 [14912] libc: Rename non-installed bits/*.h headers
2272 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2273 [15367] math: Let gcc use __builtin_isinf
2274 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2275 [15421] math: lgamma wrongly sets signgam for ISO C
2276 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2277 FE_INVALID with argument out of range
2278 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2279 all exceptions
2280 [15786] dynamic-link: ifunc resolver functions can smash function
2281 arguments
2282 [15918] math: Unnecessary check for equality in hypotf()
2283 [16061] localedata: Review / update transliteration data
2284 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2285 should include
2286 [16141] time: strptime %z offset restriction
2287 [16171] math: drem should be alias of remainder
2288 [16296] math: fegetround is pure?
2289 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2290 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2291 GNU/Linux
2292 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2293 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2294 arguments
2295 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2296 "Exception "Inexact" set" on ppc32
2297 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2298 [16517] math: Missing underflow exception from tanf/tan/tanl
2299 [16519] math: Missing underflow exception from sinhf
2300 [16520] math: Missing underflow exception from tanhf
2301 [16521] math: Missing underflow exception from exp2
2302 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2303 rounding results
2304 [16734] stdio: fopen calls mmap to allocate its buffer
2305 [16961] math: nan function incorrect handling of bad sequences
2306 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2307 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2308 [16985] locale: localedef: confusing error message when opening output
2309 fails
2310 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2311 [17197] locale: Redundant shift character in iconv conversion output at
2312 block boundary
2313 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2314 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2315 [17250] dynamic-link: static linking breaks nss loading
2316 (getaddrinfo/getpwnam/etc...)
2317 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2318 4.7?
2319 [17441] math: isnan() should use __builtin_isnan() in GCC
2320 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2321 (related to lock elision)
2322 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2323 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2324 [17887] time: strptime should be able to parse "+01:00" style timezones
2325 [17905] libc: catopen() Multiple unbounded stack allocations
2326 (CVE-2015-8779)
2327 [18084] libc: backtrace (..., 0) dumps core on x86
2328 [18086] libc: nice() sets errno to 0 on success
2329 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2330 is too large (CVE-2015-8778)
2331 [18251] dynamic-link: SONAME missing when audit modules provides path
2332 [18265] libc: add attributes for wchar string and memory functions
2333 [18370] math: csqrt missing underflows
2334 [18421] libc: [hppa] read-only segment has dynamic relocations
2335 [18472] libc: Obsolete syscall wrappers should be compat symbols
2336 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2337 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2338 [18525] localedata: Remove locale timezone information
2339 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2340 [18568] localedata: Update locale data to Unicode 8.0
2341 [18589] locale: sort-test.sh fails at random
2342 [18595] math: ctan, ctanh missing underflows
2343 [18604] libc: assert macro-expands its argument
2344 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2345 contains a vector instruction exception.
2346 [18611] math: j1, jn missing errno setting on underflow
2347 [18618] localedata: sync Chechen locale definitions with other *_RU
2348 locales
2349 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2350 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2351 [18665] network: In send_dg, the recvfrom function is NOT always using the
2352 buffer size of a newly created buffer (CVE-2015-7547)
2353 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2354 missing break ?
2355 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2356 32bit processes
2357 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2358 [18699] math: tilegx cproj() for various complex infinities does not yield
2359 infinity
2360 [18724] libc: Harden put*ent functions against data injection
2361 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2362 [18755] build: build errors with -DNDEBUG
2363 [18757] stdio: fmemopen fails to set errno on failure
2364 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2365 be forced unloaded
2366 [18781] libc: openat64 lacks O_LARGEFILE
2367 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2368 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2369 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2370 [18790] math: [ldbl-128ibm] tanhl inaccurate
2371 [18795] libc: stpncpy fortification misses buffer lengths that are
2372 statically too large
2373 [18796] build: build fails for --disable-mathvec
2374 [18803] math: hypot missing underflows
2375 [18820] stdio: fmemopen may leak memory on failure
2376 [18823] math: csqrt spurious underflows
2377 [18824] math: fma spurious underflows
2378 [18825] math: pow missing underflows
2379 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2380 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2381 [18870] build: sem_open.c fails to compile with missing symbol
2382 FUTEX_SHARED
2383 [18872] stdio: Fix memory leak in printf_positional
2384 [18873] libc: posix_fallocate overflow check ineffective
2385 [18875] math: Excess precision leads incorrect libm
2386 [18877] libc: arm: mmap offset regression
2387 [18887] libc: memory corruption when using getmntent on blank lines
2388 [18918] localedata: hu_HU: change time to HH:MM:SS format
2389 [18921] libc: Regression: extraneous stat() and fstat() performed by
2390 opendir()
2391 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2392 binaries (CVE-2015-8777)
2393 [18951] math: tgamma missing underflows
2394 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2395 signgam
2396 [18953] localedata: lt_LT: change currency symbol to the euro
2397 [18956] math: powf inaccuracy
2398 [18961] math: [i386] exp missing underflows
2399 [18966] math: [i386] exp10 missing underflows
2400 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2401 [18969] build: multiple string test failures due to missing locale
2402 dependencies
2403 [18970] libc: Reference of pthread_setcancelstate in libc.a
2404 [18977] math: float / long double Bessel functions not in XSI POSIX
2405 [18980] math: i386 libm functions return with excess range and precision
2406 [18981] math: i386 scalb*, ldexp return with excess range and precision
2407 [18982] stdio: va_list and vprintf
2408 [18985] time: Passing out of range data to strftime() causes a segfault
2409 (CVE-2015-8776)
2410 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2411 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2412 [19012] locale: iconv_open leaks memory on error path
2413 [19016] math: clog, clog10 inaccuracy
2414 [19018] nptl: Mangle function pointers in tls_dtor_list
2415 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2416 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2417 [19048] malloc: malloc: arena free list can become cyclic, increasing
2418 contention
2419 [19049] math: [powerpc] erfc incorrect zero sign
2420 [19050] math: [powerpc] log* incorrect zero sign
2421 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2422 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2423 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2424 [19074] network: Data race in _res_hconf_reorder_addrs
2425 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2426 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2427 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2428 rounding modes
2429 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2430 ILP32
2431 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2432 [19086] manual: posix_fallocate64 documented argument order is wrong.
2433 [19088] math: lround, llround missing exceptions close to overflow
2434 threshold
2435 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2436 [19095] math: dbl-64 lrint incorrect for 64-bit long
2437 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2438 [19124] dynamic-link: ld.so failed to build with older assmebler
2439 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2440 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2441 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2442 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2443 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2444 pthread_setaffinity_np
2445 [19156] math: [ldbl-128] j0l spurious underflows
2446 [19164] nptl: tst-getcpu fails with many possible CPUs
2447 [19168] math: math/test-ildoubl and math/test-ldouble failure
2448 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2449 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2450 prelink
2451 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2452 (FE_NOMASK_ENV) do not clear SSE exceptions
2453 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2454 _int_new_arena/reused_arena
2455 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2456 [19201] math: dbl-64 remainder incorrect sign of zero result
2457 [19205] math: bits/math-finite.h conditions do not match math.h and
2458 bits/mathcalls.h
2459 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2460 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2461 for C99-based standards
2462 [19212] libc: features.h not -Wundef clean
2463 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2464 math-only
2465 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2466 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2467 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2468 disabled
2469 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2470 "inexact" exceptions
2471 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2472 arguments
2473 [19242] libc: strtol incorrect in Turkish locales
2474 [19243] malloc: reused_arena can pick an arena on the free list, leading
2475 to an assertion failure and reference count corruption
2476 [19253] time: tzset() ineffective when temporary TZ did not include DST
2477 rules
2478 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2479 [19270] math: [hppa] Shared libm missing __isnanl
2480 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2481 MAP_HUGETLB and MAP_STACK defines
2482 [19313] nptl: Wrong __cpu_mask for x32
2483 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2484 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2485 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2486 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2487 [19363] time: x32: times() return value wrongly truncates/sign extends
2488 from 32bit
2489 [19367] dynamic-link: Improve branch prediction on Silvermont
2490 [19369] network: Default domain name not reset by res_ninit when "search"
2491 / "domain" entry is removed from resolv.conf
2492 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2493 subnormals
2494 [19385] localedata: bg_BG: time separator should be colon, not comma
2495 [19408] libc: linux personality syscall wrapper may erroneously return an
2496 error on 32-bit architectures
2497 [19415] libc: dladdr returns wrong names on hppa
2498 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2499 IBM905, IBM907, and IBM909
2500 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2501 [19443] build: build failures with -DDEBUG
2502 [19451] build: Make check fails on test-double-vlen2
2503 [19462] libc: Glibc failed to build with -Os
2504 [19465] math: Wrong code with -Os
2505 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2506 -Os
2507 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2508 CPU's.
2509 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2510 [19486] math: S390: Math tests fail with "Exception Inexact set".
2511 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2512 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2513 architectures
2514 [19590] math: Fail to build shared objects that use libmvec.so functions.
2515 \f
2516 Version 2.22
2517
2518 * The following bugs are resolved with this release:
2519
2520 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2521 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2522 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2523 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2524 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2525 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2526 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2527 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2528 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2529 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2530 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2531 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2532 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2533 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2534 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2535 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2536 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2537 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2538 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2539 18657, 18676, 18694, 18696, 18887.
2540
2541 * Cache information can be queried via sysconf() function on s390 e.g. with
2542 _SC_LEVEL1_ICACHE_SIZE as argument.
2543
2544 * A buffer overflow in gethostbyname_r and related functions performing DNS
2545 requests has been fixed. If the NSS functions were called with a
2546 misaligned buffer, the buffer length change due to pointer alignment was
2547 not taken into account. This could result in application crashes or,
2548 potentially arbitrary code execution, using crafted, but syntactically
2549 valid DNS responses. (CVE-2015-1781)
2550
2551 * The time zone file parser has been made more robust against crafted time
2552 zone files, avoiding heap buffer overflows related to the processing of
2553 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2554 large time zone data files. Overly long time zone specifiers in the TZ
2555 variable no longer result in stack overflows and crashes.
2556
2557 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2558 for LD and GD on x86 and x86-64, has been implemented. You will need
2559 binutils-2.24 or later to enable this optimization.
2560
2561 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2562 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2563 Hat). These updates cause user visible changes, such as the fix for bug
2564 17998.
2565
2566 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2567 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2568 condition in some applications.
2569
2570 * Added vector math library named libmvec with the following vectorized x86_64
2571 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2572 pow, powf.
2573 The library can be disabled with --disable-mathvec. Use of the functions is
2574 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2575 Shared library libmvec.so is linked in as needed when using -lm (no need to
2576 specify -lmvec explicitly for not static builds).
2577 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2578
2579 * A new fmemopen implementation has been added with the goal of POSIX
2580 compliance. The new implementation fixes the following long-standing
2581 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2582 old implementation is still present for use be by existing binaries.
2583
2584 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2585 releases. It has been fixed to match 2.19 and older, but binaries built
2586 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2587
2588 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2589 Contributed by Roland McGrath (Google).
2590
2591 * The header <regexp.h> is deprecated, and will be removed in a future
2592 release. Use of this header will trigger a deprecation warning.
2593 Application developers should update their code to use <regex.h> instead.
2594
2595 This header was formerly part of SUS, but was deprecated in 1994 and
2596 removed from the standard in 2001. Also, the glibc implementation
2597 leaks memory. See BZ#18681 for more details.
2598 \f
2599 Version 2.21
2600
2601 * The following bugs are resolved with this release:
2602
2603 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2604 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2605 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2606 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2607 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2608 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2609 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2610 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2611 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2612 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2613 17892.
2614
2615 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2616 too little memory for the to-be-scanned arguments and overflow the
2617 allocated buffer. The implementation now correctly computes the required
2618 buffer size when using malloc, and switches to malloc from alloca as
2619 intended.
2620
2621 * A new semaphore algorithm has been implemented in generic C code for all
2622 machines. Previous custom assembly implementations of semaphore were
2623 difficult to reason about or ensure that they were safe. The new version
2624 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2625 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2626 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2627
2628 * Port to Altera Nios II has been contributed by Mentor Graphics.
2629
2630 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2631 implementations for powerpc64/powerpc64le.
2632 Implemented by Adhemerval Zanella (IBM).
2633
2634 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2635 and powerpc64le. This may improve lock scaling of existing programs on
2636 HTM capable systems. The lock elision code is only enabled with
2637 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2638 powerpc will issue a transaction abort on every syscall to avoid side
2639 effects being visible outside transactions.
2640
2641 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2642 AArch64. Contributed by ARM Ltd.
2643
2644 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2645
2646 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2647 under certain input conditions resulting in the execution of a shell for
2648 command substitution when the application did not request it. The
2649 implementation now checks WRDE_NOCMD immediately before executing the
2650 shell and returns the error WRDE_CMDSUB as expected.
2651
2652 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2653 processing format strings with a large number of format specifiers.
2654
2655 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2656 infinite loop if the DNS response contained a PTR record of an unexpected
2657 format.
2658
2659 * The minimum GCC version that can be used to build this version of the GNU
2660 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2661 still be used to compile programs using the GNU C Library.
2662
2663 * The GNU C Library is now built with -Werror by default. This can be
2664 disabled by configuring with --disable-werror.
2665
2666 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2667
2668 * The obsolete sigvec function has been removed. This was the original
2669 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2670 programs have been using instead for about 25 years. Of course, ABI
2671 compatibility for old binaries using sigvec remains intact.
2672
2673 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2674 with newer versions of bison.
2675
2676 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2677 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2678 registers overlay two consecutive single-precision registers. MIPS32R2
2679 introduced a new FPU mode (FR=1) where double-precision registers extend the
2680 corresponding single-precision registers which is incompatible with the
2681 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2682 require the use of FR=1 making a transition necessary. New o32 ABI
2683 extensions enable users to migrate over time from the original o32 ABI
2684 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2685 tracks the ABI of any loaded object and verifies that new objects are
2686 compatible. Mode transitions will also be requested as required and
2687 unsupportable objects will be rejected. The ABI checks include both soft and
2688 hard float ABIs for o32, n32 and n64.
2689
2690 GCC 5 with GNU binutils 2.25 onwards:
2691 It is strongly recommended that all o32 system libraries are built using the
2692 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2693 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2694 compiler using --with-fp-32=xx to set this by default.
2695 \f
2696 Version 2.20
2697
2698 * The following bugs are resolved with this release:
2699
2700 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2701 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2702 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2703 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2704 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2705 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2706 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2707 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2708 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2709 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2710 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2711 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2712 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2713 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2714 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2715
2716 * Reverted change of ABI data structures for s390 and s390x:
2717 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2718 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2719 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2720 are preserved pointing straight to the same implementation as the old ones.
2721 Given that, new callers will simply provide a too-big buffer to these
2722 functions. Any applications/libraries out there that embed jmp_buf or
2723 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2724 against 2.19 headers will have to rebuilt again. This is necessary in any
2725 case to revert the breakage in their ABI caused by the glibc change.
2726
2727 * Support for file description locks is added to systems running the
2728 Linux kernel. The standard file locking interfaces are extended to
2729 operate on file descriptions, not file descriptors, via the use of
2730 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2731 are associated with an open file instead of a process.
2732
2733 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2734
2735 * The minimum Linux kernel version that this version of the GNU C Library
2736 can be used with is 2.6.32.
2737
2738 * Running the testsuite no longer terminates as soon as a test fails.
2739 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2740 with PASS or FAIL lines for individual tests. A summary of the results is
2741 printed, including a list of failing lists, and "make check" exits with
2742 error status if there were any unexpected failures. "make check
2743 stop-on-test-failure=y" may be used to keep the old behavior.
2744
2745 * The am33 port, which had not worked for several years, has been removed
2746 from ports.
2747
2748 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2749 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2750 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2751 conflicted with POSIX (support for which was removed in 2.19), the
2752 interfaces those macros enabled remain available when compiling with
2753 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2754 test macros defined.
2755
2756 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2757
2758 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2759 This may improve lock scaling of existing programs on TX capable systems.
2760 The lock elision code is only built with --enable-lock-elision=yes and
2761 then requires a GCC version supporting the TX builtins. With lock elision
2762 default mutexes are elided via __builtin_tbegin, if the cpu supports
2763 transactions. By default lock elision is not enabled and the elision code
2764 is not built.
2765
2766 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2767 copy the path argument. This allowed programs to cause posix_spawn to
2768 deference a dangling pointer, or use an unexpected pathname argument if
2769 the string was modified after the posix_spawn_file_actions_addopen
2770 invocation.
2771
2772 * All supported architectures now use the main glibc sysdeps directory
2773 instead of some being in a separate "ports" directory (which was
2774 distributed separately before glibc 2.17).
2775
2776 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2777 On configurations that support it (all Linux configurations), it's now
2778 used regardless of the --enable-add-ons switch to configure. It is no
2779 longer possible to build such configurations without pthreads support.
2780
2781 * Locale names, including those obtained from environment variables (LANG
2782 and the LC_* variables), are more tightly checked for proper syntax.
2783 setlocale will now fail (with EINVAL) for locale names that are overly
2784 long, contain slashes without starting with a slash, or contain ".." path
2785 components. (CVE-2014-0475) Previously, some valid locale names were
2786 silently replaced with the "C" locale when running in AT_SECURE mode
2787 (e.g., in a SUID program). This is no longer necessary because of the
2788 additional checks.
2789
2790 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2791 with application code using Intel MPX instructions. (With all previous
2792 versions, the MPX register state could be clobbered when making calls
2793 into or out of a shared library.) Note that while the new dynamic
2794 linker is compatible with all known x86 hardware whether or not it
2795 supports Intel MPX, some x86 instruction-set emulators might fail to
2796 handle the new instruction encodings. This is known to affect Valgrind
2797 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2798 release), and might affect other tools that do instruction emulation.
2799
2800 * Support for loadable gconv transliteration modules has been removed.
2801 The support for transliteration modules has been non-functional for
2802 over a decade, and the removal is prompted by security defects. The
2803 normal gconv conversion modules are still supported. Transliteration
2804 with //TRANSLIT is still possible, and the //IGNORE specifier
2805 continues to be supported. (CVE-2014-5119)
2806
2807 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2808 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2809 resulting a denial-of-service security vulnerability in applications which
2810 use functions related to iconv. (CVE-2014-6040)
2811 \f
2812 Version 2.19
2813
2814 * The following bugs are resolved with this release:
2815
2816 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2817 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2818 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2819 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2820 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2821 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2822 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2823 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2824 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2825 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2826 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2827 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2828 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2829 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2830 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2831 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2832 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2833 16431, 16453, 16474, 16506, 16510, 16529
2834
2835 * Slovenian translations for glibc messages have been contributed by the
2836 Translation Project's Slovenian team of translators.
2837
2838 * The public headers no longer use __unused nor __block. This change is to
2839 support compiling programs that are derived from BSD sources and use
2840 __unused internally, and to support compiling with Clang's -fblock
2841 extension which uses __block.
2842
2843 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2844 large collation sequences to optimize multiple passes. This cache
2845 computation may overflow for large collation sequences and may cause a
2846 stack or buffer overflow. This is now fixed to use a slower algorithm
2847 which does not use a cache if there is an integer overflow.
2848
2849 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2850 rules for large collation sequences to optimize multiple passes and falls
2851 back to alloca if malloc fails, resulting in a possible stack overflow.
2852 The implementation now falls back to an uncached collation sequence lookup
2853 if malloc fails.
2854
2855 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2856 initialized for static applications resulting in the security feature
2857 being disabled. The pointer guard is now correctly initialized to a
2858 random value for static applications. Existing static applications need
2859 to be recompiled to take advantage of the fix (bug 15754).
2860
2861 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2862 to the d_name member of struct dirent, or omit the terminating NUL
2863 character. (Bugzilla #14699).
2864
2865 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2866 aligned_alloc functions could allocate too few bytes or corrupt the
2867 heap when passed very large allocation size values (Bugzilla #15855,
2868 #15856, #15857).
2869
2870 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2871 for AF_INET6 has been fixed (Bugzilla #16072).
2872
2873 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2874 pap_CW, quz_PE, the_NP.
2875
2876 * Substantially revised locales: gd_GB, ht_HT
2877
2878 * The LC_ADDRESS field was updated to support country_car for almost all
2879 supported locales.
2880
2881 * ISO 1427 definitions were updated.
2882
2883 * ISO 3166 definitions were updated.
2884
2885 * The localedef utility now supports --big-endian and --little-endian
2886 command-line options to generate locales for a different system from that
2887 for which the C library was built.
2888
2889 * Binary locale files now only depend on the endianness of the system for
2890 which they are generated and not on other properties of that system. As a
2891 consequence, binary files generated with new localedef may be incompatible
2892 with old versions of the GNU C Library, and binary files generated with
2893 old localedef may be incompatible with this version of the GNU C Library,
2894 in the following circumstances:
2895
2896 + Locale files may be incompatible on m68k systems.
2897
2898 + Locale archive files (but not separate files for individual locales) may
2899 be incompatible on systems where plain "char" is signed.
2900
2901 * The configure option --disable-versioning has been removed. Builds with
2902 --disable-versioning had not worked for several years.
2903
2904 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2905 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2906
2907 * SystemTap probes for malloc have been introduced.
2908
2909 * SystemTap probes for slow multiple precision fallback paths of
2910 transcendental functions have been introduced.
2911
2912 * Support for powerpc64le has been added.
2913
2914 * The soft-float powerpc port now supports e500 processors.
2915
2916 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2917
2918 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2919 set of header declarations that are enabled by default, even when other
2920 feature test macros or compiler options such as -std=c99 would otherwise
2921 disable some of those declarations.
2922
2923 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2924 conflict with POSIX. The libbsd-compat library (which was a dummy library
2925 that did nothing) has also been removed.
2926
2927 * Preliminary documentation about Multi-Thread, Async-Signal and
2928 Async-Cancel Safety has been added.
2929
2930 * Change of ABI data structures for s390 and s390x:
2931 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2932 allow for future hardware extensions. All library functions that accept or
2933 return these structures were versioned in order to provide backwards
2934 compatibility with existing code. However, not all cases can be handled
2935 with interface versioning. If an application embeds either structure into
2936 another structure and that structure is passed to another compilation unit
2937 which expects the newer larger-sized object then any access to the new
2938 fields will result in undefined behaviour. Similarly any access to fields
2939 that were moved by the enlarging of the structures will cause undefined
2940 behaviour. To fix the undefined behaviour all objects that are part of the
2941 final in-memory image must agree on the size of structures, and this may
2942 require recompilation.
2943 \f
2944 Version 2.18
2945
2946 * The following bugs are resolved with this release:
2947
2948 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2949 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2950 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2951 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2952 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2953 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2954 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2955 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2956 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2957 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2958 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2959 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2960 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2961 15755, 15759.
2962
2963 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2964 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2965 Distributions can re-enable building and using pt_chown via the new configure
2966 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2967 considerable security risks and should only be used if the distribution
2968 understands and accepts the risks.
2969
2970 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2971 #15078).
2972
2973 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2974 fixed (Bugzilla #15330).
2975
2976 * Add support for calling C++11 thread_local object destructors on thread
2977 and program exit. This needs compiler support for offloading C++11
2978 destructor calls to glibc.
2979
2980 * Improved worst case performance of libm functions with double inputs and
2981 output.
2982
2983 * Support for priority inherited mutexes in pthread condition variables on
2984 non-x86 architectures.
2985
2986 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2987
2988 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2989
2990 * Optimized string functions for ARM. Implemented by Will Newton and
2991 Richard Henderson.
2992
2993 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2994
2995 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2996 Richard Henderson.
2997
2998 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2999 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3000
3001 * Added a benchmark framework to track performance of functions in glibc.
3002
3003 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3004 It is based on draft TS 18661 and currently enabled as a GNU extension.
3005
3006 * On Linux, the clock function now uses the clock_gettime system call
3007 for improved precision, rather than old times system call.
3008
3009 * Added support for version-3 tz format files. This is needed when using
3010 the tz database release 2013e or later, and affects a few unusual cases --
3011 currently only TZ='America/Godthab' for time stamps after 2037.
3012
3013 * Added new API functions pthread_getattr_default_np and
3014 pthread_setattr_default_np to get and set the default pthread
3015 attributes of a process.
3016
3017 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3018 This may improve lock scaling of existing programs on TSX capable systems.
3019 When the --enable-lock-elision=yes parameter is specified at configure
3020 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3021 mutexes.
3022
3023 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3024 availability). Implemented by Ryan S. Arnold.
3025
3026 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3027 \f
3028 Version 2.17
3029
3030 * The following bugs are resolved with this release:
3031
3032 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3033 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3034 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3035 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3036 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3037 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3038 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3039 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3040 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3041 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3042 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3043 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3044 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3045 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3046 14879, 14889, 14893, 14898, 14914.
3047
3048 * Optimization of memcpy for MIPS.
3049
3050 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3051 EMFILE has been fixed (Bugzilla #14889).
3052
3053 * The add-on ports collection is now distributed in the "ports" subdirectory
3054 of the main GNU C Library distribution, rather than separately.
3055
3056 * Port to ARM AArch64 contributed by Linaro.
3057
3058 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3059 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3060 zEnterprise z196.
3061 Implemented by Andreas Krebbel.
3062
3063 * The new function secure_getenv allows secure access to the environment,
3064 returning NULL if running in a SUID/SGID process. This function replaces
3065 the internal function __secure_getenv.
3066
3067 * SystemTap static probes have been added into the dynamic linker.
3068 Implemented by Gary Benson.
3069
3070 * Optimizations of string functions strstr, strcasestr and memmem.
3071 Implemented by Maxim Kuvyrkov.
3072
3073 * The minimum Linux kernel version that this version of the GNU C Library
3074 can be used with is 2.6.16.
3075
3076 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3077 powerpc POWER7. Implemented by Will Schmidt.
3078
3079 * New configure option --disable-nscd builds the C library such that it
3080 never attempts to contact the Name Service Caching Daemon (nscd).
3081 New configure option --disable-build-nscd avoids building nscd itself;
3082 this is the default if --disable-nscd is used.
3083
3084 * Improved support for cross-compilation, including cross-testing and
3085 bootstrap builds without a previously built glibc.
3086
3087 * Several testsuite tests are now able to test multiple IFUNC variants of an
3088 interface, rather than just testing the one that would be chooen by
3089 default.
3090
3091 * New configure options --with-bugurl and --with-pkgversion, for
3092 distributors to use to embed their bug-reporting and package version
3093 information in --help and --version output.
3094
3095 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3096 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3097 allows creation of chroots without the procfs mounted on /proc.
3098
3099 * The `crypt' function now fails if passed salt bytes that violate the
3100 specification for those values. On Linux, the `crypt' function will
3101 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3102 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3103 when the mode is enabled.
3104
3105 * The `clock_*' suite of functions (declared in <time.h>) is now available
3106 directly in the main C library. Previously it was necessary to link with
3107 -lrt to use these functions. This change has the effect that a
3108 single-threaded program that uses a function such as `clock_gettime' (and
3109 is not linked with -lrt) will no longer implicitly load the pthreads
3110 library at runtime and so will not suffer the overheads associated with
3111 multi-thread support in other code such as the C++ runtime library.
3112
3113 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3114 sat_IN, and szl_PL.
3115 \f
3116 Version 2.16
3117
3118 * The following bugs are resolved with this release:
3119
3120 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3121 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3122 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3123 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3124 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3125 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3126 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3127 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3128 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3129 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3130 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3131 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3132 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3133 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3134 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3135 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3136 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3137 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3138 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3139 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3140 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3141 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3142 14277, 14278.
3143
3144 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3145 configuring glibc with:
3146 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3147 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3148 Implemented by H.J. Lu.
3149
3150 * ISO C11 support:
3151
3152 + define static_assert
3153
3154 + do not declare gets
3155
3156 + declare at_quick_exit and quick_exit also for ISO C11
3157
3158 + aligned_alloc. NB: The code is deliberately allows the size parameter
3159 to not be a multiple of the alignment. This is a moronic requirement
3160 in the standard but it is only a requirement on the caller, not the
3161 implementation.
3162
3163 + timespec_get added
3164
3165 + uchar.h support added
3166
3167 + CMPLX, CMPLXF, CMPLXL added
3168
3169 Implemented by Ulrich Drepper.
3170
3171 * Support for the IA-64 has been moved to ports.
3172
3173 * Remove support for anything but ELF binary format
3174
3175 * Checking versions of poll, ppoll added.
3176 Implemented by Ulrich Drepper.
3177
3178 * More generic and 64-bit performance optimizations to math functions.
3179 Implemented by Ulrich Drepper.
3180
3181 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3182 headers and functions available at compile time as they were before
3183 version 2.14. This option will be removed at some time in the future
3184 after the TI-RPC library becomes fully sufficient for the needs of
3185 existing applications.
3186
3187 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3188 Note that glibc is not expected to work with any Linux kernel version
3189 before 2.6.
3190
3191 * New header <sys/auxv.h> and function getauxval allowing easy access to
3192 the AT_* key-value pairs passed from the Linux kernel. The header also
3193 defines the HWCAP_* bits associated with the AT_HWCAP key.
3194
3195 * New locales: mag_IN
3196
3197 * New configure option --enable-systemtap builds SystemTap static probes
3198 into libc for setjmp and longjmp and into libpthread for various operations.
3199 So far the setjmp/longjmp probes and some of the libpthread probes are
3200 provided only for i*86 and x86_64.
3201 Implemented by Roland McGrath and Rayson Ho.
3202
3203 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3204
3205 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3206 and Will Schmidt.
3207
3208 * More optimized functions for SPARC. Implemented by David S. Miller.
3209
3210 * Improved support for cross-compilation, especially bootstrap builds
3211 without a previously built glibc.
3212
3213 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3214 Chris Metcalf from Tilera.
3215
3216 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3217 now supported for ARM processors.
3218
3219 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3220 as the name of the dynamic linker, to distinguish it from the
3221 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3222
3223 * Support for CRIS has been removed from ports.
3224
3225 * A new class of installed header has been documented for low-level
3226 platform-specific functionality. PowerPC added the first instance with a
3227 function to provide time base register access. Contributed by Tulio
3228 Magno Quites Machado Filho.
3229
3230 * ix86 configurations now install header files that are consistent with
3231 what x86-64 configurations install. These same header files can be used
3232 for -m32, -m64, or -mx32 builds.
3233 Contributed by H.J. Lu.
3234
3235 * Math library bug fixes. A thorough audit of all open math library bugs was
3236 conducted by Joseph Myers. Significant progress was made on many math
3237 library bugs resulting in more accurate exceptions and function results.
3238 Many thanks to all those that contributed including Andreas Jaeger for his
3239 patch review and work on the x87 trigonometric instruction issues.
3240
3241 * Timezone data is no longer installed. Timezone-related binaries and scripts
3242 will continue to be installed. Users should obtain their timezone data from
3243 their distribution provider or from the tzdata package at
3244 <ftp://munnari.oz.au/pub/>.
3245 \f
3246 Version 2.15
3247
3248 * The following bugs are resolved with this release:
3249
3250 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3251 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3252 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3253 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3254 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3255 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3256 13484, 13506, 13515, 13523, 13524, 13538, 13540
3257
3258 * New program pldd to list loaded object of a process
3259 Implemented by Ulrich Drepper.
3260
3261 * Add nss_db support back to glibc. No more dependency on Berkeley db
3262 and support for initgroups lookups.
3263 Implemented by Ulrich Drepper.
3264
3265 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3266 Contributed by HJ Lu.
3267
3268 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3269 Contributed by HJ Lu.
3270
3271 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3272 on x86-32 and x86-64.
3273 Contributed by Liubov Dmitrieva.
3274
3275 * Optimized strchr and strrchr for SSE on x86-32.
3276 Contributed by Liubov Dmitrieva.
3277
3278 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3279 for x86-64 and x86-32.
3280 Contributed by Liubov Dmitrieva.
3281
3282 * New interfaces: scandirat, scandirat64
3283 Implemented by Ulrich Drepper.
3284
3285 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3286 Implemented by Ulrich Drepper.
3287
3288 * nscd now also caches the netgroup database.
3289 Implemented by Ulrich Drepper.
3290
3291 * Integrate libm with gcc's -ffinite-math-only option.
3292 Implemented by Ulrich Drepper.
3293
3294 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3295 to math functions. Implemented by Ulrich Drepper.
3296
3297 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3298 Implemented by Ulrich Drepper.
3299
3300 * New Linux interfaces: process_vm_readv, process_vm_writev
3301
3302 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3303 Implemented by Ulrich Drepper.
3304
3305 * Optimized nearbyint and strcasecmp for PPC.
3306 Implemented by Adhemerval Zanella.
3307
3308 * New locales: bho_IN, unm_US, es_CU, ta_LK
3309 \f
3310 Version 2.14
3311
3312 * The following bugs are resolved with this release:
3313
3314 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3315 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3316 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3317 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3318 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3319 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3320 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3321 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3322 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3323 12795, 12811, 12813, 12814, 12841
3324
3325 * The RPC implementation in libc is obsoleted. Old programs keep working
3326 but new programs cannot be linked with the routines in libc anymore.
3327 Programs in need of RPC functionality must be linked against TI-RPC.
3328 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3329
3330 Visible changes of this change include (obviously) the inability to link
3331 programs using RPC functions without referencing the TI-RPC library and the
3332 removal of the RPC headers from the glibc headers.
3333 Implemented by Ulrich Drepper.
3334
3335 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3336 syncfs, setns, sendmmsg
3337
3338 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3339 yue_HK, lij_IT, mhr_RU
3340
3341 * New encodings: CP770, CP771, CP772, CP773, CP774
3342
3343 * New program sotruss to trace calls through PLTs
3344 Implemented by Ulrich Drepper.
3345
3346 * The malloc hook implementation is marked deprecated and will be removed
3347 from the default implementation in the next version. The design never
3348 worked ever since the introduction of threads. Even programs which do
3349 not create threads themselves can use multiple threads created internally.
3350 \f
3351 Version 2.13
3352
3353 * The following bugs are resolved with this release:
3354
3355 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3356 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3357 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3358 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3359 12378, 12394, 12397
3360
3361 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3362
3363 * POWER7 optimizations: memset, memcmp, strncmp
3364
3365 * New optimized string functions for x86-64: strnlen (SSE2),
3366 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3367 Implemented by Ulrich Drepper.
3368 \f
3369 Version 2.12
3370
3371 * The following bugs are resolved with this release:
3372
3373 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3374 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3375 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3376 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3377 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3378 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3379 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3380 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3381
3382 * New interfaces: pthread_getname_np, pthread_setname_np
3383
3384 * New Linux interface: recvmmsg
3385
3386 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3387
3388 * The dynamic linker now recognizes supported ABI versions from the
3389 EI_ABIVERSION field in the ELF header.
3390 Implemented by Ulrich Drepper.
3391
3392 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3393 will not be used to fill in password fields in the passwd.byname replies.
3394 Instead it is used to synthesize the shadow.byname table, should it be
3395 missing. This is a useful mode in some installations involving Solaris.
3396 Implemented by Ulrich Drepper.
3397
3398 * New locales: kok_IN, sq_MK, cv_RU
3399 \f
3400 Version 2.11
3401
3402 * The following bugs are resolved with this release:
3403
3404 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3405 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3406 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3407 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3408 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3409 10789, 10817, 10823, 10840, 10847
3410
3411 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3412 mkostemps64
3413 Implemented by Ulrich Drepper.
3414
3415 * Checking version of longjmp added that fails if an uninitialized stack
3416 frame would be created. Implemented by Ulrich Drepper.
3417
3418 * STT_GNU_IFUNC is now supported in static executables.
3419 Implemented by H.J. Lu.
3420
3421 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3422
3423 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3424 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3425 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3426 Contributed by H.J. Lu.
3427
3428 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3429 Implemented by Ulrich Drepper.
3430
3431 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3432 strstr, strcasestr.
3433 Contributed by H.J. Lu.
3434
3435 * Support for fma instruction in AVX on x86-64.
3436 Implemented by H.J. Lu and Ulrich Drepper.
3437
3438 * AVX support in x86-64 auditing support in ld.so.
3439 Implemented by H.J. Lu.
3440
3441 * STB_GNU_UNIQUE support added.
3442 Implemented by Ulrich Drepper.
3443
3444 * Implement second fallback mode for optimized DNS lookup for even more
3445 broken environments. If two requests from the same port are not
3446 handled correctly close the socket and open a new one before sending
3447 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3448 can be used to select this mode right away, instead of rediscovering the
3449 necessity is every process again.
3450 Implemented by Ulrich Drepper.
3451
3452 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3453 Implemented by Adam Tkac.
3454
3455 * Optimized iconv conversions for S390x.
3456 Implemented by Andreas Krebbel.
3457
3458 * Using condvars with PI mutexes is now more efficient due to kernel
3459 support for requeueing to PI futexes. NPTL support added for x86-64.
3460 Implemented by Ulrich Drepper.
3461
3462 * New locales: ps_AF, my_MM
3463 \f
3464 Version 2.10
3465
3466 * The following bugs are resolved with this release:
3467
3468 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3469 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3470 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3471 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3472 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3473
3474 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3475 Implemented by Ulrich Drepper.
3476
3477 * New Linux interfaces: accept4, fallocate, fallocate64.
3478 Implemented by Ulrich Drepper.
3479
3480 * Correct declarations of string function when used in C++ code. This
3481 could lead to compile errors for invalid C++ code.
3482
3483 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3484 now in POSIX.
3485
3486 * New POSIX 2008 interface: psiginfo
3487 Implemented by Ulrich Drepper.
3488
3489 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3490 Implemented by Ulrich Drepper.
3491
3492 * Support for selecting between multiple function definitions at runtime
3493 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3494
3495 * The libcrypt library can now use the hash function implementations in
3496 NSS. Implemented by Ulrich Drepper.
3497
3498 * The malloc implementation can be compiled to be less memory efficient
3499 but higher performing in multi-threaded programs.
3500 Implemented by Ulrich Drepper.
3501
3502 * New locales: nan_TW@latin, ks_IN
3503
3504 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3505 Implemented by Ulrich Drepper.
3506
3507 * Extended printf hook support. It is possible to use user-defined types
3508 and extend existing format specifiers.
3509 Implemented by Ulrich Drepper.
3510
3511 * Handling for group shadow files has been added.
3512 Implemented by Ulrich Drepper.
3513
3514 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3515 servers (the case, e.g., for some people using the built-in DNS
3516 server in ADSL modems/routers). There is a once-per-process timeout
3517 in case of a broken server. To avoid it, users can run nscd or put
3518 'options single-request' in /etc/resolv.conf.
3519 Implemented by Ulrich Drepper.
3520 \f
3521 Version 2.9
3522
3523 * The following bugs are resolved with this release:
3524
3525 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3526 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3527 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3528 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3529 6968, 6974, 6980, 6995, 7008, 7009, 7029
3530
3531 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3532 up at the same time. Implemented by Ulrich Drepper.
3533
3534 * TLS descriptors for LD and GD on x86 and x86-64.
3535 Implemented by Alexandre Oliva.
3536
3537 * getaddrinfo now handles DCCP and UDPlite.
3538 Implemented by Ulrich Drepper.
3539
3540 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3541 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3542 Implemented by Ulrich Drepper.
3543
3544 * New implementation of memmem, strstr, and strcasestr which is O(n).
3545 Implemented by Eric Blake.
3546
3547 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3548
3549 * Implement "e" option for popen to open file descriptor with the
3550 close-on-exec flag set. Implemented by Ulrich Drepper.
3551
3552 * Implement "b" mode for fmemopen. In this mode writes writes don't
3553 implicitly add a NUL byte and seeks from the end of the buffer really
3554 use the buffer end, not the string length as the basis.
3555 Implemented by Ulrich Drepper.
3556
3557 * Many functions, exported and internal, now atomically set the close-on-exec
3558 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3559
3560 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3561 Sinhala)
3562 Implemented by Pravin Satpute.
3563
3564 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3565 \f
3566 Version 2.8
3567
3568 * The following bugs are resolved with this release:
3569
3570 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3571 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3572 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3573 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3574 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3575 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3576 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3577
3578 * New locales: bo_CN, bo_IN, shs_CA.
3579
3580 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3581
3582 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3583 Implemented by Pravin Satpute.
3584
3585 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3586
3587 * nscd caches now all timeouts for DNS entries
3588 Implemented by Ulrich Drepper.
3589
3590 * nscd is more efficient and wakes up less often.
3591 Implemented by Ulrich Drepper.
3592
3593 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3594 vdprintf, and obstack_vprintf.
3595 Implemented by Jakub Jelinek.
3596
3597 * Faster memset for x86-64.
3598 Implemented by Harsha Jagasia and H.J. Lu.
3599
3600 * Faster memcpy on x86.
3601 Implemented by Ulrich Drepper.
3602
3603 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3604 Implemented by Ulrich Drepper.
3605
3606 * Faster sqrt and sqrtf implemention for some PPC variants.
3607 Implemented by Steven Munroe.
3608 \f
3609 Version 2.7
3610
3611 * The following bugs are resolved with this release:
3612
3613 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3614 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3615 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3616 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3617 5071, 5103, 5104, 5112, 5113, 5184, 5186
3618
3619 * More checking functions: fread, fread_unlocked, open*, mq_open.
3620 Implemented by Jakub Jelinek and Ulrich Drepper.
3621
3622 * Extend fortification to C++. Implemented by Jakub Jelinek.
3623
3624 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3625 by not recognizing 'a' as a modifier when those specs are requested.
3626 Implemented by Jakub Jelinek.
3627
3628 * PPC optimizations to math and string functions.
3629 Implemented by Steven Munroe.
3630
3631 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3632 options to be passed. Implemented by Ulrich Drepper.
3633
3634 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3635
3636 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3637 Implemented by Ulrich Drepper.
3638
3639 * Handle private futexes in the NPTL implementation.
3640 Implemented by Jakub Jelinek and Ulrich Drepper.
3641
3642 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3643 Implemented by Roland McGrath and Ulrich Drepper.
3644
3645 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3646
3647 * SHA-256 and SHA-512 based password encryption.
3648 Implemented by Ulrich Drepper.
3649
3650 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3651 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3652 yo_NG.
3653
3654 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3655 Implemented by Ulrich Drepper.
3656 \f
3657 Version 2.6
3658
3659 * The following bugs are resolved with this release:
3660
3661 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3662 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3663 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3664 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3665 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3666 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3667 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3668 4702, 4858
3669
3670 * New Linux interfaces: epoll_pwait, sched_getcpu.
3671
3672 * New generic interfaces: strerror_l.
3673
3674 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3675 \f
3676 Version 2.5
3677
3678 * The following bugs are resolved with this release:
3679
3680 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3681 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3682 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3683 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3684 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3685 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3686 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3687 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3688 3225, 3251, 3252, 3253, 3273
3689
3690 * For Linux, the sorting of addresses returned by getaddrinfo now also
3691 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3692 Implemented by Ulrich Drepper.
3693
3694 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3695 Implemented by Ulrich Drepper.
3696
3697 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3698
3699 * New iconv module for MIK. Contributed by Alexander Shopov.
3700
3701 * For sites with broken group and/or passwd database, the auto-propagate
3702 option of nscd can prevent creating ID lookup entries from the results
3703 of a name lookup and vice versa. This usually is no problem but some
3704 site might have problems with the default behavior.
3705 Implemented by Ulrich Drepper.
3706
3707 * Iterating over entire database in NIS can be slow. With the
3708 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3709 to trade time for memory. The entire database will be read at once.
3710 Implemented by Ulrich Drepper.
3711
3712 * The interfaces introduced in RFC 3542 have been implemented by
3713 Ulrich Drepper.
3714
3715 * Support for the new ELF hash table format was added by Ulrich Drepper.
3716
3717 * Support for priority inheritance mutexes added by Jakub Jelinek and
3718 Ulrich Drepper.
3719
3720 * Support for priority protected mutexes added by Jakub Jelinek.
3721
3722 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3723 or_IN, csb_PL, fy_NL, sr_ME.
3724 \f
3725 Version 2.4
3726
3727 * More overflow detection functions.
3728
3729 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3730 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3731
3732 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3733 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3734 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3735 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3736 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3737 by Masahide Washizawa.
3738
3739 * It is now possible to install audit libraries for the dynamic linker using
3740 LD_AUDIT. Implemented by Ulrich Drepper.
3741
3742 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3743 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3744 For a libc and libpthread that works well on Linux 2.4 kernels, we
3745 recommend using the stable 2.3 branch.
3746
3747 * The new function open_wmemstream acts like open_memstream,
3748 but uses a wchar_t wide character buffer.
3749
3750 * The new function ppoll is an improved version of poll, similar to pselect.
3751
3752 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3753 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3754 renameat, unlinkat, linkat, symlinkat, readlinkat.
3755
3756 * New Linux kernel system calls: unshare,
3757 inotify_init, inotify_add_watch, inotify_rm_watch.
3758
3759 * The euidaccess function is now also known by the alias eaccess,
3760 for compatibility with some other systems.
3761
3762 * Timezone data updated to 2006b version.
3763 \f
3764 Version 2.3.6
3765
3766 * The following bugs are resolved with this release:
3767
3768 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3769 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3770 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3771 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3772 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3773 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3774
3775 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3776
3777 * As of this release, GCC 4 can be used to compile the C Library.
3778
3779 * Timezone data updated to 2005m version.
3780 \f
3781 Version 2.3.5
3782
3783 * The following bugs are resolved with this release:
3784
3785 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3786 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3787 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3788 777, 787, 821, 822, 823, 825
3789
3790 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3791 \f
3792 Version 2.3.4
3793
3794 * Support for RFC 3678. Real implementations exist only for Linux so far.
3795 Implemented by Ulrich Drepper.
3796
3797 * nscd can now cache entries persistently. Expiring entries are reloaded.
3798 For speedups the cache can be shared in memory with client processes.
3799 Implemented by Ulrich Drepper.
3800
3801 * nscd can now perform SELinux checks.
3802 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3803
3804 * getaddrinfo queries are now cached. Canonical name lookup is performed
3805 efficiently.
3806 Implemented by Ulrich Drepper.
3807
3808 * The nothrow function attribute is used when headers are used by gcc when
3809 compiling C code. This can avoid significant amounts of exception
3810 handling data.
3811
3812 * The malloc functions perform more error checking and are stricter when
3813 it comes to reacting on errors. The default action is to terminate
3814 the process after showing an error message. Implemented by Ulrich Drepper.
3815
3816 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3817 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3818
3819 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3820 namespaces using the new function dlmopen(). This feature is of course,
3821 like most other dynamic loading functionality, not available in statically
3822 linked applications. Implemented by Ulrich Drepper.
3823
3824 * Low-overhead boundary checking variants of string and some stdio functions
3825 were added. These are to be used in conjunction with a gcc patch by
3826 Jakub Jelinek which adds calls to these functions if possible.
3827 Implemented by Jakub Jelinek and Ulrich Drepper.
3828
3829 * Old code for several operating systems and machine architectures that
3830 have not been in working condition in a long time have been removed from
3831 the main source tree maintained by the GNU C Library's maintainers.
3832 These files are now reside in the separate `ports' source module
3833 that is usable as an add-on when building the library.
3834 \f
3835 Version 2.3.3
3836
3837 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3838 interrogate the dynamic linker, compatible with the Solaris interface.
3839
3840 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3841 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3842
3843 * getifaddrs now uses the netlink interface on Linux to get its information.
3844 Implemented by Thorsten Kukuk.
3845
3846 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3847 Implemented by Ulrich Drepper.
3848
3849 * support for non-executable stacks on x86 has been added. Changes mostly
3850 by Roland McGrath.
3851
3852 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3853 and Ulrich Drepper.
3854
3855 * getaddrinfo now performs destination address selection according to
3856 RFC 3484.
3857 \f
3858 Version 2.3.2
3859
3860 * Thread-safe interfaces for many functions that access locale data
3861 were added in version 2.3, but these features were omitted from NEWS.
3862 Many functions have variants with an `_l' suffix that take a `locale_t'
3863 object as a parameter rather than consulting the current locale.
3864 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3865 create and maintain `locale_t' objects. Additionally, the new function
3866 `uselocale' sets "the current locale" (as used by functions not so
3867 parameterized) set for an individual thread. These features were added
3868 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3869
3870 * The functions getresuid, getresgid, setresuid, and setresgid, which
3871 have long been available on Linux, are now declared in <unistd.h>
3872 and are now also available on the Hurd.
3873
3874 * ELF thread-local storage support (TLS) now works on x86-64.
3875
3876 * The new dynamic string token $LIB is expanded in shared library names.
3877 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3878
3879 * Aldy Hernandez contributed complete software floating point support for
3880 PowerPC machines with no FPU.
3881
3882 * fexecve is implemented on Linux.
3883
3884 * The `btowc' function should work at least twice as fast due to
3885 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3886
3887 * With approriate thread add-ons cancelable functions are now implemented
3888 in libc.so as well. No need to call the function in libpthread. This
3889 change allowed to finally disable the incorrect and expensive handling
3890 of weak definition in ld.so.
3891
3892 * Yet more PLT entries in libc.so have been removed. We finally arrived
3893 at the bare minimum. Startup times improved appropriately.
3894
3895 * Support for the new Linux/x86 system call interface was added. The
3896 AT_SYSINFO auxiliary vector entry is recognized and handled.
3897 \f
3898 Version 2.3
3899
3900 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3901 charsets.
3902
3903 * iconv (the program and the interface) now accepts empty names (excluding
3904 options like //TRANSLIT) to mean "use charset of current locale".
3905
3906 * localedef can now transliterate characters in strings which are not in
3907 the provided charmap. The information from the input locale is used.
3908
3909 * Prelinking support was added for ELF targets. This requires additional
3910 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3911
3912 * Read-only stdio streams now use mmap to speed up operation by eliminating
3913 copying and buffer underflows. To use add 'm' to the mode string of
3914 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3915
3916 * The malloc functions were completely rewritten by Wolfram Gloger based
3917 on Doug Lea's malloc-2.7.0.c.
3918
3919 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3920 implementation of regex.
3921
3922 * Bruno Haible upgraded the iconv and locale implementation to support
3923 Unicode 3.2.
3924
3925 * Contents of the LC_* and LANG environment variables in the CEN style are
3926 not recognized anymore. It never was used. Change by Ulrich Drepper.
3927
3928 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3929 thread-local storage (TLS) ABI on some platforms.
3930 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3931
3932 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3933 EUC-JISX0213, and TSCII.
3934
3935 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3936 BSD-compatible interface for getting all network interface addresses.
3937 Implementation for IPv4 by Roland McGrath.
3938
3939 * Loading of locale data is faster due to the introduction of a locale
3940 archive. Implemented by Roland McGrath and Ulrich Drepper.
3941
3942 * Startup times are significantly reduced by not using exported functions
3943 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3944 and Ulrich Drepper.
3945
3946 * Steven Munroe contributed a port to PowerPC64/Linux.
3947 \f
3948 Version 2.2.6
3949
3950 * The Hurd now uses the GNU libio implementation of stdio rather than the
3951 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3952
3953 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3954 with the same behavior as the Linux system call of the same name.
3955 \f
3956 Version 2.2.5
3957
3958 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3959 128-bit long double format.
3960
3961 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3962 IBM1161, and IBM1162 charsets.
3963
3964 * Andreas Jaeger contributed a port to x86-64/Linux.
3965
3966 * Peter Bruin contributed a port to PowerPC/Hurd.
3967
3968 * libc functions using I/O streams now can handle wide-oriented streams
3969 as well.
3970
3971 * optimizations in the dynamic linker. Binaries created by recent binutils
3972 versions start up quicker due to reduced time spent on relocations.
3973
3974 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3975 \f
3976 Version 2.2.4
3977
3978 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3979 asinh, atanh, j0 for the 128-bit long double format.
3980
3981 * Bruno Haible updated all the code handling Unicode in some form to
3982 support Unicode 3.1.
3983
3984 * Speed of regex for single-byte locales is back to previous levels.
3985 Patch by Isamu Hasegawa.
3986
3987 * Alpha, SPARC, and IA-64 now also using floating stacks.
3988
3989 * Startup time of internationalized applications greatly improved through
3990 iconv cache. Use iconvconfig to generate the cache file.
3991 Contributed by Ulrich Drepper.
3992
3993 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3994 severe performance problems. Patch by David Mosberger.
3995
3996 * The Hurd port got a lot more functionality like AIO, various stdio
3997 extensions, etc. Mainly done by Roland McGrath.
3998
3999 * mtrace can now lookup symbols in shared libraries.
4000 \f
4001 Version 2.2.3
4002
4003 * Intel's IA-64 math library is largely integrated. It provides fast and
4004 accurate implementatations for most basic and standard math functions
4005 in float, double, and long double format.
4006
4007 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4008 and asin for the 96-bit long double format and asin, log, tan for the
4009 128-bit long double format.
4010
4011 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4012 The basic double functions exist today. Contributed by Abraham Ziv
4013 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4014 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4015
4016 * An asynchronous name lookup library was added. The interface is designed
4017 after POSIX AIO. The proposal was circulated beforehand to get comments.
4018 No negative ones came in. Implemented by Ulrich Drepper.
4019
4020 * Port to S390/64bit contributed by Martin Schwidefsky
4021 <schwidefsky@de.ibm.com>.
4022
4023 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4024 of functions for Linux/IA-64.
4025
4026 * The RPC code is now thread safe. Threads can now use the same service
4027 of different services at the same time. Patch by Eric Norum
4028 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4029
4030 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4031 family of functions for Linux/S390.
4032
4033 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4034 of functions for Linux/x86.
4035
4036 * Port to Linux/CRIS contributed by Axis Communications.
4037 \f
4038 Version 2.2.2
4039
4040 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4041 we can now check for namespace violations and missing declarations. The
4042 result is that almost all headers are now Unix-compliant (as defined in
4043 the upcoming XPG6). The negative side is that some programs might need
4044 corrections, too, if they depend on the incorrect form of the headers in
4045 previous versions which defined too many symbols and included too many
4046 other headers.
4047
4048 * regex now handles multibyte character sets correctly.
4049 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4050
4051 * iconv (the program) does now conform to the upcoming XPG6 and handles
4052 charmaps. Instead of the charset names the path of charmaps can be
4053 provided and the conversion happens based on this data.
4054 Contributed by Ulrich Drepper.
4055
4056 * The locale program now provides detailed information about the installed
4057 locales. While
4058
4059 locale -a
4060
4061 only lists the names of the supported locales
4062
4063 locale -a --verbose
4064
4065 provides details such as country, language, and codeset name.
4066 Contributed by Ulrich Drepper.
4067 \f
4068 Version 2.2.1
4069
4070 * The gencat program now parses the input file according to the charset
4071 selected by the LC_CTYPE category. This is important for stateful
4072 character sets. To make generating catalogs easier there is a way
4073 to overwrite the charset selected by the locale: before the first
4074 message or $ quote line the catalog can contain a line like
4075
4076 $ codeset=ISO-8859-2
4077
4078 to select the charset (ISO-8859-2 in this case).
4079
4080 Implemented by Shinya Hanataka and Ulrich Drepper.
4081
4082 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4083 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4084 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4085
4086 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4087 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4088
4089 * The LANGUAGE environment variable is now ignored unless the locale is
4090 changed from the default "C" locale.
4091
4092 * The usual bug fixes.
4093 \f
4094 Version 2.2
4095
4096 * Greg McGary added runtime support for bounds checking using gcc's
4097 new -fbounded-pointers option. ix86 target is complete. PowerPC
4098 is in progress.
4099
4100 * Thorsten Kukuk added secure mode support to nscd.
4101
4102 * The Berkeley db libraries have been removed.
4103
4104 Related, the nss_db module is now in a separate package since it
4105 obviously requires a database library being available.
4106
4107 * Wide character I/O streams implemented by Ulrich Drepper.
4108
4109 * Functions from the extended socket API added by Ulrich Drepper.
4110
4111 * Functions feenableexcept and fedisableexcept to control the
4112 behaviour of individual exceptions have been added by Andreas Jaeger.
4113
4114 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4115
4116 * The resolver code has been updated from bind 8.2.3-T5B which supports
4117 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4118 and Mark Kettenis.
4119
4120 This change could in some situations effect backward compatibility. Since
4121 now `_res' is a thread-local instead of a global variable, modifying it
4122 in one thread does not have any effect in other threads.
4123
4124 The resolver library was also extended to allow IPv6 as the transport
4125 protocol for the requests. This work was done by Stig Venaas.
4126
4127 * Compatibility code for K&R C compilers has been removed from the
4128 header files. A ISO C compiler is needed to use the library
4129 (conforming to either C89 or C99 standard).
4130
4131 * Complete rewrite of the localedef program to support multibyte character
4132 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4133 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4134 Implemented by Ulrich Drepper.
4135
4136 Bruno Haible significantly improved the generation and use of the data
4137 structures for the wide character tables.
4138
4139 * Plural handling in gettext implemented by Ulrich Drepper.
4140
4141 * The utmp daemon has been removed.
4142
4143 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4144
4145 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4146 and Yutaka Niibe.
4147
4148 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4149
4150 * POSIX barriers implemented by Kaz Kylheku.
4151
4152 * POSIX spawn function family implemented by Ulrich Drepper.
4153
4154 * POSIX spinlocks are now available.
4155
4156 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4157
4158 * the configure option --enable-kernel=X.Y.Z allows to strip out
4159 compatibility for kernel versions before X.Y.Z. This is currently only
4160 implemented for Linux.
4161
4162 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4163 field sin6_scope_id. This means that all programs using IPv6 should be
4164 recompiled. Don't expect binary compatibility with previous glibc
4165 versions.
4166
4167 * various conversion modules for IBM character sets contributed by
4168 Masahide Washizawa.
4169
4170 * IA-64 port by Jes Sorensen and HJ Lu.
4171 \f
4172 Version 2.1.3
4173
4174 * bug fixes
4175
4176 \f
4177 Version 2.1.2
4178
4179 * bug fixes
4180
4181 \f
4182 Version 2.1.1
4183
4184 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4185
4186 * New xdr functions are added; some rpc functions are now 64bit clean.
4187
4188 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4189
4190 * Fixed known incompatibilities with glibc 2.0.
4191
4192 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4193
4194 * Optimized a number of functions (especially the ELF dynamic loader).
4195
4196 * Update timezone data files.
4197
4198 * lots of charmaps corrections
4199
4200 * some new locale definitions and charmaps
4201
4202 \f
4203 Version 2.1
4204
4205 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4206 conform to POSIX member type requirements. Symbol versions have been
4207 adjusted as needed within the library, and for direct use by applications,
4208 but there is potential for problems if third-party libraries use
4209 struct timeval as part of their interface. This does not present
4210 a problem for X and other "essential" system libraries.
4211
4212 * An additional locale model to support C++ Standard Library locale
4213 model and probably more was implemented by Ulrich Drepper.
4214
4215 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4216 symbol level.
4217
4218 * Miles Bader provided the `argp' function family to support hierachical
4219 command line argument parsing, layered on top of getopt.
4220
4221 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4222
4223 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4224 numbers.
4225
4226 * scanf recognizes the %a and %A format for scanning floating point numbers.
4227
4228 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4229 information and interfaces for the available integer types.
4230
4231 * about 130 new math functions were added to implement the ISO C9x math
4232 library.
4233
4234 * the new header <complex.h> contains definitions of the complex math
4235 functions from ISO C 9X.
4236
4237 * the new header <tgmath.h> defines generic macros to use complex or
4238 real valued functions.
4239
4240 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4241
4242 * Andreas Jaeger provided a test suite for the math library.
4243
4244 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4245
4246 * Ulrich Drepper added character set conversion functions (iconv).
4247
4248 * Optimized string functions have been added.
4249
4250 * The localedata addon is now part of glibc.
4251
4252 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4253
4254 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4255 daemon for NSS (nscd).
4256
4257 Missing a better place here are some numbers on improvements. Under
4258 Linux 2.1.125 un-tar-ing the kernel sources takes
4259
4260 user system wall
4261
4262 using local files 12.19s 6.88s 22.91s
4263
4264 using NIS 13.92s 8.91s 26.34s
4265
4266 using NIS & nscd 10.37s 7.34s 25.30s
4267
4268 using NIS+ 27.57s 30.37s 640.46s
4269
4270 using NIS+ & nscd 10.25s 7.83s 26.51s
4271
4272 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4273
4274 Keep in mind that non-namelookup related operations dominate above times.
4275 It was just a common complain that using NIS+ unpacking the kernel is
4276 horribly slow.
4277
4278 [1] The old nscd implementation is not available anymore since it was
4279 distributed with glibc up to version 2.0.98 and thus is now replaced.
4280
4281 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4282
4283 * Mark Kettenis provided a Hesiod NSS module.
4284
4285 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4286 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4287
4288 * rcmd can now handle netgroups (Dick Streefland).
4289
4290 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4291 Bambrough.
4292
4293 * Support for the IPv6 protocol has been added to the socket API, as per the
4294 latest draft standards.
4295
4296 * Support for Linux 2.2 has been added.
4297
4298 * Interface changes relative to the latest 2.0.x release:
4299 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4300 addseverity NEW: Unix98
4301 alphasort64 NEW: LFS
4302 argp_err_exit_status NEW: argp, GNU ext
4303 argp_error NEW: argp, GNU ext
4304 argp_failure NEW: argp, GNU ext
4305 argp_help NEW: argp, GNU ext
4306 argp_parse NEW: argp, GNU ext
4307 argp_program_bug_address NEW: argp, GNU ext
4308 argp_program_version NEW: argp, GNU ext
4309 argp_program_version_hook NEW: argp, GNU ext
4310 argp_state_help NEW: argp, GNU ext
4311 argp_usage NEW: argp, GNU ext
4312 authdes_create NEW: Secure RPC
4313 authdes_getucred NEW: Secure RPC
4314 authdes_pk_create NEW: Secure RPC
4315 backtrace NEW: GNU ext.
4316 backtrace_symbols NEW: GNU ext.
4317 backtrace_symbols_fd NEW: GNU ext.
4318 cacos NEW: ISO C 9x
4319 cacosf NEW: ISO C 9x
4320 cacosh NEW: ISO C 9x
4321 cacoshf NEW: ISO C 9x
4322 cacoshl NEW: ISO C 9x
4323 cacosl NEW: ISO C 9x
4324 capget NEW: kernel
4325 capset NEW: kernel
4326 carg NEW: ISO C 9x
4327 cargf NEW: ISO C 9x
4328 cargl NEW: ISO C 9x
4329 casin NEW: ISO C 9x
4330 casinf NEW: ISO C 9x
4331 casinh NEW: ISO C 9x
4332 casinhf NEW: ISO C 9x
4333 casinhl NEW: ISO C 9x
4334 casinl NEW: ISO C 9x
4335 catan NEW: ISO C 9x
4336 catanf NEW: ISO C 9x
4337 catanh NEW: ISO C 9x
4338 catanhf NEW: ISO C 9x
4339 catanhl NEW: ISO C 9x
4340 catanl NEW: ISO C 9x
4341 cbc_crypt NEW: Secure RPC
4342 ccos NEW: ISO C 9x
4343 ccosf NEW: ISO C 9x
4344 ccosh NEW: ISO C 9x
4345 ccoshf NEW: ISO C 9x
4346 ccoshl NEW: ISO C 9x
4347 ccosl NEW: ISO C 9x
4348 cexp NEW: ISO C 9x
4349 cexpf NEW: ISO C 9x
4350 cexpl NEW: ISO C 9x
4351 cimag NEW: ISO C 9x
4352 cimagf NEW: ISO C 9x
4353 cimagl NEW: ISO C 9x
4354 clearerr_locked REMOVED
4355 clntunix_create NEW: sunrpc ext
4356 clog NEW: ISO C 9x
4357 clog10 NEW: ISO C 9x
4358 clog10f NEW: ISO C 9x
4359 clog10l NEW: ISO C 9x
4360 clogf NEW: ISO C 9x
4361 clogl NEW: ISO C 9x
4362 conj NEW: ISO C 9x
4363 conjf NEW: ISO C 9x
4364 conjl NEW: ISO C 9x
4365 cpow NEW: ISO C 9x
4366 cpowf NEW: ISO C 9x
4367 cpowl NEW: ISO C 9x
4368 cproj NEW: ISO C 9x
4369 cprojf NEW: ISO C 9x
4370 cprojl NEW: ISO C 9x
4371 creal NEW: ISO C 9x
4372 crealf NEW: ISO C 9x
4373 creall NEW: ISO C 9x
4374 creat64 NEW: LFS
4375 csin NEW: ISO C 9x
4376 csinf NEW: ISO C 9x
4377 csinh NEW: ISO C 9x
4378 csinhf NEW: ISO C 9x
4379 csinhl NEW: ISO C 9x
4380 csinl NEW: ISO C 9x
4381 csqrt NEW: ISO C 9x
4382 csqrtf NEW: ISO C 9x
4383 csqrtl NEW: ISO C 9x
4384 ctan NEW: ISO C 9x
4385 ctanf NEW: ISO C 9x
4386 ctanh NEW: ISO C 9x
4387 ctanhf NEW: ISO C 9x
4388 ctanhl NEW: ISO C 9x
4389 ctanl NEW: ISO C 9x
4390 des_setparity NEW: Secure RPC
4391 ecb_crypt NEW: Secure RPC
4392 endutxent NEW: Unix98
4393 exp10 NEW: ISO C 9x
4394 exp10f NEW: ISO C 9x
4395 exp10l NEW: ISO C 9x
4396 exp2 NEW: ISO C 9x
4397 exp2f NEW: ISO C 9x
4398 exp2l NEW: ISO C 9x
4399 fattach NEW: STREAMS
4400 fdetach NEW: STREAMS
4401 fdim NEW: ISO C 9x
4402 fdimf NEW: ISO C 9x
4403 fdiml NEW: ISO C 9x
4404 feclearexcept NEW: ISO C 9x
4405 fegetenv NEW: ISO C 9x
4406 fegetexceptflag NEW: ISO C 9x
4407 fegetround NEW: ISO C 9x
4408 feholdexcept NEW: ISO C 9x
4409 feof_locked REMOVED
4410 feraiseexcept NEW: ISO C 9x
4411 ferror_locked REMOVED
4412 fesetenv NEW: ISO C 9x
4413 fesetexceptflag NEW: ISO C 9x
4414 fesetround NEW: ISO C 9x
4415 fetestexcept NEW: ISO C 9x
4416 feupdateenv NEW: ISO C 9x
4417 fflush_locked REMOVED
4418 ffsl NEW: GNU ext.
4419 ffsll NEW: GNU ext.
4420 fgetpos64 NEW: LFS
4421 fgets_unlocked NEW: GNU ext.
4422 fileno_locked REMOVED
4423 fma NEW: ISO C 9x
4424 fmaf NEW: ISO C 9x
4425 fmal NEW: ISO C 9x
4426 fmax NEW: ISO C 9x
4427 fmaxf NEW: ISO C 9x
4428 fmaxl NEW: ISO C 9x
4429 fmin NEW: ISO C 9x
4430 fminf NEW: ISO C 9x
4431 fminl NEW: ISO C 9x
4432 fmtmsg NEW: Unix98
4433 fopen64 NEW: LFS
4434 fputc_locked REMOVED
4435 fputs_unlocked NEW: GNU ext.
4436 fread_unlocked NEW: GNU ext.
4437 freopen64 NEW: LFS
4438 fseeko NEW: Unix98
4439 fsetpos64 NEW: LFS
4440 fstatfs64 NEW: LFS
4441 fstatvfs NEW: Unix98
4442 fstatvfs64 NEW: LFS
4443 ftello NEW: Unix98
4444 ftello64 NEW: LFS
4445 ftruncate64 NEW: LFS
4446 ftw64 NEW: LFS
4447 fwrite_unlocked NEW: GNU ext.
4448 gai_strerror NEW: IPv6
4449 gamma_r REMOVED
4450 gammaf_r REMOVED
4451 gammal_r REMOVED
4452 getchar_locked REMOVED
4453 getdate NEW: Unix98
4454 getdate_err NEW: Unix98
4455 getdate_r NEW: GNU ext.
4456 getmsg NEW: STREAMS
4457 getnameinfo NEW: IPv6
4458 getnetname NEW: Secure RPC
4459 getpmsg NEW: STREAMS
4460 getpt NEW: Unix98 PTY
4461 getrlimit64 NEW: LFS
4462 getutxent NEW: Unix98
4463 getutxid NEW: Unix98
4464 getutxline NEW: Unix98
4465 glob64 NEW: GNU ext.
4466 globfree64 NEW: GNU ext.
4467 gnu_get_libc_release NEW: GNU ext.
4468 gnu_get_libc_version NEW: GNU ext.
4469 grantpt NEW: Unix98 PTY
4470 host2netname NEW: Secure RPC
4471 iconv NEW: iconv
4472 iconv_close NEW: iconv
4473 iconv_open NEW: iconv
4474 if_freenameindex NEW: IPv6
4475 if_indextoname NEW: IPv6
4476 if_nameindex NEW: IPv6
4477 if_nametoindex NEW: IPv6
4478 in6addr_any NEW: IPv6
4479 in6addr_loopback NEW: IPv6
4480 inet6_isipv4mapped NEW: IPv6
4481 isastream NEW: STREAMS
4482 iswblank NEW: GNU ext.
4483 key_decryptsession NEW: Secure RPC
4484 key_decryptsession_pk NEW: Secure RPC
4485 key_encryptsession NEW: Secure RPC
4486 key_encryptsession_pk NEW: Secure RPC
4487 key_gendes NEW: Secure RPC
4488 key_get_conv NEW: Secure RPC
4489 key_secretkey_is_set NEW: Secure RPC
4490 key_setnet NEW: Secure RPC
4491 key_setsecret NEW: Secure RPC
4492 llrint NEW: ISO C 9x
4493 llrintf NEW: ISO C 9x
4494 llrintl NEW: ISO C 9x
4495 llround NEW: ISO C 9x
4496 llroundf NEW: ISO C 9x
4497 llroundl NEW: ISO C 9x
4498 log2 NEW: ISO C 9x
4499 log2f NEW: ISO C 9x
4500 log2l NEW: ISO C 9x
4501 lrint NEW: ISO C 9x
4502 lrintf NEW: ISO C 9x
4503 lrintl NEW: ISO C 9x
4504 lround NEW: ISO C 9x
4505 lroundf NEW: ISO C 9x
4506 lroundl NEW: ISO C 9x
4507 lseek64 NEW: LFS
4508 makecontext NEW: Unix98
4509 mempcpy NEW: GNU ext.
4510 mmap64 NEW: LFS
4511 moncontrol REMOVED
4512 modify_ldt NEW: kernel
4513 nan NEW: ISO C 9x
4514 nanf NEW: ISO C 9x
4515 nanl NEW: ISO C 9x
4516 nearbyint NEW: ISO C 9x
4517 nearbyintf NEW: ISO C 9x
4518 nearbyintl NEW: ISO C 9x
4519 netname2host NEW: Secure RPC
4520 netname2user NEW: Secure RPC
4521 nexttoward NEW: ISO C 9x
4522 nexttowardf NEW: ISO C 9x
4523 nexttowardl NEW: ISO C 9x
4524 nftw NEW: Unix98
4525 nftw64 NEW: LFS
4526 open64 NEW: LFS
4527 passwd2des NEW: Secure RPC
4528 pow10 NEW: GNU ext.
4529 pow10f NEW: GNU ext.
4530 pow10l NEW: GNU ext.
4531 pread NEW: Unix98
4532 pread64 NEW: LFS
4533 printf_size NEW: GNU ext.
4534 printf_size_info NEW: GNU ext.
4535 profil_counter REMOVED
4536 pthread_mutexattr_getkind_np REPLACED
4537 pthread_mutexattr_setkind_np REPLACED
4538 ptsname NEW: Unix98 PTY
4539 ptsname_r NEW: Unix98 PTY
4540 putc_locked REMOVED
4541 putchar_locked REMOVED
4542 putgrent NEW: GNU ext.
4543 putmsg NEW: STREAMS
4544 putpmsg NEW: STREAMS
4545 pututxline NEW: Unix98
4546 pwrite NEW: Unix98
4547 pwrite64 NEW: LFS
4548 readdir64 NEW: LFS
4549 readdir64_r NEW: LFS
4550 remquo NEW: ISO C 9x
4551 remquof NEW: ISO C 9x
4552 remquol NEW: ISO C 9x
4553 round NEW: ISO C 9x
4554 roundf NEW: ISO C 9x
4555 roundl NEW: ISO C 9x
4556 rtime NEW: GNU ext.
4557 scalbln NEW: ISO C 9x
4558 scalblnf NEW: ISO C 9x
4559 scalblnl NEW: ISO C 9x
4560 scandir64 NEW: LFS
4561 sendfile NEW: kernel
4562 setcontext NEW: Unix98
4563 setrlimit64 NEW: LFS
4564 setutxent NEW: Unix98
4565 sighold NEW: Unix98
4566 sigignore NEW: Unix98
4567 sigqueue NEW: POSIX.1b
4568 sigrelse NEW: Unix98
4569 sigset NEW: POSIX.1b
4570 sigtimedwait NEW: POSIX.1b
4571 sigwaitinfo NEW: POSIX.1b
4572 sincos NEW: GNU ext.
4573 sincosf NEW: GNU ext.
4574 sincosl NEW: GNU ext.
4575 statfs64 NEW: LFS
4576 statvfs NEW: Unix98
4577 statvfs64 NEW: LFS
4578 strcasestr NEW: GNU ext.
4579 strtoimax NEW: ISO C 9x
4580 strtoumax NEW: ISO C 9x
4581 strverscmp NEW: GNU ext.
4582 svcauthdes_stats NEW: Secure RPC
4583 svcunix_create NEW: sunrpc etx
4584 svcunixfd_create NEW: sunrpc ext
4585 swapcontext NEW: Unix98
4586 tcgetsid NEW: Unix98 PTY
4587 tdestroy NEW: GNU ext.
4588 tgamma NEW: ISO C 9x
4589 tgammaf NEW: ISO C 9x
4590 tgammal NEW: ISO C 9x
4591 tmpfile64 NEW: LFS
4592 trunc NEW: ISO C 9x
4593 truncate64 NEW: LFS
4594 truncf NEW: ISO C 9x
4595 truncl NEW: ISO C 9x
4596 umount2 NEW: kernel
4597 unlockpt NEW: Unix98 PTY
4598 updwtmpx NEW: Unix98
4599 user2netname NEW: Secure RPC
4600 utmpxname NEW: Unix98
4601 versionsort NEW: GNU ext.
4602 versionsort64 NEW: GNU ext.
4603 waitid NEW: Unix98
4604 wcscasecmp NEW: GNU ext.
4605 wcsncasecmp NEW: GNU ext.
4606 wcsnlen NEW: GNU ext.
4607 wcstoimax NEW: ISO C 9x
4608 wcstoll NEW: ISO C 9x
4609 wcstoull NEW: ISO C 9x
4610 wcstoumax NEW: ISO C 9x
4611 wcswcs NEW: Unix98
4612 wordexp NEW: POSIX.2
4613 wordfree NEW: POSIX.2
4614 write_profiling REMOVED
4615 xdecrypt NEW: Secure RPC
4616 xdr_authdes_cred NEW: Secure RPC
4617 xdr_authdes_verf NEW: Secure RPC
4618 xdr_cryptkeyarg NEW: Secure RPC
4619 xdr_cryptkeyarg2 NEW: Secure RPC
4620 xdr_cryptkeyres NEW: Secure RPC
4621 xdr_getcredres NEW: Secure RPC
4622 xdr_key_netstarg NEW: Secure RPC
4623 xdr_key_netstres NEW: Secure RPC
4624 xdr_keybuf NEW: Secure RPC
4625 xdr_keystatus NEW: Secure RPC
4626 xdr_netnamestr NEW: Secure RPC
4627 xdr_sizeof NEW: Secure RPC
4628 xdr_unixcred NEW: sunrpc ext
4629 xencrypt NEW: Secure RPC
4630 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4631 \f
4632 Version 2.0.6
4633
4634 * more bug fixes
4635
4636 \f
4637 Version 2.0.5
4638
4639 * more bug fixes
4640
4641 * inet_ntoa is thread-safe
4642
4643 * updwtmp is moved from libutil to libc
4644
4645 * rewrite of cbrt function
4646
4647 * update of timezone data
4648 \f
4649 Version 2.0.4
4650
4651 * more bug fixes
4652 \f
4653 Version 2.0.3
4654
4655 * more bug fixes
4656 \f
4657 Version 2.0.2
4658
4659 * more bug fixes
4660
4661 * add atoll function
4662
4663 * fix complex problems in Berkeley DB code
4664
4665 * fix math functions
4666 \f
4667 Version 2.0.1
4668
4669 * fixed lots of header problems (especially Linux/GNU specific)
4670
4671 * dynamic loader preserves all registers
4672
4673 * Roland McGrath provided support for handling of auxiliary objects in
4674 the ELF dynamic loader.
4675
4676 * support for parallel builds is improved
4677 \f
4678 Version 2.0
4679
4680 * GNU extensions are no longer declared by default. To enable them you
4681 must define the macro `_GNU_SOURCE' in your program or compile with
4682 `-D_GNU_SOURCE'.
4683
4684 * The library has changed from using GNU ld symbol aliases to using weak
4685 symbols where available. The ELF object file format supports weak
4686 symbols; GNU ld also supports weak symbols in the a.out format. (There
4687 is also now support for other GNU ld extensions in ELF. Use the
4688 `--with-elf' option to configure to indicate you have ELF, and
4689 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4690 of many files which contained only symbol aliases, reducing the size of
4691 the source and the compiled library; many other files were renamed to
4692 less cryptic names previously occupied by the symbol alias files.
4693 There is a new header file <elf.h> for programs which operate on
4694 files in the ELF format.
4695
4696 * Converted to Autoconf version 2, so `configure' has more options.
4697 Run `configure --help' to see the details.
4698
4699 * The library can now be configured to build profiling, highly-optimized
4700 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4701 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4702 `configure' enable building these extra libraries. The shared library is
4703 built by default when using both ELF and GNU ld. When shared libraries
4704 are enabled, the new library `-ldl' is available for arbitrary run-time
4705 loading of shared objects; its interface is defined in <dlfcn.h>. The
4706 new header file <link.h> gives access to the internals of the run-time
4707 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4708 application of same name on other systems and it provides information
4709 about dynamically linked binaries.
4710
4711 * The C library now provides the run-time support code for profiling
4712 executables compiled with `-pg'. Programs can control the profiling code
4713 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4714 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4715 the support for this file format was contributed by David Mosberger-Tang.
4716
4717 * The math code has been replaced with a math library based on fdlibm from
4718 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4719 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4720 functions. The math functions now reside in a separate library, so
4721 programs using them will need to use `-lm' their linking commands.
4722
4723 * John C. Bowman contributed optimized ix87 assembler inline functions.
4724
4725 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4726 mechanism similar to that found in Solaris 2. This is now used for the
4727 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4728 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4729 file controls what services are used for each individual database. This
4730 works by loading shared libraries with names specified in `nsswitch.conf',
4731 so service modules can be changed or added at any time without even
4732 relinking any program. Currently there are the file, db, and NIS based
4733 NSS services available.
4734
4735 * The new functions `strtoq' and `strtouq' parse integer values from
4736 strings, like `strtol' and `strtoul', but they return `long long int' and
4737 `unsigned long long int' values, respectively (64-bit quantities).
4738
4739 * The new functions `strtof' and `strtold' parse floating-point values from
4740 strings, like `strtod', but they return `float' and `long double' values,
4741 respectively (on some machines `double' and `long double' are the same).
4742
4743 * Ulrich Drepper has contributed new implementations of the floating-point
4744 printing and reading code used in the `printf' family of functions and
4745 `strtod', `strtof', and `strtold'. These new functions are perfectly
4746 accurate, and much faster than the old ones.
4747
4748 * The implementation of the POSIX locale model was completely rewritten by
4749 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4750 to compile the POSIX locale definition.
4751
4752 * The former dummy implementations of the strcoll and strxfrm function are
4753 now replaced by fully functional code contributed by Ulrich Drepper. The
4754 collation information comes from the POSIX locale definitions.
4755
4756 * The new header <langinfo.h> defines an interface for accessing
4757 various locale-dependent data (using the locale chosen with `setlocale').
4758
4759 * Ulrich Drepper has contributed a new suite of functions for operation on
4760 wide-character and multibyte-character strings, in <wchar.h>;
4761 and classification and case conversion of wide characters, in <wctype.h>.
4762 These new functions are conforming to the ISO C, Amendement 1 specification.
4763
4764 * There is now a second implementation of the standard I/O library available.
4765 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4766 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4767
4768 * You can now use positional parameter specifications in format strings
4769 for the `printf' and `scanf' families of functions. For example,
4770 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4771 ``Number 6, Mr Jones''. This is mainly useful when providing different
4772 format strings for different languages, whose grammars may dictate
4773 different orderings of the values being printed. To support this
4774 feature, the interface for `register_printf_handler' has changed; see
4775 the header file <printf.h> for details.
4776
4777 * The `printf' and `scanf' families of functions now understand a new
4778 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4779 to group numbers as indicated by the locale; for `scanf' and friends, this
4780 says to accept as valid only a number with all the proper grouping
4781 separators in the right places. In the default "C" locale, numbers are
4782 not grouped; but locales for specific countries will define the usual
4783 conventions (i.e. separate thousands with `,' in the US locale).
4784
4785 * The pgrp functions have been regularized, slightly incompatibly but much
4786 less confusingly. The core functions are now `getpgid' and `setpgid',
4787 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4788 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4789 provided for compatibility. There is no longer an incompatible `getpgrp'
4790 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4791
4792 * The new header file <fts.h> and suite of functions simplify programs that
4793 operate on directory trees. This code comes from 4.4 BSD.
4794
4795 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4796 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4797 NSS scheme used in glibc.
4798
4799 * There is a new malloc debugging hook `__memalign_hook'.
4800
4801 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4802 `unsigned int' in <sys/types.h>. These are for compatibility only and
4803 their use is discouraged.
4804
4805 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4806 done differently, so that it works even without GNU ld.
4807
4808 * New function `euidaccess' checks allowed access to a file like `access',
4809 but using the effective IDs instead of the real IDs.
4810
4811 * The time zone data files have been updated for the latest and greatest
4812 local time conventions of the countries of the world.
4813
4814 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4815 see <dirent.h>.
4816
4817 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4818 for formatting floating-point numbers. They are provided only for
4819 compatibility; new programs should use `sprintf' instead. There are
4820 also equivalent function for the `long double' floating-point type and
4821 all functions also exist in a reentrant form.
4822
4823 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4824 functions for maintaining the login-record files (primarily of use to
4825 system programs such as `login'), and convenient functions for
4826 allocating and initializing a pseudo-terminal (pty) device.
4827
4828 * Ulrich Drepper has contributed new support for System V style
4829 shared memory and IPC on systems that support it.
4830
4831 * Ulrich Drepper has contributed several miscellaneous new functions found
4832 in System V: The `hsearch' family of functions provide an effective
4833 implementation of hash tables; `a64l' and `l64a' provide a very simple
4834 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4835 number generator.
4836
4837 * Ulrich Drepper has contributed new reentrant counterparts for the
4838 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4839
4840 * Ulrich Drepper has contributed new, highly-optimized versions of several
4841 string functions for the i486/Pentium family of processors.
4842
4843 * Ulrich Drepper has updated the Linux-specific code, based largely
4844 on work done in Hongjiu Lu's version of GNU libc for Linux.
4845 The GNU library now supports Linux versions 2.0.10 and later,
4846 using the ELF object file format (i[3456]86-*-linux).
4847
4848 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4849
4850 * David Mosberger-Tang and Richard Henderson have ported the C library
4851 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4852 dynamic linking support for ELF/Alpha.
4853
4854 * Richard Henderson contributed several Alpha optimized assembler function
4855 for arithmetic and string handling.
4856
4857 * Ulrich Drepper has contributed a new set of message catalog functions to
4858 support multiple languages using the <libintl.h> interface, for use with
4859 his new package GNU gettext. Translation volunteers have contributed
4860 catalogs of the library's messages in Spanish, German, and Korean.
4861
4862 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4863 program and the `catgets' function for reading the catalog files it
4864 creates. (The <libintl.h> interface is preferred; we include the
4865 <nl_types.h> interface using `catgets' only for source compatibility with
4866 programs already written to use it.)
4867
4868 * New header file <values.h> gives SVID-compatible names for <limits.h>
4869 constants.
4870
4871 * Various new macros, declarations, and small header files for compatibility
4872 with 4.4 BSD.
4873
4874 * New function `group_member' is a convenient way to check if a process has
4875 a given effective group ID.
4876
4877 * When using GCC 2.7 and later, the socket functions are now declared in a
4878 special way so that passing an argument of type `struct sockaddr_in *',
4879 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4880 `struct sockaddr *' type, does not generate a type-clash warning.
4881
4882 * New function `error' declared in header file <error.h> is a convenient
4883 function for printing error messages and optionally exiting; this is the
4884 canonical function used in GNU programs. The new functions `err', `warn',
4885 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4886 doing the same thing.
4887
4888 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4889 POSIX.2 `glob' function to do ~ and {...} expansion.
4890
4891 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4892 `clearenv' which is used in POSIX.9 is also available.
4893
4894 * New function `getsid' returns session ID number on systems that support it.
4895
4896 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4897 files <db.h> and <mpool.h> provide a rich set of functions for several
4898 types of simple databases stored in memory and in files, and <ndbm.h> is
4899 an old `ndbm'-compatible interface using the `db' functions. Link with
4900 `-ldb' to get these functions.
4901
4902 * New macro `strdupa' copies a string like `strdup', but uses local stack
4903 space from `alloca' instead of dynamic heap space from `malloc'.
4904
4905 * New function `strnlen' is like `strlen' but searches only a given maximum
4906 number of characters for the null terminator. `stpncpy', `strndup' and
4907 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4908 function.
4909
4910 * New function `statfs' in header <sys/statfs.h>.
4911
4912 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4913 provide convenient functions for operating on blocks of null-terminated
4914 strings.
4915
4916 * A new suite of functions in <utmp.h> handle all the details of reading
4917 and writing the utmp file.
4918
4919 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4920 Thorsten Kukuk.
4921
4922 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4923 completely POSIX compliant and also implemented the extended functionality
4924 to handle alternate digit representation and alternate era date formats.
4925
4926 * Ulrich Drepper provided an implementation of the `strptime' function
4927 defined in XPG4.2 which transforms a string into a `struct tm' value.
4928
4929 * Paul Eggert provided the tzselect shell script as part of the timezone
4930 code. The shell script makes it easy to select the correct timezone
4931 specification.
4932
4933 * The implementation of the malloc family of functions is completely replaced
4934 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4935 The implementation uses the mmap function (if available) and it is
4936 optimized for the use in multi threaded programs.
4937
4938 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4939 functions. This new functionality is usable by specifying a special
4940 salt string and it is compatible with implementation on *BSD systems.
4941
4942 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4943 `getsubopt' to handle second level command line options, `bsd_signal'
4944 to access BSD style `signal' functionality, the obsolete `regexp' style
4945 expression matcher.
4946
4947 * the `lchown' function is available on system which support this
4948 functionality.
4949
4950 * The implementation of the shadow password handling function was contributed
4951 by Ulrich Drepper.
4952
4953 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4954
4955 * POSIX.1g support was added. The <sys/select.h> header is available,
4956 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4957 implementation of `getaddrinfo'.
4958 \f
4959 Version 1.09
4960
4961 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4962
4963 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4964 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4965
4966 * The new function `daemon' from 4.4 BSD is useful for server programs that
4967 want to put themselves in the background.
4968
4969 * Joel Sherrill has contributed support for several standalone boards that
4970 run without an operating system.
4971
4972 * `printf', `scanf' and friends now accept a `q' type modifier for long
4973 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4974
4975 * All of the code taken from BSD (notably most of the math and networking
4976 routines) has been updated from the BSD 4.4-Lite release.
4977
4978 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4979
4980 * The new functions `getdomainname' and `setdomainname' fetch or change the
4981 YP/NIS domain name. These are system calls which exist on systems which
4982 have YP (aka NIS).
4983
4984 * The time zone data files have been updated for the latest international
4985 conventions.
4986
4987 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4988 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4989 \f
4990 Version 1.08
4991
4992 * The C library now includes support for Sun RPC, from Sun's free
4993 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4994 are included. (There is still no support for YP.)
4995
4996 * Tom Quinn has contributed a port of the C library to SGI machines running
4997 Irix 4 (mips-sgi-irix4).
4998
4999 * The new `lockf' function is a simplified interface to the locking
5000 facilities of `fcntl', included for compatibility.
5001
5002 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5003
5004 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5005
5006 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5007 compatibility.
5008
5009 * The error message printed by `assert' for a failed assertion now includes
5010 the name of the program (if using GNU ld) and the name of the calling
5011 function (with versions of GCC that support this).
5012
5013 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5014
5015 * The library now includes the <sys/mman.h> header file and memory
5016 management functions `mmap', `munmap', `mprotect', `msync', and
5017 `madvise', on systems that support those facilities.
5018
5019 * The interface for `mcheck' has changed slightly: the function called to
5020 abort the program when an allocation inconsistency is detected now takes
5021 an argument that indicates the type of failure. The new function
5022 `mprobe' lets you request a consistency check for a particular block at
5023 any time (checks are normally done only when you call `free' or `realloc'
5024 on a block).
5025
5026 * It is now possible to easily cross-compile the C library, building on one
5027 system a library to run on another machine and/or operating system. All
5028 you need to do is set the variable `HOST_CC' in `configparms' to the
5029 native compiler for programs to run on the machine you are building on (a
5030 few generator programs are used on Unix systems); set `CC' to the
5031 cross-compiler.
5032
5033 * The new function `fexecve' (only implemented on the GNU system) executes
5034 a program file given a file descriptor already open on the file.
5035 \f
5036 Version 1.07
5037
5038 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5039 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5040
5041 * You can set the variable `libprefix' in `configparms' to specify a prefix
5042 to be prepended to installed library files; this makes it easy to install
5043 the GNU C library to be linked as `-lgnuc' or whatever.
5044
5045 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5046 copies a limited number of characters from a string, and returns the
5047 address of the last character written.
5048
5049 * You no longer need to check for whether the installed `stddef.h' is
5050 compatible with the GNU C library. configure now checks for you.
5051
5052 * You can now define a per-stream `fileno' function to convert the
5053 stream's cookie into an integral file descriptor.
5054
5055 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5056 allocates zero bytes of storage, and returns a unique pointer which
5057 you can pass to `realloc' or `free'. The behavior is undefined if
5058 you dereference this pointer.
5059
5060 * The C library now runs on Sony NEWS m68k machines running either
5061 NewsOS 3 or NewsOS 4.
5062
5063 * The new `syscall' function is a system-dependent primitive function
5064 for invoking system calls. It has the canonical behavior on Unix
5065 systems, including unreliable return values for some calls (such as
5066 `pipe', `fork' and `getppid').
5067
5068 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5069 to `EAGAIN', which is the preferred name. On systems whose kernels
5070 use two distinct codes, the C library now translates EWOULDBLOCK to
5071 EAGAIN in every system call function.
5072 \f
5073 Version 1.06
5074
5075 * The GNU C Library Reference Manual is now distributed with the library.
5076 `make dvi' will produce a DVI file of the printed manual.
5077 `make info' will produce Info files that you can read on line using C-h i
5078 in Emacs or the `info' program.
5079 Please send comments on the manual to bug-glibc-manual@gnu.org.
5080
5081 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5082
5083 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5084
5085 * Jason Merrill has contributed a port to the Sequent Symmetry running
5086 Dynix version 3 (i386-sequent-dynix).
5087
5088 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5089 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5090
5091 * New function `memory_warnings' lets you arrange to get warnings when
5092 malloc is running out of memory to allocate, like Emacs gives you.
5093
5094 * The C library now contains the relocating allocator used in Emacs 19 for
5095 its editing buffers. This allocator (ralloc) minimizes allocation
5096 overhead and fragmentation by moving allocated regions around whenever it
5097 needs to. You always refer to a ralloc'd region with a "handle" (a
5098 pointer to a pointer--an object of type `void **').
5099
5100 * There is a new `printf' format: `%m' gives you the string corresponding
5101 to the error code in `errno'.
5102
5103 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5104 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5105 pass, the `a' modifier says to fill in a `char **' you pass with a
5106 malloc'd string.
5107
5108 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5109 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5110 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5111
5112 * `mkstemp' is a traditional Unix function to atomically create and open a
5113 uniquely-named temporary file.
5114 \f
5115 Version 1.05
5116
5117 * The standard location for the file that says what the local timezone is
5118 has changed again. It is now `/usr/local/etc/localtime' (or more
5119 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5120
5121 * The distribution no longer contains any files with names longer than 14
5122 characters.
5123
5124 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5125 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5126
5127 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5128 \f
5129 Version 1.04
5130
5131 * The configuration process has changed quite a bit. The `configure'
5132 script is now used just like the configuration scripts for other GNU
5133 packages. The `sysdeps' directory hierarchy is much rearranged.
5134 The file `INSTALL' explains the new scheme in detail.
5135
5136 * The header files no longer need to be processed into ANSI C and
5137 traditional C versions. There is just one set of files to install, and
5138 it will work with ANSI or old C compilers (including `gcc -traditional').
5139
5140 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5141 MIPS DECStation running Ultrix 4.
5142
5143 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5144 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5145 made itself into a shared library.
5146
5147 * Yet further improved support for the i386, running 4.3 BSD-like systems
5148 (such as Mach 3 with the Unix single-server), or System V.
5149
5150 * New function `strncasecmp' to do case-insensitive string comparison
5151 with limited length.
5152
5153 * New function `strsep' is a reentrant alternative to `strtok'.
5154
5155 * New functions `scandir' and `alphasort' for searching directories.
5156
5157 * New function `setenv' is a better interface to `putenv'.
5158
5159 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5160 function for traversing a directory tree.
5161
5162 * The GNU obstack package is now also part of the C library.
5163 The new function `open_obstack_stream' creates a stdio stream that
5164 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5165 formatted output directly to an obstack.
5166
5167 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5168 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5169
5170 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5171
5172 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5173 things to your strings.
5174
5175 * There are some new test programs: test-fseek, testmb, and testrand.
5176
5177 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5178 These ports are not finished, but are a good starting place for really
5179 supporting those systems.
5180
5181 * `/etc/localtime' is now the standard location for the file that says what
5182 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5183 This follows the general principle that `/etc' is the place for all local
5184 configuration files.
5185
5186 * The C library header files now use `extern "C"' when used by the C++
5187 compiler, so the C library should now work with C++ code.
5188
5189 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5190 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5191 in <strings.h>.)
5192
5193 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5194 function which uses the `merge sort' algorithm, and is said to be
5195 significantly faster than the old GNU `qsort' function. Merge sort is
5196 now the standard `qsort' function. The new algorithm can require a lot
5197 of temporary storage; so, the old sorting function is called when the
5198 required storage is not available.
5199
5200 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5201 provides the Unix `crypt' function, plus some other entry points.
5202
5203 * The code and header files taken from 4.4 BSD have been updated with the
5204 latest files released from Berkeley.
5205 \f
5206 ----------------------------------------------------------------------
5207 Copyright information:
5208
5209 Copyright (C) 1992-2018 Free Software Foundation, Inc.
5210
5211 Permission is granted to anyone to make or distribute verbatim copies
5212 of this document as received, in any medium, provided that the
5213 copyright notice and this permission notice are preserved,
5214 thus giving the recipient permission to redistribute in turn.
5215
5216 Permission is granted to distribute modified versions
5217 of this document, or of portions of it,
5218 under the above conditions, provided also that they
5219 carry prominent notices stating who last changed them.
5220 \f
5221 Local variables:
5222 version-control: never
5223 fill-column: 76
5224 End: