]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
nptl: Fix testcases for new pthread cancellation mechanism
[thirdparty/glibc.git] / NEWS
1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2019 Free Software Foundation, Inc.
3 See the end for copying conditions.
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 * The getcpu wrapper function has been added, which returns the currently
13 used CPU and NUMA node. This function is Linux-specific.
14
15 * A new convenience target has been added for distribution maintainers
16 to build and install all locales as directories with files. The new
17 target is run by issuing the following command in your build tree:
18 'make localedata/install-locale-files', with an optional DESTDIR
19 to set the install root if you wish to install into a non-default
20 configured location.
21
22 * Optimized generic exp, exp2, log, log2, pow, sinf, cosf, sincosf and tanf.
23
24 * The reallocarray function is now declared under _DEFAULT_SOURCE, not just
25 for _GNU_SOURCE, to match BSD environments.
26
27 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
28 indicates that it will abort the transaction prior to entering the kernel
29 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
30 suspended, and this caused some undefined side-effects issues by aborting
31 transactions manually. Glibc avoided it by abort transactions manually on
32 each syscall, but it lead to performance issues on newer kernels where the
33 HTM state is saved and restore lazily (the state being saved even when the
34 process actually does not use HTM).
35
36 * The functions posix_spawn_file_actions_addchdir_np and
37 posix_spawn_file_actions_addfchdir_np have been added, enabling
38 posix_spawn and posix_spawnp to run the new process in a different
39 directory. These functions are GNU extensions. The function
40 posix_spawn_file_actions_addchdir_np is similar to the Solaris function
41 of the same name.
42
43 * The popen and system do not run atfork handlers anymore (BZ#17490).
44 Although it is a possible POSIX violation, the POSIX rationale in
45 pthread_atfork documentation regarding atfork handlers is to handle
46 inconsistent mutex state after a fork call in a multi-threaded process.
47 In both popen and system there is no direct access to user-defined mutexes.
48
49 * Support for the C-SKY ABIV2 running on Linux has been added. This port
50 requires at least binutils-2.32, gcc-9.0, and linux-4.20. Two ABIs are
51 supported:
52 - C-SKY ABIV2 soft-float little-endian
53 - C-SKY ABIV2 hard-float little-endian
54
55 Deprecated and removed features, and other changes affecting compatibility:
56
57 * The glibc.tune tunable namespace has been renamed to glibc.cpu and the
58 tunable glibc.tune.cpu has been renamed to glibc.cpu.name.
59
60 * The type of the pr_uid and pr_gid members of struct elf_prpsinfo, defined
61 in <sys/procfs.h>, has been corrected to match the type actually used by
62 the Linux kernel. This affects the size and layout of that structure on
63 MicroBlaze, MIPS (n64 ABI only), Nios II and RISC-V.
64
65 * For the MIPS n32 ABI, the type of the pr_sigpend and pr_sighold members of
66 struct elf_prstatus, and the pr_flag member of struct elf_prpsinfo,
67 defined in <sys/procfs.h>, has been corrected to match the type actually
68 used by the Linux kernel. This affects the size and layout of those
69 structures.
70
71 * An archaic GNU extension to scanf, under which '%as', '%aS', and '%a[...]'
72 meant to scan a string and allocate space for it with malloc, is now
73 restricted to programs compiled in C89 or C++98 mode with _GNU_SOURCE
74 defined. This extension conflicts with C99's use of '%a' to scan a
75 hexadecimal floating-point number, which is now available to programs
76 compiled as C99 or C++11 or higher, regardless of _GNU_SOURCE.
77
78 POSIX.1-2008 includes the feature of allocating a buffer for string input
79 with malloc, using the modifier letter 'm' instead. Programs using
80 '%as', '%aS', or '%a[...]' with the old GNU meaning should change to
81 '%ms', '%mS', or '%m[...]' respectively. Programs that wish to use the
82 C99 '%a' no longer need to avoid _GNU_SOURCE.
83
84 GCC's -Wformat warnings can detect most uses of this extension, as long
85 as all functions that call vscanf, vfscanf, or vsscanf are annotated with
86 __attribute__ ((format (scanf, ...))).
87
88 Changes to build and runtime requirements:
89
90 * Python 3.4 or later is required to build the GNU C Library.
91
92 * On most architectures, GCC 5 or later is required to build the GNU C
93 Library. (On powerpc64le, GCC 6.2 or later is still required, as before.)
94
95 Older GCC versions and non-GNU compilers are still supported when
96 compiling programs that use the GNU C Library.
97
98 Security related changes:
99
100 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
101 denial of service due to resource exhaustion when processing getaddrinfo
102 calls with crafted host names. Reported by Guido Vranken.
103
104 The following bugs are resolved with this release:
105
106 [The release manager will add the list generated by
107 scripts/list-fixed-bugs.py just before the release.]
108
109 \f
110 Version 2.28
111
112 Major new features:
113
114 * The localization data for ISO 14651 is updated to match the 2016
115 Edition 4 release of the standard, this matches data provided by
116 Unicode 9.0.0. This update introduces significant improvements to the
117 collation of Unicode characters. This release deviates slightly from
118 the standard in that the collation element ordering for lowercase and
119 uppercase LATIN script characters is adjusted to ensure that regular
120 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
121 is not matched by [a-z]. With the update many locales have been
122 updated to take advantage of the new collation information. The new
123 collation information has increased the size of the compiled locale
124 archive or binary locales.
125
126 * The GNU C Library can now be compiled with support for Intel CET, AKA
127 Intel Control-flow Enforcement Technology. When the library is built
128 with --enable-cet, the resulting glibc is protected with indirect
129 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
130 compatible with all existing executables and shared libraries. This
131 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
132 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
133 capable of multi-byte NOPs, like x86-64 processors as well as Intel
134 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
135 x86_64 and x32 on non-CET processors. --enable-cet has been tested
136 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
137 validated for i686.
138
139 * The GNU C Library now has correct support for ABSOLUTE symbols
140 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
141 relocated incorrectly or in some cases discarded. The GNU linker can
142 make use of the newer semantics, but it must communicate it to the
143 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
144 field) to indicate such support is required.
145
146 * Unicode 11.0.0 Support: Character encoding, character type info, and
147 transliteration tables are all updated to Unicode 11.0.0, using
148 generator scripts contributed by Mike FABIAN (Red Hat).
149
150 * <math.h> functions that round their results to a narrower type are added
151 from TS 18661-1:2014 and TS 18661-3:2015:
152
153 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
154 fMxaddfNx functions.
155
156 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
157 fMxsubfNx functions.
158
159 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
160 fMxmulfNx functions.
161
162 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
163 fMxdivfNx functions.
164
165 * Two grammatical forms of month names are now supported for the following
166 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
167 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
168 support two grammatical forms in abbreviated month names: Catalan, Greek,
169 and Kashubian.
170
171 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
172 include the support for two grammatical forms of month names.
173
174 * Building and running on GNU/Hurd systems now works without out-of-tree
175 patches.
176
177 * The renameat2 function has been added, a variant of the renameat function
178 which has a flags argument. If the flags are zero, the renameat2 function
179 acts like renameat. If the flag is not zero and there is no kernel
180 support for renameat2, the function will fail with an errno value of
181 EINVAL. This is different from the existing gnulib function renameatu,
182 which performs a plain rename operation in case of a RENAME_NOREPLACE
183 flags and a non-existing destination (and therefore has a race condition
184 that can clobber the destination inadvertently).
185
186 * The statx function has been added, a variant of the fstatat64
187 function with an additional flags argument. If there is no direct
188 kernel support for statx, glibc provides basic stat support based on
189 the fstatat64 function.
190
191 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
192 library if installed. libidn2 version 2.0.5 or later is recommended. If
193 libidn2 is not available, internationalized domain names are not encoded
194 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
195 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
196 with an encoding error.) Flags which used to change the IDN encoding and
197 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
198 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
199 deprecated. They no longer have any effect.
200
201 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
202 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
203 range of ELF gABI expressions including such constructs as
204 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
205 have been further restricted, and where in the past a dynamic string
206 token sequence may have been interpreted as a literal string it will
207 now cause a load failure. These load failures were always considered
208 unspecified behaviour from the perspective of the dynamic loader, and
209 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
210 results in a load failure now.
211
212 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
213 implementation includes all the standard functions provided by
214 <threads.h>:
215
216 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
217 thrd_detach, thrd_exit, and thrd_join for thread management.
218
219 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
220 mtx_destroy for mutual exclusion.
221
222 - call_once for function call synchronization.
223
224 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
225 cnd_wait for conditional variables.
226
227 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
228
229 Application developers must link against libpthread to use ISO C threads.
230
231 Deprecated and removed features, and other changes affecting compatibility:
232
233 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
234 installed. Software that was using either header should be updated to
235 use standard <stdio.h> interfaces instead.
236
237 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
238 This was never required by the C standard, and the macros just expanded
239 to call alternative names for the same functions. If you hoped getc and
240 putc would provide performance improvements over fgetc and fputc, instead
241 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
242 necessary, flockfile and funlockfile.
243
244 * All stdio functions now treat end-of-file as a sticky condition. If you
245 read from a file until EOF, and then the file is enlarged by another
246 process, you must call clearerr or another function with the same effect
247 (e.g. fseek, rewind) before you can read the additional data. This
248 corrects a longstanding C99 conformance bug. It is most likely to affect
249 programs that use stdio to read interactive input from a terminal.
250 (Bug #1190.)
251
252 * The macros 'major', 'minor', and 'makedev' are now only available from
253 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
254 headers that happen to include <sys/types.h>. These macros are rarely
255 used, not part of POSIX nor XSI, and their names frequently collide with
256 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
257 further explanation.
258
259 <sys/sysmacros.h> is a GNU extension. Portable programs that require
260 these macros should first include <sys/types.h>, and then include
261 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
262
263 * The tilegx*-*-linux-gnu configurations are no longer supported.
264
265 * The obsolete function ustat is no longer available to newly linked
266 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
267 function has been deprecated in favor of fstatfs and statfs.
268
269 * The obsolete function nfsservctl is no longer available to newly linked
270 binaries. This function was specific to systems using the Linux kernel
271 and could not usefully be used with the GNU C Library on systems with
272 version 3.1 or later of the Linux kernel.
273
274 * The obsolete function name llseek is no longer available to newly linked
275 binaries. This function was specific to systems using the Linux kernel
276 and was not declared in a header. Programs should use the lseek64 name
277 for this function instead.
278
279 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
280 getaddrinfo and getnameinfo functions have been deprecated. The behavior
281 previously selected by them is now always enabled.
282
283 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
284 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
285 restriction (rejecting '_' in host names, among other things) has been
286 removed, for increased compatibility with non-IDN name resolution.
287
288 * The fcntl function now have a Long File Support variant named fcntl64. It
289 is added to fix some Linux Open File Description (OFD) locks usage on non
290 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
291 fcntl and LFS support is handled transparently. Also for Linux, the OFD
292 locks act as a cancellation entrypoint.
293
294 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
295 ecb_crypt, and des_setparity are no longer available to newly linked
296 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
297 longer installed. These functions encrypted and decrypted data with the
298 DES block cipher, which is no longer considered secure. Software that
299 still uses these functions should switch to a modern cryptography library,
300 such as libgcrypt.
301
302 * Reflecting the removal of the encrypt and setkey functions above, the
303 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
304 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
305 enabled.
306
307 * The obsolete function fcrypt is no longer available to newly linked
308 binaries. It was just another name for the standard function crypt,
309 and it has not appeared in any header file in many years.
310
311 * We have tentative plans to hand off maintenance of the passphrase-hashing
312 library, libcrypt, to a separate development project that will, we hope,
313 keep up better with new passphrase-hashing algorithms. We will continue
314 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
315 'crypt_r' should not need to change at all; however, distributions will
316 need to install <crypt.h> and libcrypt from a separate project.
317
318 In this release, if the configure option --disable-crypt is used, glibc
319 will not install <crypt.h> or libcrypt, making room for the separate
320 project's versions of these files. The plan is to make this the default
321 behavior in a future release.
322
323 Changes to build and runtime requirements:
324
325 GNU make 4.0 or later is now required to build glibc.
326
327 Security related changes:
328
329 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
330 been fixed by removing the glibc-internal IDNA implementation and using
331 the system-provided libidn2 library instead. Originally reported by Hanno
332 Böck and Christian Weisgerber.
333
334 CVE-2017-18269: An SSE2-based memmove implementation for the i386
335 architecture could corrupt memory. Reported by Max Horn.
336
337 CVE-2018-11236: Very long pathname arguments to realpath function could
338 result in an integer overflow and buffer overflow. Reported by Alexey
339 Izbyshev.
340
341 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
342 architecture could write beyond the target buffer, resulting in a buffer
343 overflow. Reported by Andreas Schwab.
344
345 The following bugs are resolved with this release:
346
347 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
348 [6889] manual: 'PWD' mentioned but not specified
349 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
350 when __WORDSIZE != 64
351 [13762] regex: re_search etc. should return -2 on memory exhaustion
352 [13888] build: /tmp usage during testing
353 [13932] math: dbl-64 pow unexpectedly slow for some inputs
354 [14092] nptl: Support C11 threads
355 [14095] localedata: Review / update collation data from Unicode / ISO
356 14651
357 [14508] libc: -Wformat warnings
358 [14553] libc: Namespace pollution loff_t in sys/types.h
359 [14890] libc: Make NT_PRFPREG canonical.
360 [15105] libc: Extra PLT references with -Os
361 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
362 conversion is given
363 [16335] manual: Feature test macro documentation incomplete and out of
364 date
365 [16552] libc: Unify umount implementations in terms of umount2
366 [17082] libc: htons et al.: statement-expressions prevent use on global
367 scope with -O1 and higher
368 [17343] libc: Signed integer overflow in /stdlib/random_r.c
369 [17438] localedata: pt_BR: wrong d_fmt delimiter
370 [17662] libc: please implement binding for the new renameat2 syscall
371 [17721] libc: __restrict defined as /* Ignore */ even in c11
372 [17979] libc: inconsistency between uchar.h and stdint.h
373 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
374 [18023] libc: extend_alloca is broken (questionable pointer comparison,
375 horrible machine code)
376 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
377 last constant.
378 [18471] libc: llseek should be a compat symbol
379 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
380 symbols
381 [18991] nss: nss_files skips large entry in database
382 [19239] libc: Including stdlib.h ends up with macros major and minor being
383 defined
384 [19463] libc: linknamespace failures when compiled with -Os
385 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
386 [19527] locale: Normalized charset name not recognized by setlocale
387 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
388 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
389 fpscr.c'
390 [19728] network: out of bounds stack read in libidn function
391 idna_to_ascii_4i (CVE-2016-6261)
392 [19729] network: out of bounds heap read on invalid utf-8 inputs in
393 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
394 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
395 the base address
396 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
397 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
398 [20419] dynamic-link: files with large allocated notes crash in
399 open_verify
400 [20530] libc: bswap_16 should use __builtin_bswap16() when available
401 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
402 [20980] manual: CFLAGS environment variable replaces vital options
403 [21163] regex: Assertion failure in pop_fail_stack when executing a
404 malformed regexp (CVE-2015-8985)
405 [21234] manual: use of CFLAGS makes glibc detect no optimization
406 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
407 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
408 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
409 [21508] locale: intl/tst-gettext failure with latest msgfmt
410 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
411 [21812] network: getifaddrs() returns entries with ifa_name == NULL
412 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
413 dlopen
414 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
415 AT_SECURE=1
416 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
417 [22247] network: Integer overflow in the decode_digit function in
418 puny_decode.c in libidn (CVE-2017-14062)
419 [22342] nscd: NSCD not properly caching netgroup
420 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
421 [22550] localedata: es_ES locale (and other es_* locales): collation
422 should treat ñ as a primary different character, sync the collation
423 for Spanish with CLDR
424 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
425 by gcc configured with --enable-default-pie
426 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
427 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
428 crossing 2GB threshold (CVE-2017-18269)
429 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
430 es_BO
431 [22735] time: Misleading typo in time.h source comment regarding
432 CLOCKS_PER_SECOND
433 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
434 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
435 `__assert_perror_fail ()' to `__assert_fail_base ()'
436 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
437 dlopen failures
438 [22786] libc: Stack buffer overflow in realpath() if input size is close
439 to SSIZE_MAX (CVE-2018-11236)
440 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
441 through an absolute DT_NEEDED path
442 [22792] build: tcb-offsets.h dependency dropped
443 [22797] libc: pkey_get() uses non-reserved name of argument
444 [22807] libc: PTRACE_* constants missing for powerpc
445 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
446 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
447 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
448 [22848] localedata: ca_ES: update date definitions from CLDR
449 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
450 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
451 [22896] localedata: Update locale data for an_ES
452 [22902] math: float128 test failures with GCC 8
453 [22918] libc: multiple common of `__nss_shadow_database'
454 [22919] libc: sparc32: backtrace yields infinite backtrace with
455 makecontext
456 [22926] libc: FTBFS on powerpcspe
457 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
458 required
459 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
460 [22947] libc: FAIL: misc/tst-preadvwritev2
461 [22963] localedata: cs_CZ: Add alternative month names
462 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
463 [22996] localedata: change LC_PAPER to en_US in es_BO locale
464 [22998] dynamic-link: execstack tests are disabled when SELinux is
465 disabled
466 [23005] network: Crash in __res_context_send after memory allocation
467 failure
468 [23007] math: strtod cannot handle -nan
469 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
470 [23036] regex: regex equivalence class regression
471 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
472 [23069] libc: sigaction broken on riscv64-linux-gnu
473 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
474 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
475 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
476 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
477 and libc build with -Os)
478 [23140] localedata: More languages need two forms of month names
479 [23145] libc: _init/_fini aren't marked as hidden
480 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
481 [23171] math: C++ iseqsig for long double converts arguments to double
482 [23178] nscd: sudo will fail when it is run in concurrent with commands
483 that changes /etc/passwd
484 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
485 (CVE-2018-11237)
486 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
487 [23208] localedata: New locale - Lower Sorbian (dsb)
488 [23233] regex: Memory leak in build_charclass_op function in file
489 posix/regcomp.c
490 [23236] stdio: Harden function pointers in _IO_str_fields
491 [23250] nptl: Offset of __private_ss differs from GCC
492 [23253] math: tgamma test suite failures on i686 with -march=x86-64
493 -mtune=generic -mfpmath=sse
494 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
495 AT_SECURE
496 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
497 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
498 parser.c
499 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
500 [23277] math: nan function should not have const attribute
501 [23279] math: scanf and strtod wrong for some hex floating-point
502 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
503 directed rounding
504 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
505 [23303] build: undefined reference to symbol
506 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
507 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
508 lookup
509 [23313] stdio: libio vtables validation and standard file object
510 interposition
511 [23329] libc: The __libc_freeres infrastructure is not properly run across
512 DSO boundaries.
513 [23349] libc: Various glibc headers no longer compatible with
514 <linux/time.h>
515 [23351] malloc: Remove unused code related to heap dumps and malloc
516 checking
517 [23363] stdio: stdio-common/tst-printf.c has non-free license
518 [23396] regex: Regex equivalence regression in single-byte locales
519 [23422] localedata: oc_FR: More updates of locale data
520 [23442] build: New warning with GCC 8
521 [23448] libc: Out of bounds access in IBM-1390 converter
522 [23456] libc: Wrong index_cpu_LZCNT
523 [23458] build: tst-get-cpu-features-static isn't added to tests
524 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
525 processors
526 [23467] dynamic-link: x86/CET: A property note parser bug
527
528 \f
529 Version 2.27
530
531 Major new features:
532
533 * The GNU C Library can now be compiled with support for building static
534 PIE executables (See --enable-static-pie in INSTALL). These static PIE
535 executables are like static executables but can be loaded at any address
536 and provide additional security hardening benefits at the cost of some
537 memory and performance. When the library is built with --enable-static-pie
538 the resulting libc.a is usable with GCC 8 and above to create static PIE
539 executables using the GCC option '-static-pie'. This feature is currently
540 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
541 aarch64 with binutils 2.30 or later.
542
543 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
544 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
545 H.J. Lu from Intel.
546
547 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
548
549 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
550
551 * In order to support faster and safer process termination the malloc API
552 family of functions will no longer print a failure address and stack
553 backtrace after detecting heap corruption. The goal is to minimize the
554 amount of work done after corruption is detected and to avoid potential
555 security issues in continued process execution. Reducing shutdown time
556 leads to lower overall process restart latency, so there is benefit both
557 from a security and performance perspective.
558
559 * The abort function terminates the process immediately, without flushing
560 stdio streams. Previous glibc versions used to flush streams, resulting
561 in deadlocks and further data corruption. This change also affects
562 process aborts as the result of assertion failures.
563
564 * On platforms where long double has the IEEE binary128 format (aarch64,
565 alpha, mips64, riscv, s390 and sparc), the math library now implements
566 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
567 These are the same interfaces added in version 2.26 for some platforms where
568 this format is supported but is not the format of long double.
569
570 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
571 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
572 implements interfaces for that type, as defined by ISO/IEC TS
573 18661-3:2015. These are corresponding interfaces to those supported for
574 _Float128.
575
576 * The math library now implements interfaces for the _Float32, _Float64 and
577 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
578 corresponding interfaces to those supported for _Float128.
579
580 * glibc now implements the memfd_create and mlock2 functions on Linux.
581
582 * Support for memory protection keys was added. The <sys/mman.h> header now
583 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
584 pkey_get.
585
586 * The copy_file_range function was added.
587
588 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
589
590 * The ldconfig utility now processes `include' directives using the C/POSIX
591 collation ordering. Previous glibc versions used locale-specific
592 ordering, the change might break systems that relied on that.
593
594 * Support for two grammatical forms of month names has been added.
595 In a call to strftime, the "%B" and "%b" format specifiers will now
596 produce the grammatical form required when the month is used as part
597 of a complete date. New "%OB" and "%Ob" specifiers produce the form
598 required when the month is named by itself. For instance, in Greek
599 and in many Slavic and Baltic languages, "%B" will produce the month
600 in genitive case, and "%OB" will produce the month in nominative case.
601
602 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
603 are all valid and will all accept any known form of month
604 name---standalone or complete, abbreviated or full. In a call to
605 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
606 the strings used by "%B" and "%b", respectively. New query
607 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
608 used by "%OB" and "%Ob", respectively.
609
610 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
611 define the strings for %OB and %Ob, respectively; these have the
612 same syntax as "mon" and "abmon". These arrays are optional; if they
613 are not provided then they have the same content as "mon" and "abmon",
614 respectively.
615
616 These features are provided for locales which define "alt_mon" and/or
617 "ab_alt_mon" in their locale source data. This release includes such
618 alternative month name data for the following languages: Belarusian,
619 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
620
621 This feature is currently a GNU extension, but it is expected to
622 be added to the next revision of POSIX, and it is also already
623 available on some BSD-derived operating systems.
624
625 This feature will cause existing statically compiled applications
626 to fail to load locales and fall back to the builtin C/POSIX locales.
627 See notes below for other changes affecting compatibility.
628
629 * Support for the RISC-V ISA running on Linux has been added. This port
630 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
631 for the following ISA and ABI pairs:
632
633 - rv64imac lp64
634 - rv64imafdc lp64
635 - rv64imafdc lp64d
636
637 Deprecated and removed features, and other changes affecting compatibility:
638
639 * Statically compiled applications attempting to load locales compiled for the
640 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
641 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
642 support for two grammatical forms of the month names, also extends the locale
643 data binary format. Static applications needing locale support must be
644 recompiled to match the runtime and data they are deployed with. In some
645 distributions there is an upgrade window where dynamically linked applications
646 may use a new library but the old locale data and also fall back to the
647 builtin C/POSIX locales; restarting the application process is sufficient to
648 fix this.
649
650 * Support for statically linked applications which call dlopen is deprecated
651 and will be removed in a future version of glibc. Applications which call
652 dlopen need to be linked dynamically instead.
653
654 * Support for old programs which use internal stdio data structures and
655 functions is deprecated. This includes programs which use the C++ streams
656 provided by libstdc++ in GCC 2.95. Programs which use the internal
657 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
658 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
659 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
660 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
661 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
662 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
663 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
664 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
665 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
666 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
667 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
668 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
669 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
670 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
671 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
672 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
673 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
674 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
675 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
676 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
677 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
678 future version of glibc. Unlike other symbol removals, these old
679 applications will not be supported using compatibility symbols.
680
681 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
682 defined by <sys/ptrace.h>.
683
684 * libm no longer supports SVID error handling (calling a user-provided
685 matherr function on error) or the _LIB_VERSION variable to control error
686 handling. (SVID error handling and the _LIB_VERSION variable still work
687 for binaries linked against older versions of the GNU C Library.) The
688 libieee.a library is no longer provided. math.h no longer defines struct
689 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
690 TLOSS, PLOSS and HUGE.
691
692 * The libm functions pow10, pow10f and pow10l are no longer supported for
693 new programs. Programs should use the standard names exp10, exp10f and
694 exp10l for these functions instead.
695
696 * The mcontext_t type is no longer the same as struct sigcontext. On
697 platforms where it was previously the same, this changes the C++ name
698 mangling for interfaces involving this type.
699
700 * The add-ons mechanism for building additional packages at the same time as
701 glibc has been removed. The --enable-add-ons configure option is now
702 ignored.
703
704 * The --without-fp configure option is now ignored. Whether hardware
705 floating-point instructions are used is now configured based on whether
706 the compiler used at configure time (without any options implied by a
707 --with-cpu= configure option) uses such instructions.
708
709 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
710 the specified string can be parsed as a domain name.
711
712 * In the malloc_info output, the <heap> element may contain another <aspace>
713 element, "subheaps", which contains the number of sub-heaps.
714
715 * The libresolv function p_secstodate is no longer supported for new
716 programs.
717
718 * The tilepro-*-linux-gnu configuration is no longer supported.
719
720 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
721 and will be removed in a future release. Software that is still using
722 either header should be updated to use standard <stdio.h> interfaces
723 instead.
724
725 libio.h was originally the header for a set of supported GNU extensions,
726 but they have not been maintained as such in many years, they are now
727 standing in the way of improvements to stdio, and we don't think there are
728 any remaining external users. _G_config.h was never intended for public
729 use, but predates the bits convention.
730
731 Changes to build and runtime requirements:
732
733 * bison version 2.7 or later is required to generate code in the 'intl'
734 subdirectory.
735
736 Security related changes:
737
738 CVE-2009-5064: The ldd script would sometimes run the program under
739 examination directly, without preventing code execution through the
740 dynamic linker. (The glibc project disputes that this is a security
741 vulnerability; only trusted binaries must be examined using the ldd
742 script.)
743
744 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
745 suffered from a one-byte overflow during ~ operator processing (either
746 on the stack or the heap, depending on the length of the user name).
747 Reported by Tim Rühsen.
748
749 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
750 would sometimes fail to free memory allocated during ~ operator
751 processing, leading to a memory leak and, potentially, to a denial
752 of service.
753
754 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
755 without GLOB_NOESCAPE, could write past the end of a buffer while
756 unescaping user names. Reported by Tim Rühsen.
757
758 CVE-2017-17426: The malloc function, when called with an object size near
759 the value SIZE_MAX, would return a pointer to a buffer which is too small,
760 instead of NULL. This was a regression introduced with the new malloc
761 thread cache in glibc 2.26. Reported by Iain Buclaw.
762
763 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
764 to the allocation of too much memory. (This is not a security bug per se,
765 it is mentioned here only because of the CVE assignment.) Reported by
766 Qualys.
767
768 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
769 of the number of search path components. (This is not a security
770 vulnerability per se because no trust boundary is crossed if the fix for
771 CVE-2017-1000366 has been applied, but it is mentioned here only because
772 of the CVE assignment.) Reported by Qualys.
773
774 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
775 for AT_SECURE or SUID binaries could be used to load libraries from the
776 current directory.
777
778 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
779 succeeds without returning an absolute path due to unexpected behaviour
780 of the Linux kernel getcwd syscall. Reported by halfdog.
781
782 CVE-2018-6485: The posix_memalign and memalign functions, when called with
783 an object size near the value of SIZE_MAX, would return a pointer to a
784 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
785
786 CVE-2018-6551: The malloc function, when called with an object size near
787 the value of SIZE_MAX, would return a pointer to a buffer which is too
788 small, instead of NULL.
789
790 The following bugs are resolved with this release:
791
792 [866] glob: glob should match dangling symlinks
793 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
794 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
795 locale
796 [5997] math: Very slow execution of sinf function
797 [10580] localedata: hr_HR: updated locale
798 [10871] locale: 'mon' array should contain both nominative and genitive
799 cases
800 [12349] localedata: eu_ES: incorrect thousands separator
801 [13605] localedata: shn_MM: new Shan locale
802 [13805] localedata: ru_RU: currency should use ',' as radix point
803 [13953] localedata: km_KH: locale update
804 [13994] localedata: mjw_IN: new locale
805 [14121] build: make writes .mo files in po directory
806 [14333] libc: Fix the race between atexit() and exit()
807 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
808 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
809 "Bangla"
810 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
811 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
812 width Latin characters
813 [15332] localedata: es_CU: locale update
814 [15436] stdio: Don't close or flush stdio streams on abort
815 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
816 letters
817 [16148] localedata: ca_ES: incorrect thousands separator
818 [16750] dynamic-link: ldd should not try to execute the binaries
819 (CVE-2009-5064)
820 [16777] localedata: pl_PL: incorrect thousands separator in locale
821 [16905] localedata: hanzi: new collation
822 [17563] localedata: cmn_TW: add hanzi collation
823 [17750] localedata: wrong collation order of diacritics in most locales
824 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
825 if malloc succeeded
826 [17956] build: Build fails on missing definitions from header file
827 nss/nss.h when Mozilla NSS is used for cryptography
828 [18203] libc: realpath() does not handle unreachable paths correctly
829 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
830 [18812] localedata: kab_DZ: new Kabyle Algeria locale
831 [18822] libc: Internal functions are called via PLT
832 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
833 [19170] libc: __gmon_start__ defined in hppa in crtn.S
834 [19574] libc: glibc should support building static PIE binaries
835 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
836 U+309A
837 [19971] glob: glob: Do not skip entries with zero d_ino values
838 [19982] localedata: fr.po: spelling mistake for error code EXDEV
839 [20008] localedata: km_KH: convert to translit_neutral
840 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
841 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
842 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
843 [20482] localedata: de_CH: abbreviated weekdays should be two letters
844 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
845 for Nicaragua
846 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
847 they're set, wrong errors returned by gaih_inet when lookup functions
848 are not found.
849 [20756] localedata: [PATCH] Use Unicode wise thousands separator
850 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
851 access
852 [20952] localedata: yuw_PG: new locale
853 [21084] localedata: charmaps/IBM858: new codepage
854 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
855 [21242] libc: assert gives pedantic warning in old gcc versions
856 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
857 __regcall calling convention
858 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
859 [21326] libc: C99 functions are not declared for C++11 and later if
860 _GNU_SOURCE is not predefined by g++
861 [21457] libc: sys/ucontext.h namespace
862 [21530] libc: tmpfile() should be implemented using O_TMPFILE
863 [21660] math: GCC fails to compile a formula with tgmath.h
864 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
865 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
866 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
867 [21684] math: tgmath.h handling of complex integers
868 [21685] math: tgmath.h handling of bit-fields
869 [21686] math: tgmath.h handling of __int128
870 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
871 locale
872 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
873 protector=all
874 [21750] localedata: column width of characters incompatible with classical
875 wcwidth
876 [21754] malloc: malloc: Perform as little work as possible after heap
877 consistency check failures
878 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
879 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
880 [21791] string: Unused XXX_chk_XXX functions in libc.a
881 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
882 PIE
883 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
884 locales
885 [21845] localedata: Added new Locale bho_NP
886 [21853] localedata: Fix abday Which looks same as day in zh_SG
887 [21854] localedata: Added New Locale en_SC
888 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
889 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
890 _dl_runtime_resolve_avx_slow
891 [21885] network: getaddrinfo: gethosts does not release resolver context
892 on memory allocation failure
893 [21899] libc: XPG4.2 sigaction namespace
894 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
895 the suspect)
896 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
897 default (i386)
898 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
899 errno value, causing getaddrinfo to fail
900 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
901 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
902 instead of EAI_NODATA
903 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
904 constant PTRACE_SEIZE_DEVEL
905 [21930] math: C-only gcc builtins used in <math.h> isinf
906 [21932] network: Unpaired __resolv_context_get in generic get*_r
907 implementation
908 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
909 [21944] libc: sigval namespace
910 [21951] localedata: Update hanzi collation by stroke
911 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
912 sysdeps/x86_64/fpu/e_expf.S
913 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
914 allocation in syscall loops)
915 [21959] localedata: Fix Country name for xh_ZA
916 [21960] localedata: Fix abmon for bem_ZM
917 [21966] math: AVX2 mathvec functions use FMA without checking
918 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
919 [21971] localedata: Added New Locale for mfe_MU
920 [21972] libc: assert macro requires operator== (int) for its argument type
921 [21973] math: [sparc] libm missing sqrtl compat symbol
922 [21974] libc: Remove __bb_init_func and __bb_exit_func
923 [21982] string: stratcliff.c: error: assuming signed overflow does not
924 occur with -O3
925 [21986] stdio: __guess_grouping is called incorrectly
926 [21987] math: [sparc32] wrong bits/long-double.h installed
927 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
928 amounts)
929 [22022] localedata: Missing country_name for mni_IN
930 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
931 niu_NZ
932 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
933 checks
934 [22026] locale: iconv_open: heap overflow on gconv_init failure
935 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
936 [22035] math: [m68k] bits/math-inline.h macro namespace
937 [22038] localedata: Fix abbreviated weeks and months for Somali
938 [22044] localedata: Remove redundant data for Limburgish Language
939 [22050] malloc: Linking with -lmcheck does not hook
940 __malloc_initialize_hook correctly
941 [22051] libc: zero terminator in the middle of glibc's .eh_frame
942 [22052] malloc: malloc failed to compile with GCC 7 and -O3
943 [22070] localedata: charmaps/UTF-8: wcwidth for
944 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
945 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
946 Jungseong and Jongseong) should be 0
947 [22078] nss: nss_files performance issue in multi mode
948 [22082] math: bits/math-finite.h exp10 condition
949 [22086] libc: pcprofiledump incorrect cross-endian condition
950 [22093] dynamic-link: ld.so no longer searches in .../x86_64
951 [22095] network: Name server address allocation memory leak in resolv.conf
952 parsing after OOM
953 [22096] network: __resolv_conf_attach can incorrectly free passed conf
954 object
955 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
956 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
957 e.g. ET_GNU_DEBUG_*
958 [22111] malloc: malloc: per thread cache is not returned when thread exits
959 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
960 [22134] libc: [linux] implement fexecve with execveat
961 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
962 ppc64le
963 [22145] libc: ttyname() gives up too early in the face of namespaces
964 [22146] math: C++ build issue with float128 on x86_64
965 [22153] nptl: nptl: save error code before process termination
966 [22156] libc: [hppa,ia64,microblaze] Executable stack default
967 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
968 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
969 [22165] libc: [hppa] Text relocations in libc.so
970 [22180] libc: destructor registered via __cxa_atexit is called twice
971 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
972 symlinks") cause "make" segfaults
973 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
974 math_force_eval
975 [22207] libc: FAIL: stdlib/test-atexit-race
976 [22225] math: nearbyint arithmetic moved before feholdexcept
977 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
978 [22235] math: iscanonical in C++ and float128
979 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
980 the svid compat wrapper
981 [22244] math: ynf and yn are wrong without the svid compat wrapper
982 [22273] libc: Improper assert in Linux posix_spawn implementation
983 [22284] libc: -pg -pie doesn't work
984 [22292] locale: localedef exits with error 4 when it should be error 1
985 [22294] locale: Allow "" for int_currency_symbol definition in locales.
986 [22295] locale: Don't warn on non-symbolic characters in locale sources in
987 --verbose.
988 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
989 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
990 to 2.26
991 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
992 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
993 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
994 [22322] libc: [mips64] wrong bits/long-double.h installed
995 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
996 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
997 (CVE-2017-15804)
998 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
999 [22343] malloc: Integer overflow in posix_memalign
1000 [22347] libc: getrandom() returns the number of bytes that were copied to
1001 the buffer even though the comments say "Return 0 on success and -1 on
1002 failure."
1003 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1004 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1005 -m32
1006 [22370] dynamic-link: Incorrect note padding check
1007 [22375] libc: malloc returns pointer from tcache_get when should return
1008 NULL (CVE-2017-17426)
1009 [22377] math: iseqsig, float128 and C++
1010 [22382] localedata: Error in tpi_PG locale
1011 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1012 inside the ASCII printable range
1013 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1014 -mlong-double-64
1015 [22403] localedata: Slash needs escaping in some locales
1016 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1017 [22409] network: res_hnok does not accept some host names used on the
1018 Internet
1019 [22412] network: res_dnok, res_hnok should perform syntax checks
1020 [22413] network: ns_name_pton ignores syntactically invalid trailing
1021 backslash
1022 [22415] stdio: setvbuf can lead to invalid free/segfault
1023 [22432] build: Non-deterministic build
1024 [22439] malloc: malloc_info should compute summary statistics for all sub-
1025 heaps in an arena
1026 [22442] network: if_nametoindex could report index for the wrong
1027 networking interface
1028 [22446] build: aliasing violation calling readlink in handle_request
1029 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1030 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1031 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1032 __nscd_hash/__nss_hash
1033 [22463] network: p_secstodate overflow handling
1034 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1035 [22478] libc: sigwait can fail with EINTR
1036 [22505] libc: ldconfig processes include directive in locale-specific
1037 order
1038 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1039 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1040 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1041 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1042 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1043 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1044 same as for Croatian
1045 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1046 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1047 [22577] libc: missing newline after "cannot allocate TLS data structures
1048 for initial thread"
1049 [22588] manual: manual/conf.texi: missing underscore in front of
1050 SC_SSIZE_MAX
1051 [22593] math: nextafter and nexttoward are declared with const attribute
1052 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1053 [22603] string: ia64 memchr overflows internal pointer check
1054 [22605] libc: SH clone does not set the exit code correctly
1055 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1056 (CVE-2017-1000408)
1057 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1058 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1059 modified in case of success
1060 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1061 [22615] manual: manual: ambiguous wording about errno value in case of
1062 success
1063 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1064 stack-protector=all
1065 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1066 binaries or if /proc is not mounted (CVE-2017-16997)
1067 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1068 [22630] build: $(no-pie-ldflag) is no longer effective
1069 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1070 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1071 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1072 [22637] nptl: guard size is subtracted from thread stack size instead of
1073 adding it on top
1074 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1075 [22657] localedata: hu_HU: Avoid double space in date
1076 [22660] math: fmax, fmin sNaN handling on alpha
1077 [22664] libc: New warning of GCC8
1078 [22665] math: alpha: ceil and floor raise inexact exceptions
1079 [22666] math: alpha: trunc raise inexact exceptions
1080 [22667] libc: makecontext lacks stack alignment on i386
1081 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1082 [22679] libc: getcwd(3) can succeed without returning an absolute path
1083 (CVE-2018-1000001)
1084 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1085 elision and tunables
1086 [22687] math: [powerpc-nofpu] complex long double functions spurious
1087 "invalid" exception
1088 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1089 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1090 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1091 exception
1092 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1093 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1094 power4
1095 [22701] nis: Incomplete removal of libnsl
1096 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1097 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1098 [22715] dynamic-link: FAIL: elf/tst-audit10
1099 [22719] libc: Backtrace tests fail on hppa
1100 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1101 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1102 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1103 before the first call to crypt_r () in crypt/badsalttest.c
1104
1105 \f
1106 Version 2.26
1107
1108 Major new features:
1109
1110 * A per-thread cache has been added to malloc. Access to the cache requires
1111 no locks and therefore significantly accelerates the fast path to allocate
1112 and free small amounts of memory. Refilling an empty cache requires locking
1113 the underlying arena. Performance measurements show significant gains in a
1114 wide variety of user workloads. Workloads were captured using a special
1115 instrumented malloc and analyzed with a malloc simulator. Contributed by
1116 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1117
1118 * Unicode 10.0.0 Support: Character encoding, character type info, and
1119 transliteration tables are all updated to Unicode 10.0.0, using
1120 generator scripts contributed by Mike FABIAN (Red Hat).
1121 These updates cause user visible changes, especially the changes in
1122 wcwidth for many emoji characters cause problems when emoji sequences
1123 are rendered with pango, see for example:
1124 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1125
1126 * Collation of Hungarian has been overhauled and is now consistent with "The
1127 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1128 Egmont Koblinger.
1129
1130 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1131
1132 - The GNU C Library will now detect when /etc/resolv.conf has been
1133 modified and reload the changed configuration. The new resolver option
1134 “no-reload” (RES_NORELOAD) disables this behavior.
1135
1136 - The GNU C Library now supports an arbitrary number of search domains
1137 (configured using the “search” directive in /etc/resolv.conf);
1138 previously, there was a hard limit of six domains. For backward
1139 compatibility, applications that directly modify the ‘_res’ global
1140 object are still limited to six search domains.
1141
1142 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1143 Library will now randomly pick a name server from the configuration as a
1144 starting point. (Previously, the second name server was always used.)
1145
1146 * The tunables feature is now enabled by default. This allows users to tweak
1147 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1148
1149 * New function reallocarray, which resizes an allocated block (like realloc)
1150 to the product of two sizes, with a guaranteed clean failure upon integer
1151 overflow in the multiplication. Originally from OpenBSD, contributed by
1152 Dennis Wölfing and Rüdiger Sonderfeld.
1153
1154 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1155 These are extended versions of preadv and pwritev, respectively, taking an
1156 additional flags argument. The set of supported flags depends on the
1157 running kernel; full support currently requires kernel 4.7 or later.
1158
1159 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1160 create a new session ID for the spawned process. This feature is
1161 scheduled to be added to the next major revision of POSIX; for the time
1162 being, it is available under _GNU_SOURCE.
1163
1164 * errno.h is now safe to use from C-preprocessed assembly language on all
1165 supported operating systems. In this context, it will only define the
1166 Exxxx constants, as preprocessor macros expanding to integer literals.
1167
1168 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1169 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1170 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1171 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1172
1173 To compile programs that use this feature, the compiler must support
1174 128-bit floating point with the type name _Float128 (as defined by TS
1175 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1176 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1177 must be defined to make the new interfaces visible.
1178
1179 The new functions and macros correspond to those present for other
1180 floating-point types (except for a few obsolescent interfaces not
1181 supported for the new type), with F128 or f128 suffixes; for example,
1182 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1183 printf or scanf formats for the new type; the strfromf128 and strtof128
1184 interfaces should be used instead.
1185
1186 Deprecated and removed features, and other changes affecting compatibility:
1187
1188 * The synchronization that pthread_spin_unlock performs has been changed to
1189 now be equivalent to a C11 atomic store with release memory order to the
1190 spin lock's memory location. Previously, several (but not all)
1191 architectures used stronger synchronization (e.g., containing what is
1192 often called a full barrier). This change can improve performance, but
1193 may affect odd fringe uses of spin locks that depend on the previous
1194 behavior (e.g., using spin locks as atomic variables to try to implement
1195 Dekker's mutual exclusion algorithm).
1196
1197 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1198 removed.
1199
1200 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1201 will only be built and installed when the GNU C Library is configured with
1202 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1203 as TIRPC or rpcsvc-proto, to be used.
1204
1205 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1206 libnss_compat, are deprecated, and will not be built or installed by
1207 default.
1208
1209 The NIS(+) support library, libnsl, is also deprecated. By default, a
1210 compatibility shared library will be built and installed, but not headers
1211 or development libraries. Only a few NIS-related programs require this
1212 library. (In particular, the GNU C Library has never required programs
1213 that use 'gethostbyname' to be linked with libnsl.)
1214
1215 Replacement implementations based on TIRPC, which additionally support
1216 IPv6, are available from <https://github.com/thkukuk/>. The configure
1217 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1218 name service modules, to be built and installed.
1219
1220 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1221 support is enabled, the configured recursive resolver must support EDNS.
1222 (Responding to EDNS-enabled queries with responses which are not
1223 EDNS-enabled is fine, but FORMERR responses are not.)
1224
1225 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1226 servers have not supported this opcode for a long time.
1227
1228 * The _res_opcodes variable has been removed from libresolv. It had been
1229 exported by accident.
1230
1231 * <string.h> no longer includes inline versions of any string functions,
1232 as this kind of optimization is better done by the compiler. The macros
1233 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1234
1235 * The nonstandard header <xlocale.h> has been removed. Most programs should
1236 use <locale.h> instead. If you have a specific need for the definition of
1237 locale_t with no other declarations, please contact
1238 libc-alpha@sourceware.org and explain.
1239
1240 * The obsolete header <sys/ultrasound.h> has been removed.
1241
1242 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1243
1244 * The obsolete function cfree has been removed. Applications should use
1245 free instead.
1246
1247 * The stack_t type no longer has the name struct sigaltstack. This changes
1248 the C++ name mangling for interfaces involving this type.
1249
1250 * The ucontext_t type no longer has the name struct ucontext. This changes
1251 the C++ name mangling for interfaces involving this type.
1252
1253 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1254 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1255 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1256 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1257 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1258 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1259 fpregset_t type no longer has the name struct fpu. This changes the C++
1260 name mangling for interfaces involving those types.
1261
1262 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1263 synced with the kernel:
1264
1265 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1266 are not supported on this architecture and have been removed.
1267
1268 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1269 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1270 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1271
1272 Programs that assume the GET/SETREGS ptrace requests are universally
1273 available will now fail to build, instead of malfunctioning at runtime.
1274
1275 Changes to build and runtime requirements:
1276
1277 * Linux kernel 3.2 or later is required at runtime, on all architectures
1278 supported by that kernel. (This is a change from version 2.25 only for
1279 x86-32 and x86-64.)
1280
1281 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1282
1283 * On most architectures, GCC 4.9 or later is required to build the GNU C
1284 Library. On powerpc64le, GCC 6.2 or later is required.
1285
1286 Older GCC versions and non-GNU compilers are still supported when
1287 compiling programs that use the GNU C Library. (We do not know exactly
1288 how old, and some GNU extensions to C may be _de facto_ required. If you
1289 are interested in helping us make this statement less vague, please
1290 contact libc-alpha@sourceware.org.)
1291
1292 Security related changes:
1293
1294 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1295 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1296
1297 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1298 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1299
1300 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1301 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1302 (CVE-2010-3192).
1303
1304 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1305 fixed (CVE-2017-12133).
1306
1307 The following bugs are resolved with this release:
1308
1309 [984] network: Respond to changed resolv.conf in gethostbyname
1310 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1311 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1312 fields
1313 [12189] libc: __stack_chk_fail should not attempt a backtrace
1314 (CVE-2010-3192)
1315 [14096] time: Race condition on timezone/tst-timezone.out
1316 [14172] localedata: az_IR: new locale
1317 [14995] build: glibc fails to build if gold is the default linker, even if
1318 ld.bfd is available
1319 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1320 [16637] network: inet_pton function is accepting IPv6 with bad format
1321 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1322 x86 and other generic code
1323 [16875] localedata: ko_KR: fix lang_name
1324 [17225] localedata: ar_SY: localized month names for May and June are
1325 incorrect
1326 [17297] localedata: da_DK: wrong date_fmt string
1327 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1328 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1329 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1330 [19066] localedata: ar_SA abbreviated day and month names are in English
1331 [19569] network: resolv: Support an arbitrary number of search domains
1332 [19570] network: Implement random DNS server selection in the stub
1333 resolver
1334 [19838] locale: localedef fails on PA-RISC
1335 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1336 order of 0D36 and 0D37
1337 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1338 chillu characters
1339 [20098] libc: FAIL: debug/backtrace-tst on hppa
1340 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1341 receiving data
1342 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1343 whitespace
1344 [20313] localedata: Update locale data to Unicode 9.0
1345 [20424] manual: Document how to provide a malloc replacement
1346 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
1347 for Peru
1348 [20686] locale: Add el_GR@euro to SUPPORTED.
1349 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1350 failures consistently
1351 [21015] dynamic-link: Document and fix --enable-bind-now
1352 [21016] nptl: pthread_cond support is broken on hppa
1353 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1354 frame-pointer on i386
1355 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1356 register
1357 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1358 [21088] libc: Build fails with --enable-static-nss
1359 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1360 generic c code is used
1361 [21109] libc: Tunables broken on big-endian
1362 [21112] math: powf has large ulp errors with base close to 1 and exponent
1363 around 4000
1364 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1365 (CVE-2017-12133)
1366 [21120] malloc: glibc malloc is incompatible with GCC 7
1367 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1368 with -lieee
1369 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1370 (-0.0) when linking with -lieee
1371 [21171] math: log10, log2 and lgamma return incorrect results
1372 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1373 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1374 [21207] localedata: ce_RU: update weekdays from CLDR
1375 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1376 [21217] localedata: Update months from CLDR-31
1377 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1378 [21243] libc: support_delete_temp_file should issue warning for failed
1379 remove()
1380 [21244] libc: support resolv_test_start() socket fd close should be
1381 checked for errors.
1382 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1383 new posix_spawn implementation
1384 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1385 leads to lower CPU frequency
1386 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1387 [21261] libc: [sparc64] bits/setjmp.h namespace
1388 [21267] network: [mips] bits/socket.h IOC* namespace
1389 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1390 [21270] libc: mmap64 silently truncates large offset values
1391 [21275] libc: posix_spawn always crashes on ia64 now
1392 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1393 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1394 [21289] libc: Incorrect declaration for 32-bit platforms with
1395 _FILE_OFFSET_BITS=64 causes build error
1396 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1397 not support gethostbyname4_r
1398 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1399 switching
1400 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1401 arenas
1402 [21340] libc: Support POSIX_SPAWN_SETSID
1403 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1404 [21359] network: ns_name_pack needs additional byte in destination buffer
1405 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1406 against fragmentation attacks (CVE-2017-12132)
1407 [21369] network: resolv: Remove EDNS fallback
1408 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1409 and _POSIX_C_SOURCE
1410 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1411 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1412 [21393] stdio: Missing dup3 error check in freopen, freopen64
1413 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1414 [21399] localedata: Bad description for U00EC in
1415 localedata/charmaps/CP1254
1416 [21411] malloc: realloc documentation error
1417 [21426] network: sys/socket.h uio.h namespace
1418 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1419 [21445] libc: signal.h bsd_signal namespace
1420 [21455] network: Network headers stdint.h namespace
1421 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1422 retry value
1423 [21475] network: resolv: Overlong search path is truncated mid-label
1424 [21511] libc: sigstack namespace
1425 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1426 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1427 bits/syscall.h: No such file or directory
1428 [21517] libc: struct sigaltstack namespace
1429 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1430 [21533] localedata: Update locale data to Unicode 10.0
1431 [21537] libc:
1432 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1433 junk at end of line, first unrecognized character is `@'
1434 [21538] libc: SIG_HOLD missing for XPG4
1435 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1436 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1437 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1438 [21543] libc: sigevent namespace
1439 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1440 size for general purpose registers in mcontext_t structure
1441 [21550] libc: sigwait namespace
1442 [21552] libc: XPG4 bsd_signal namespace
1443 [21554] libc: sigpause namespace
1444 [21560] libc: sys/wait.h signal.h namespace
1445 [21561] libc: waitid namespace
1446 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1447 '/usr/include/stdlib.h': Permission denied
1448 [21575] libc: sys/wait.h missing struct rusage definition
1449 [21584] libc: sigaltstack etc namespace
1450 [21597] libc: siginterrupt namespace
1451 [21607] math: hppa: FAIL: math/test-tgmath
1452 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1453 issue on x86-64
1454 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1455 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1456 stack and heap (CVE-2017-1000366)
1457 [21625] libc: wait3 namespace
1458 [21654] nss: Incorrect pointer alignment in NSS group merge result
1459 construction
1460 [21657] network: Parse interface zone id for node-local multicast
1461 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1462 between_2_3
1463 [21666] libc: .symver is used on common symbol
1464 [21668] network: resolv: res_init cross-thread broadcast introduces race
1465 conditions
1466 [21687] math: tgmath.h totalorder, totalordermag return type
1467 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1468 Hindi Locale
1469 [21696] libc: Incorrect assumption of of __cpu_mask in
1470 posix/sched_cpucount.c
1471 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1472 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1473 locale
1474 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1475 leading to relocation crash
1476 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1477 `resolv_conf_matches (resp, conf)' failed.
1478 [21710] localedata: Added Samoan language locale for Samoa
1479 [21711] localedata: Pashto yesstr/nostr locale are missing
1480 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1481 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1482 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1483 [21723] localedata: yesstr/nostr missing for Chinese language locale
1484 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1485 locale
1486 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1487 locale
1488 [21728] localedata: New Locale for Tongan language
1489 [21729] localedata: incorrect LC_NAME fields for hi_IN
1490 [21733] localedata: yesstr and nostr are missing for zh_HK
1491 [21734] localedata: Missing yesstr and nostr are for kw_GB
1492 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1493 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1494 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1495 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1496 strings
1497 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1498 protector=all
1499 [21749] localedata: Wrong abbreviated day name (“abday”) for
1500 ar_JO/ar_LB/ar_SY
1501 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1502 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1503 [21759] localedata: missing yesstr and nostr for Tigrinya
1504 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1505 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1506 [21767] localedata: Missing Bislama locales
1507 [21768] localedata: Missing yesstr and nostr for aa_ET
1508 [21770] localedata: Missing Field in li_NL
1509 [21778] nptl: Robust mutex may deadlock
1510 [21779] libc: MicroBlaze segfaults when loading libpthread
1511 [21783] localedata: Fix int_select international_call_prefixes
1512 [21784] localedata: Inconsistency in country_isbn
1513 [21788] localedata: Missing Country Postal Abbreviations
1514 [21794] localedata: Added-country_isbn-for-Italy
1515 [21795] localedata: Add/Fix country_isbn for France
1516 [21796] localedata: Added country_isbn for Republic of Korea
1517 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1518 [21799] localedata: Added int_select international_call_prefixes
1519 [21801] localedata: Added int_select international_call_prefixes
1520 [21804] nptl: Double semicolon in thread-shared-types.h
1521 [21807] localedata: LC_ADDRESS fix for pap_CW
1522 [21808] localedata: Fix LC_ADDRESS for pap_AW
1523 [21821] localedata: Added country_name in mai_IN
1524 [21822] localedata: Fix LC_TIME for mai_IN
1525 [21823] localedata: missing yesstr, nostr for sa_IN
1526 [21825] localedata: Fix name_mrs for mag_IN
1527 [21828] localedata: 2.26 changelog should mention user visible changes
1528 with unicode 9.0
1529 [21835] localedata: Added Maithili language locale for Nepal
1530 [21838] localedata: Removed redundant data for the_NP
1531 [21839] localedata: Fix LC_MONETARY for ta_LK
1532 [21844] localedata: Fix Latin characters and Months Sequence.
1533 [21848] localedata: Fix mai_NP Title Name
1534
1535 \f
1536 Version 2.25
1537
1538 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1539 24731-2:2010, is supported to enable declarations of functions from that
1540 TR. Note that not all functions from that TR are supported by the GNU C
1541 Library.
1542
1543 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1544 18661-1:2014, is supported to enable declarations of functions and macros
1545 from that TS. Note that not all features from that TS are supported by
1546 the GNU C Library.
1547
1548 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1549 18661-4:2015, is supported to enable declarations of functions and macros
1550 from that TS. Note that most features from that TS are not supported by
1551 the GNU C Library.
1552
1553 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1554 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1555 Since the GNU C Library defaults to a much newer revision of POSIX, this
1556 will only affect programs that specifically request an old conformance
1557 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1558 see a change in the visible declarations, but a program compiled with
1559 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1560 will not.
1561
1562 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1563 defined by all multithreaded code, but glibc has not required this for
1564 many years.
1565
1566 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1567 means that in a future release, the macros “major”, “minor”, and “makedev”
1568 will only be available from <sys/sysmacros.h>.
1569
1570 These macros are not part of POSIX nor XSI, and their names frequently
1571 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1572 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1573 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1574 problem.
1575
1576 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1577 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1578 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1579
1580 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1581 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1582 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1583 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1584 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1585 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1586 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1587 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1588 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1589 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1590 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1591 WINT_WIDTH.
1592
1593 * New <math.h> features are added from TS 18661-1:2014:
1594
1595 - Signaling NaN macros: SNANF, SNAN, SNANL.
1596
1597 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1598 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1599 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1600
1601 - llogb functions: the llogb, llogbf and llogbl functions, and the
1602 FP_LLOGB0 and FP_LLOGBNAN macros.
1603
1604 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1605 fminmagf, fminmagl.
1606
1607 - Comparison macros: iseqsig.
1608
1609 - Classification macros: iscanonical, issubnormal, iszero.
1610
1611 - Total order functions: totalorder, totalorderf, totalorderl,
1612 totalordermag, totalordermagf, totalordermagl.
1613
1614 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1615
1616 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1617 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1618
1619 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1620 are added to libc. They convert a floating-point number into string.
1621
1622 * Most of glibc can now be built with the stack smashing protector enabled.
1623 It is recommended to build glibc with --enable-stack-protector=strong.
1624 Implemented by Nick Alcock (Oracle).
1625
1626 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1627 intended to be used instead of memset() to erase sensitive data after use;
1628 the compiler will not optimize out calls to explicit_bzero even if they
1629 are "unnecessary" (in the sense that no _correct_ program can observe the
1630 effects of the memory clear).
1631
1632 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1633 to float instead of double. This does not affect the ABI of any libraries
1634 that are part of the GNU C Library, but may affect the ABI of other
1635 libraries that use this type in their interfaces.
1636
1637 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1638 float_t and double_t types are now defined to long double instead of float
1639 and double. These options are not the default, and this does not affect
1640 the ABI of any libraries that are part of the GNU C Library, but it may
1641 affect the ABI of other libraries that use this type in their interfaces,
1642 if they are compiled or used with those options.
1643
1644 * The getentropy and getrandom functions, and the <sys/random.h> header file
1645 have been added.
1646
1647 * The buffer size for byte-oriented stdio streams is now limited to 8192
1648 bytes by default. Previously, on Linux, the default buffer size on most
1649 file systems was 4096 bytes (and thus remains unchanged), except on
1650 network file systems, where the buffer size was unpredictable and could be
1651 as large as several megabytes.
1652
1653 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1654 for the Linux quota interface which predates kernel version 2.4.22 has
1655 been removed.
1656
1657 * The malloc_get_state and malloc_set_state functions have been removed.
1658 Already-existing binaries that dynamically link to these functions will
1659 get a hidden implementation in which malloc_get_state is a stub. As far
1660 as we know, these functions are used only by GNU Emacs and this change
1661 will not adversely affect already-built Emacs executables. Any undumped
1662 Emacs executables, which normally exist only during an Emacs build, should
1663 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1664
1665 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1666 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1667 “no-ip6-dotint” had already been the default, and support for the
1668 “ip6-dotint” option was removed from the Internet in 2006.
1669
1670 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1671 flag from <resolv.h> have been removed. The option relied on a
1672 backwards-incompatible DNS extension which was never deployed on the
1673 Internet.
1674
1675 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1676 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1677 They were already unimplemented.
1678
1679 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1680 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1681 removed again from the IETF name lookup interface specification in RFC
1682 2553. Applications should use getaddrinfo instead.
1683
1684 * DNSSEC-related declarations and definitions have been removed from the
1685 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1686 decode the data part of DNSSEC record types. Previous versions of glibc
1687 only implemented minimal support for the previous version of DNSSEC, which
1688 is incompatible with the currently deployed version.
1689
1690 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1691 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1692 <arpa/nameser.h> header file because the distinction between RR types and
1693 meta-RR types is not officially standardized, subject to revision, and
1694 thus not suitable for encoding in a macro.
1695
1696 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1697 and rhook members of the res_state type in <resolv.h> have been removed.
1698 The glibc stub resolver did not support these hooks, but the header file
1699 did not reflect that.
1700
1701 * For multi-arch support it is recommended to use a GCC which has
1702 been built with support for GNU indirect functions. This ensures
1703 that correct debugging information is generated for functions
1704 selected by IFUNC resolvers. This support can either be enabled by
1705 configuring GCC with '--enable-gnu-indirect-function', or by
1706 enabling it by default by setting 'default_gnu_indirect_function'
1707 variable for a particular architecture in the GCC source file
1708 'gcc/config.gcc'.
1709
1710 * GDB pretty printers have been added for mutex and condition variable
1711 structures in POSIX Threads. When installed and loaded in gdb these pretty
1712 printers show various pthread variables in human-readable form when read
1713 using the 'print' or 'display' commands in gdb.
1714
1715 * Tunables feature added to allow tweaking of the runtime for an application
1716 program. This feature can be enabled with the '--enable-tunables' configure
1717 flag. The GNU C Library manual has details on usage and README.tunables has
1718 instructions on adding new tunables to the library.
1719
1720 * A new version of condition variables functions have been implemented in
1721 the NPTL implementation of POSIX Threads to provide stronger ordering
1722 guarantees.
1723
1724 * A new version of pthread_rwlock functions have been implemented to use a more
1725 scalable algorithm primarily through not using a critical section anymore to
1726 make state changes.
1727
1728 Security related changes:
1729
1730 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1731 have been created with makecontext could fail to terminate due to a
1732 missing .cantunwind annotation. This has been observed to lead to a hang
1733 (denial of service) in some Go applications compiled with gccgo. Reported
1734 by Andreas Schwab. (CVE-2016-6323)
1735
1736 * The DNS stub resolver functions would crash due to a NULL pointer
1737 dereference when processing a query with a valid DNS question type which
1738 was used internally in the implementation. The stub resolver now uses a
1739 question type which is outside the range of valid question type values.
1740 (CVE-2015-5180)
1741
1742 The following bugs are resolved with this release:
1743
1744 [4099] stdio: Overly agressive caching by stream i/o functions.
1745 [7065] build: Support building glibc with -fstack-protector or -fstack-
1746 protector-all
1747 [9842] localedata: en_CA: incorrect date format
1748 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1749 before it started waiting
1750 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1751 [15765] nptl: sem_open is wrongly a cancellation point
1752 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1753 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1754 wrong type
1755 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1756 library linked with pthread
1757 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1758 "FIXME: Ingo" issue)
1759 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1760 [17252] libc: getrandom and getentropy syscall
1761 [17730] malloc: thread-local storage is sometimes improperly free()'d
1762 after being __libc_memalign()'d
1763 [18241] stdio: failed fseek on memstream does not set errno and can fail
1764 when it shouldnt
1765 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1766 not
1767 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1768 PTHREAD_PRIO_INHERIT mutex on ARM
1769 [18784] network: res_query and related function crash for special record
1770 type queries (CVE-2015-5180)
1771 [19380] math: strtod does not raise "inexact"
1772 [19387] string: Integer overflow in memchr
1773 [19390] string: Integer overflow in strncat
1774 [19398] build: linknamespace tests fail with massively parallel build
1775 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1776 termination
1777 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1778 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1779 [19514] libc: [PATCH] Fix spelling errors in spelling
1780 "implement"/"implementation" in several places
1781 [19582] network: Deprecate RES_USE_INET6
1782 [19673] manual: clog10 docs appear to be erroneous
1783 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1784 causes a segmentation fault
1785 [19826] libc: invalid pointer returned from __tls_get_addr with static
1786 linking
1787 [20016] network: resolv: Remove hooks support from the API
1788 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1789 uninitialized GOT
1790 [20033] math: [x86_64] vectorized math function don't call the __finite
1791 versions
1792 [20116] nptl: use after free in pthread_create
1793 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1794 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1795 is always true.
1796 [20311] nptl: please install proc_service.h
1797 [20366] libc: Compilation errors in installed headers in strict-compliance
1798 modes
1799 [20370] malloc: malloc: Arena free list management is still racy
1800 (incorrect fix in bug 19243)
1801 [20386] libc: assert (X = 0) does not result in GCC warning
1802 [20432] malloc: malloc: Minimize interface required for interposition
1803 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1804 in _Unwind_Backtrace (CVE-2016-6323)
1805 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1806 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1807 libpthread.a
1808 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1809 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1810 [20477] network: resolv: incorrect double-checked locking related to
1811 _res_hconf
1812 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1813 information.
1814 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1815 penalty
1816 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1817 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1818 cause transition penalty
1819 [20517] math: sparcv9 missing fdiml compat symbol
1820 [20524] manual: strverscmp is inconsistent
1821 [20525] libc: <sys/quota.h> should be based on kernel headers
1822 [20539] math: GCC 7 -static -lm fails to link at -O3
1823 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1824 mcount.oS)
1825 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1826 Checking
1827 [20591] network: Remove obsolete DNSSEC support
1828 [20592] network: DNS resource record type classification macros in
1829 <arpa/nameser.h> are incorrect
1830 [20593] network: Update DNS RR type definitions
1831 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1832 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1833 multi-arch
1834 [20629] network: libresolv: Remove support for bitstring labels
1835 (RES_USEBSTRING)
1836 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1837 assertion: "maxidx >= 2"
1838 [20660] math: [arm] Use VSQRT
1839 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1840 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1841 [20689] libc: Test for FMA should also check for AVX.
1842 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1843 [20715] math: iszero macro breaks existing code
1844 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1845 [20728] libc: powerpc: Missing TOC stub in clone
1846 [20729] build: glibc-2.24 fails to build for i486 with -Os
1847 [20750] build: Build fails with default PIE enabled gcc-6.x
1848 [20768] math: [alpha] sqrt fegetenv namespace
1849 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1850 glibc
1851 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1852 __FLT_EVAL_METHOD__ is 2
1853 [20790] network: rpcgen buffer overrun in get_prog_declaration
1854 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1855 [20829] libc: crypt snprintf namespace
1856 [20847] libc: tst-vfork3 failure
1857 [20855] math: Default bits/mathdef.h has inappropriate float_t
1858 [20859] math: [sh4] FP_ILOGB0 invalid
1859 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1860 U+20AC), not same as GBK
1861 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1862 [20916] math: pow handling of sNaN arguments
1863 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1864 [20919] math: Bad pow (qNaN, 0) result with -lieee
1865 [20940] math: hypot sNaN handling
1866 [20947] math: fmax, fmin sNaN handling
1867 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1868 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1869 RES_USE_INET6
1870 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1871 [20973] nptl: robust mutexes: Lost wake-ups
1872 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1873 [20978] nis: strlen on null pointer in nss_nisplus
1874 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1875 wrong condition
1876 [21014] string: i686 memchr overflows internal pointer check
1877 [21019] libc: [mips] n32 lseek incorrect on overflow
1878 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1879 [21026] libc: [MIPS] readahead syscall is broken on n64
1880 [21028] math: Fallback fesetexceptflag should always succeed
1881 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1882 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1883 clang
1884 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1885 [21061] librt: [microblaze] librt lost clock_* exports
1886 [21073] libc: tunables: insecure environment variables passed to
1887 subprocesses with AT_SECURE
1888 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1889 \f
1890 Version 2.24
1891
1892 * The minimum Linux kernel version that this version of the GNU C Library
1893 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1894 kernel version 2.6.32 or later suffices (on architectures that already
1895 required kernel versions more recent than 3.2, those requirements remain
1896 unchanged). Linux 3.2 or later kernel headers are required on all
1897 architectures.
1898
1899 * The pap_AN locale has been deleted. This has been deprecated for a long
1900 time. It has been replaced by pap_AW & pap_CW, both of which have long
1901 been included in previous releases.
1902
1903 * The readdir_r and readdir64_r functions have been deprecated. It is
1904 recommended to use readdir and readdir64 instead.
1905
1906 * The type “union wait” has been removed. It was deprecated in the early
1907 1990s and never part of POSIX. Application code should use the int type
1908 instead of “union wait”.
1909
1910 * A new NSS action is added to facilitate large distributed system
1911 administration. The action, MERGE, allows remote user stores like LDAP
1912 to be merged into local user stores like /etc/groups in order to provide
1913 easy to use, updated, and managed sets of merged credentials. The new
1914 action can be used by configuring it in /etc/nsswitch.conf:
1915 group: files [SUCCESS=merge] nis
1916 Implemented by Stephen Gallagher (Red Hat).
1917
1918 * The deprecated __malloc_initialize_hook variable has been removed from the
1919 API.
1920
1921 * The long unused localedef --old-style option has been removed. It hasn't
1922 done anything in over 16 years. Scripts using this option can safely
1923 drop it.
1924
1925 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1926 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1927 return the next representable value in the direction of positive infinity
1928 and the nextdown functions return the next representable value in the
1929 direction of negative infinity. These are currently enabled as GNU
1930 extensions.
1931
1932 Security related changes:
1933
1934 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1935 could result in a stack overflow when getnetbyname was called with an
1936 overly long name. (CVE-2016-3075)
1937
1938 * Previously, getaddrinfo copied large amounts of address data to the stack,
1939 even after the fix for CVE-2013-4458 has been applied, potentially
1940 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1941 instead. Reported by Michael Petlan. (CVE-2016-3706)
1942
1943 * The glob function suffered from a stack-based buffer overflow when it was
1944 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1945 Reported by Alexander Cherepanov. (CVE-2016-1234)
1946
1947 * The Sun RPC UDP client could exhaust all available stack space when
1948 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1949 alloca plugin for GCC. (CVE-2016-4429)
1950
1951 * The IPv6 name server management code in libresolv could result in a memory
1952 leak for each thread which is created, performs a failing naming lookup,
1953 and exits. Over time, this could result in a denial of service due to
1954 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1955
1956 The following bugs are resolved with this release:
1957
1958 [1170] localedata: ne_NP: update Nepali locale definition file
1959 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1960 of MS-DOS.
1961 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1962 [6796] math: fdim() does not set errno on overflow
1963 [10354] libc: posix_spawn should use vfork() in more cases than presently
1964 [11213] localedata: localedata: add copyright disclaimer to locale files
1965 [12143] localedata: chr_US: new Cherokee locale
1966 [12450] localedata: sgs_LT: new locale
1967 [12676] localedata: ln_CD: new locale
1968 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1969 CLDR data
1970 [13304] math: fma, fmaf, fmal produce wrong results
1971 [14259] build: --localedir arg to configure is ignored
1972 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1973 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1974 [14934] localedata: es_CL: wrong first weekday chilean locale
1975 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1976 romanisation
1977 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1978 and +/-
1979 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1980 [15368] nptl: raise() is not async-signal-safe
1981 [15479] math: ceil, floor, round and trunc raise inexact exception
1982 [15578] localedata: kk_KZ: various updates
1983 [16003] localedata: pap_AN: punt old locale
1984 [16137] localedata: iw_IL: punt old locale
1985 [16190] localedata: eo: new esperanto locale
1986 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1987 [16742] malloc: race condition: pthread_atfork() called before first
1988 malloc() results in unexpected locking behaviour/deadlocks
1989 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1990 all locales
1991 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1992 [17565] localedata: pt_PT: wrong (work-)week start
1993 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1994 [17950] build: Build fails with -msse
1995 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1996 [18433] libc: posix_spawn does not return correctly upon failure to
1997 execute
1998 [18453] localedata: charmaps/IBM875: incorrect codes
1999 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2000 -Wsystem-headers
2001 [18896] localedata: he_IL: improvements for currency
2002 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2003 Romanian locale data
2004 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2005 symbol
2006 [19056] libc: Deprecate readdir_r
2007 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2008 language
2009 [19198] localedata: nl_NL: small improvements for Dutch locales
2010 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2011 nameservers (CVE-2016-5417)
2012 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2013 machine
2014 [19400] locale: Language missing in "iso-639.def", trivial fix in
2015 description
2016 [19431] malloc: Deadlock between fflush, getdelim, and fork
2017 [19505] libc: Incorrect file descriptor validity checks in
2018 posix_spawn_file_actions_add{open,close,dup2}
2019 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2020 when using RTLD_NEXT
2021 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2022 `intl/{gettextP,loadinfo}.h'
2023 [19534] libc: execle, execlp may use malloc
2024 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2025 [19573] network: res_nclose and __res_maybe_init disagree about name
2026 server initialization, breaking Hesiod
2027 [19575] localedata: Status of GB18030 tables
2028 [19581] localedata: sr_* date_fmt string contains additional newline
2029 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2030 Excavator core
2031 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2032 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2033 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2034 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2035 double range
2036 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2037 part zero incorrect
2038 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2039 equality tests
2040 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2041 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2042 [19633] locale: strfmon_l applies global locale to number formatting
2043 [19642] network: Memory leak in getnameinfo
2044 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2045 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2046 glibc-2.22
2047 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2048 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2049 'tst-numeric.c'
2050 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2051 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2052 low part
2053 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2054 result
2055 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2056 unaligned stack
2057 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2058 pointers and lengths in error-case.
2059 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2060 report errors on UTF-16 surrogates values.
2061 [19755] nscd: nscd assertion failure in gc
2062 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2063 [19759] libc: mempcpy shouldn't be inlined
2064 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2065 [19765] libc: s390 needs an optimized mempcpy
2066 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2067 NAME_MAX limit assumption (CVE-2016-1234)
2068 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2069 [19787] network: Missing and incorrect truncation checks in getnameinfo
2070 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2071 modes
2072 [19791] network: Assertion failure in res_query.c with un-connectable name
2073 server addresses
2074 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2075 [19822] math: libm.so install clobbers old version
2076 [19825] network: resolv: send_vc can return uninitialized data in second
2077 response to getaddrinfo
2078 [19830] network: nss_dns: should check RDATA length against buffer length
2079 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2080 confronted with A/AAAA records of invalid size
2081 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2082 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2083 [19853] stdio: Printing IBM long double in decimal with high precision is
2084 sometimes incorrect
2085 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2086 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2087 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2088 record types
2089 [19865] network: Assertion failure or memory leak in
2090 _nss_dns_getcanonname_r
2091 [19868] network: nss_dns: netent code does not skip over non-PTR records
2092 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2093 (CVE-2016-3075)
2094 [19881] string: Improve x86-64 memset
2095 [19907] string: Incorrect memcpy tests
2096 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2097 resolving symbols
2098 [19925] libc: termios.h XCASE namespace
2099 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2100 [19929] libc: limits.h NL_NMAX namespace
2101 [19931] stdio: Memory leak in vfprintf
2102 [19957] libc: clone(CLONE_VM) access invalid parent memory
2103 [19963] localedata: en_IL: New locale
2104 [19989] stdio: stdio.h cuserid namespace
2105 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2106 gethosts
2107 [19996] locale: langinfo.h nl_langinfo_l namespace
2108 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2109 [20010] network: getaddrinfo: Stack overflow in hostent translation
2110 (CVE-2016-3706)
2111 [20012] stdio: libio: fmemopen append mode failure
2112 [20014] stdio: stdio.h namespace for pre-threads POSIX
2113 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2114 [20023] libc: fcntl.h timespec namespace
2115 [20024] math: [x86_64] vectorized sincos trashes the stack
2116 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2117 [20041] time: sys/time.h timespec namespace
2118 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2119 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2120 [20051] libc: ttyslot in wrong header under wrong conditions
2121 [20054] libc: gethostname not declared for XPG4
2122 [20055] libc: termios.h missing tcgetsid for XPG4
2123 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2124 executable
2125 [20073] libc: sys/stat.h fchmod namespace
2126 [20074] libc: stdlib.h rand_r namespace
2127 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2128 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2129 XPG3
2130 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2131 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2132 (CVE-2016-4429)
2133 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2134 [20119] libc: Wrong mask for processors level type from CPUID
2135 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2136 AS not supporting AVX512
2137 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2138 sNaN argument
2139 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2140 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2141 argument
2142 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2143 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2144 [20178] libc: posix_spawn{p} should not call exit
2145 [20191] stdio: libio: vtables hardening
2146 [20195] string: FMA4 detection requires CPUID execution with register
2147 eax=0x80000001
2148 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2149 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2150 subnormals
2151 [20212] math: acos (sNaN) returns sNaN
2152 [20213] math: asin (sNaN) returns sNaN
2153 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2154 [20218] math: [i386] asinhl (sNaN) returns sNaN
2155 [20219] math: [i386] atanhl (sNaN) returns sNaN
2156 [20222] stdio: fopencookie: Mangle function pointers
2157 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2158 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2159 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2160 input
2161 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2162 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2163 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2164 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2165 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2166 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2167 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2168 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2169 [20240] math: modf (sNaN) returns sNaN
2170 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2171 [20250] math: frexp (sNaN) returns sNaN
2172 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2173 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2174 double rounding
2175 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2176 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2177 subscript is below array bounds [-Werror=array-bounds]
2178 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2179 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2180 (Only arm/linux)
2181 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2182 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2183 fallbacks
2184 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2185 "invalid" exceptions
2186 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2187 [20316] localedata: id_ID: Februari instead of Pebruari
2188 [20327] string: POWER8 strcasecmp returns incorrect result
2189 [20347] math: Failure: Test: j0_downward (0xap+0)
2190 [20348] libc: FAIL: misc/tst-preadvwritev64
2191 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2192 [20350] libc: There is no test for p{read,write}64
2193 [20357] math: Incorrect cos result for 1.5174239687223976
2194 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2195 \f
2196 Version 2.23
2197
2198 * Unicode 8.0.0 Support: Character encoding, character type info, and
2199 transliteration tables are all updated to Unicode 8.0.0, using new
2200 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2201 These updates cause user visible changes, such as the fixes for bugs
2202 89, 16061, and 18568.
2203
2204 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2205 kernel-internal CPU set size. This means that requests that change the
2206 CPU affinity which failed before (for example, an all-ones CPU mask) will
2207 now succeed. Applications that need to determine the effective CPU
2208 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2209 setting it because the kernel can adjust it (and the previous size check
2210 would not detect this in the majority of cases).
2211
2212 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2213 the following new symbols are used: fts64_children, fts64_close,
2214 fts64_open, fts64_read and fts64_set.
2215
2216 * getaddrinfo now detects certain invalid responses on an internal netlink
2217 socket. If such responses are received, an affected process will
2218 terminate with an error message of "Unexpected error <number> on netlink
2219 descriptor <number>" or "Unexpected netlink response of size <number> on
2220 descriptor <number>". The most likely cause for these errors is a
2221 multi-threaded application which erroneously closes and reuses the netlink
2222 file descriptor while it is used by getaddrinfo.
2223
2224 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2225 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2226 unnecessary serialization of memory allocation requests across threads.
2227 The defect is now corrected. Users should see a substantial increase in
2228 the concurent throughput of allocation requests for applications which
2229 trigger this bug. Affected applications typically create and
2230 destroy threads frequently. (Bug 19048 was reported and analyzed by
2231 Ericsson.)
2232
2233 * There is now a --disable-timezone-tools configure option for disabling the
2234 building and installing of the timezone related utilities (zic, zdump, and
2235 tzselect). This is useful for people who build the timezone data and code
2236 independent of the GNU C Library.
2237
2238 * The obsolete header <regexp.h> has been removed. Programs that require
2239 this header must be updated to use <regex.h> instead.
2240
2241 * The obsolete functions bdflush, create_module, get_kernel_syms,
2242 query_module and uselib are no longer available to newly linked binaries;
2243 the header <sys/kdaemon.h> has been removed. These functions and header
2244 were specific to systems using the Linux kernel and could not usefully be
2245 used with the GNU C Library on systems with version 2.6 or later of the
2246 Linux kernel.
2247
2248 * Optimized string, wcsmbs and memory functions for IBM z13.
2249 Implemented by Stefan Liebler.
2250
2251 * Newly linked programs that define a variable called signgam will no longer
2252 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2253 require signgam to be set by those functions must ensure that they use the
2254 variable provided by the GNU C Library and declared in <math.h>, without
2255 defining their own copy.
2256
2257 * The minimum GCC version that can be used to build this version of the GNU
2258 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2259 still be used to compile programs using the GNU C Library.
2260
2261 Security related changes:
2262
2263 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2264 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2265
2266 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2267 the pointer guard feature. It is always enabled. Previously,
2268 LD_POINTER_GUARD could be used to disable security hardening in binaries
2269 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2270 (CVE-2015-8777)
2271
2272 * An integer overflow in hcreate and hcreate_r could lead to an
2273 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2274
2275 * The catopen function no longer has unbounded stack usage. Reported by
2276 Max. (CVE-2015-8779)
2277
2278 * The nan, nanf and nanl functions no longer have unbounded stack usage
2279 depending on the length of the string passed as an argument to the
2280 functions. Reported by Joseph Myers. (CVE-2014-9761)
2281
2282 * A stack-based buffer overflow was found in libresolv when invoked from
2283 libnss_dns, allowing specially crafted DNS responses to seize control
2284 of execution flow in the DNS client. The buffer overflow occurs in
2285 the functions send_dg (send datagram) and send_vc (send TCP) for the
2286 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2287 family. The use of AF_UNSPEC triggers the low-level resolver code to
2288 send out two parallel queries for A and AAAA. A mismanagement of the
2289 buffers used for those queries could result in the response of a query
2290 writing beyond the alloca allocated buffer created by
2291 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2292 the overflow. Thanks to the Google Security Team and Red Hat for
2293 reporting the security impact of this issue, and Robert Holiday of
2294 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2295
2296 The following bugs are resolved with this release:
2297
2298 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
2299 [887] math: Math library function "logb" and "nextafter" inconsistent
2300 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2301 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2302 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2303 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2304 use `mkstemp'
2305 [4404] localedata: German translation of "Alarm clock" is misleading
2306 [6799] math: nextafter() and nexttoward() doen't set errno on
2307 overflow/underflow errors
2308 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2309 overflow/underflow
2310 [10432] nis: _nss_nis_setnetgrent assertion failure
2311 [11460] libc: fts has no LFS support
2312 [12926] network: getaddrinfo()/make_request() may spin forever
2313 [13065] nptl: Race condition in pthread barriers
2314 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2315 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2316 are not contiguous
2317 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2318 [14912] libc: Rename non-installed bits/*.h headers
2319 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2320 [15367] math: Let gcc use __builtin_isinf
2321 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2322 [15421] math: lgamma wrongly sets signgam for ISO C
2323 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2324 FE_INVALID with argument out of range
2325 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2326 all exceptions
2327 [15786] dynamic-link: ifunc resolver functions can smash function
2328 arguments
2329 [15918] math: Unnecessary check for equality in hypotf()
2330 [16061] localedata: Review / update transliteration data
2331 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2332 should include
2333 [16141] time: strptime %z offset restriction
2334 [16171] math: drem should be alias of remainder
2335 [16296] math: fegetround is pure?
2336 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2337 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2338 GNU/Linux
2339 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2340 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2341 arguments
2342 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2343 "Exception "Inexact" set" on ppc32
2344 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2345 [16517] math: Missing underflow exception from tanf/tan/tanl
2346 [16519] math: Missing underflow exception from sinhf
2347 [16520] math: Missing underflow exception from tanhf
2348 [16521] math: Missing underflow exception from exp2
2349 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2350 rounding results
2351 [16734] stdio: fopen calls mmap to allocate its buffer
2352 [16961] math: nan function incorrect handling of bad sequences
2353 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2354 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2355 [16985] locale: localedef: confusing error message when opening output
2356 fails
2357 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2358 [17197] locale: Redundant shift character in iconv conversion output at
2359 block boundary
2360 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2361 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2362 [17250] dynamic-link: static linking breaks nss loading
2363 (getaddrinfo/getpwnam/etc...)
2364 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2365 4.7?
2366 [17441] math: isnan() should use __builtin_isnan() in GCC
2367 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2368 (related to lock elision)
2369 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2370 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2371 [17887] time: strptime should be able to parse "+01:00" style timezones
2372 [17905] libc: catopen() Multiple unbounded stack allocations
2373 (CVE-2015-8779)
2374 [18084] libc: backtrace (..., 0) dumps core on x86
2375 [18086] libc: nice() sets errno to 0 on success
2376 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2377 is too large (CVE-2015-8778)
2378 [18251] dynamic-link: SONAME missing when audit modules provides path
2379 [18265] libc: add attributes for wchar string and memory functions
2380 [18370] math: csqrt missing underflows
2381 [18421] libc: [hppa] read-only segment has dynamic relocations
2382 [18472] libc: Obsolete syscall wrappers should be compat symbols
2383 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2384 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2385 [18525] localedata: Remove locale timezone information
2386 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2387 [18568] localedata: Update locale data to Unicode 8.0
2388 [18589] locale: sort-test.sh fails at random
2389 [18595] math: ctan, ctanh missing underflows
2390 [18604] libc: assert macro-expands its argument
2391 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2392 contains a vector instruction exception.
2393 [18611] math: j1, jn missing errno setting on underflow
2394 [18618] localedata: sync Chechen locale definitions with other *_RU
2395 locales
2396 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2397 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2398 [18665] network: In send_dg, the recvfrom function is NOT always using the
2399 buffer size of a newly created buffer (CVE-2015-7547)
2400 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2401 missing break ?
2402 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2403 32bit processes
2404 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2405 [18699] math: tilegx cproj() for various complex infinities does not yield
2406 infinity
2407 [18724] libc: Harden put*ent functions against data injection
2408 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2409 [18755] build: build errors with -DNDEBUG
2410 [18757] stdio: fmemopen fails to set errno on failure
2411 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2412 be forced unloaded
2413 [18781] libc: openat64 lacks O_LARGEFILE
2414 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2415 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2416 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2417 [18790] math: [ldbl-128ibm] tanhl inaccurate
2418 [18795] libc: stpncpy fortification misses buffer lengths that are
2419 statically too large
2420 [18796] build: build fails for --disable-mathvec
2421 [18803] math: hypot missing underflows
2422 [18820] stdio: fmemopen may leak memory on failure
2423 [18823] math: csqrt spurious underflows
2424 [18824] math: fma spurious underflows
2425 [18825] math: pow missing underflows
2426 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2427 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2428 [18870] build: sem_open.c fails to compile with missing symbol
2429 FUTEX_SHARED
2430 [18872] stdio: Fix memory leak in printf_positional
2431 [18873] libc: posix_fallocate overflow check ineffective
2432 [18875] math: Excess precision leads incorrect libm
2433 [18877] libc: arm: mmap offset regression
2434 [18887] libc: memory corruption when using getmntent on blank lines
2435 [18918] localedata: hu_HU: change time to HH:MM:SS format
2436 [18921] libc: Regression: extraneous stat() and fstat() performed by
2437 opendir()
2438 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2439 binaries (CVE-2015-8777)
2440 [18951] math: tgamma missing underflows
2441 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2442 signgam
2443 [18953] localedata: lt_LT: change currency symbol to the euro
2444 [18956] math: powf inaccuracy
2445 [18961] math: [i386] exp missing underflows
2446 [18966] math: [i386] exp10 missing underflows
2447 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2448 [18969] build: multiple string test failures due to missing locale
2449 dependencies
2450 [18970] libc: Reference of pthread_setcancelstate in libc.a
2451 [18977] math: float / long double Bessel functions not in XSI POSIX
2452 [18980] math: i386 libm functions return with excess range and precision
2453 [18981] math: i386 scalb*, ldexp return with excess range and precision
2454 [18982] stdio: va_list and vprintf
2455 [18985] time: Passing out of range data to strftime() causes a segfault
2456 (CVE-2015-8776)
2457 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2458 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2459 [19012] locale: iconv_open leaks memory on error path
2460 [19016] math: clog, clog10 inaccuracy
2461 [19018] nptl: Mangle function pointers in tls_dtor_list
2462 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2463 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2464 [19048] malloc: malloc: arena free list can become cyclic, increasing
2465 contention
2466 [19049] math: [powerpc] erfc incorrect zero sign
2467 [19050] math: [powerpc] log* incorrect zero sign
2468 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2469 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2470 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2471 [19074] network: Data race in _res_hconf_reorder_addrs
2472 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2473 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2474 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2475 rounding modes
2476 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2477 ILP32
2478 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2479 [19086] manual: posix_fallocate64 documented argument order is wrong.
2480 [19088] math: lround, llround missing exceptions close to overflow
2481 threshold
2482 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2483 [19095] math: dbl-64 lrint incorrect for 64-bit long
2484 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2485 [19124] dynamic-link: ld.so failed to build with older assmebler
2486 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2487 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2488 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2489 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2490 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2491 pthread_setaffinity_np
2492 [19156] math: [ldbl-128] j0l spurious underflows
2493 [19164] nptl: tst-getcpu fails with many possible CPUs
2494 [19168] math: math/test-ildoubl and math/test-ldouble failure
2495 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2496 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2497 prelink
2498 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2499 (FE_NOMASK_ENV) do not clear SSE exceptions
2500 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2501 _int_new_arena/reused_arena
2502 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2503 [19201] math: dbl-64 remainder incorrect sign of zero result
2504 [19205] math: bits/math-finite.h conditions do not match math.h and
2505 bits/mathcalls.h
2506 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2507 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2508 for C99-based standards
2509 [19212] libc: features.h not -Wundef clean
2510 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2511 math-only
2512 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2513 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2514 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2515 disabled
2516 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2517 "inexact" exceptions
2518 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2519 arguments
2520 [19242] libc: strtol incorrect in Turkish locales
2521 [19243] malloc: reused_arena can pick an arena on the free list, leading
2522 to an assertion failure and reference count corruption
2523 [19253] time: tzset() ineffective when temporary TZ did not include DST
2524 rules
2525 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2526 [19270] math: [hppa] Shared libm missing __isnanl
2527 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2528 MAP_HUGETLB and MAP_STACK defines
2529 [19313] nptl: Wrong __cpu_mask for x32
2530 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2531 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2532 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2533 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2534 [19363] time: x32: times() return value wrongly truncates/sign extends
2535 from 32bit
2536 [19367] dynamic-link: Improve branch prediction on Silvermont
2537 [19369] network: Default domain name not reset by res_ninit when "search"
2538 / "domain" entry is removed from resolv.conf
2539 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2540 subnormals
2541 [19385] localedata: bg_BG: time separator should be colon, not comma
2542 [19408] libc: linux personality syscall wrapper may erroneously return an
2543 error on 32-bit architectures
2544 [19415] libc: dladdr returns wrong names on hppa
2545 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2546 IBM905, IBM907, and IBM909
2547 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2548 [19443] build: build failures with -DDEBUG
2549 [19451] build: Make check fails on test-double-vlen2
2550 [19462] libc: Glibc failed to build with -Os
2551 [19465] math: Wrong code with -Os
2552 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2553 -Os
2554 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2555 CPU's.
2556 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2557 [19486] math: S390: Math tests fail with "Exception Inexact set".
2558 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2559 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2560 architectures
2561 [19590] math: Fail to build shared objects that use libmvec.so functions.
2562 \f
2563 Version 2.22
2564
2565 * The following bugs are resolved with this release:
2566
2567 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2568 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2569 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2570 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2571 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2572 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2573 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2574 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2575 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2576 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2577 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2578 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2579 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2580 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2581 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2582 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2583 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2584 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2585 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2586 18657, 18676, 18694, 18696, 18887.
2587
2588 * Cache information can be queried via sysconf() function on s390 e.g. with
2589 _SC_LEVEL1_ICACHE_SIZE as argument.
2590
2591 * A buffer overflow in gethostbyname_r and related functions performing DNS
2592 requests has been fixed. If the NSS functions were called with a
2593 misaligned buffer, the buffer length change due to pointer alignment was
2594 not taken into account. This could result in application crashes or,
2595 potentially arbitrary code execution, using crafted, but syntactically
2596 valid DNS responses. (CVE-2015-1781)
2597
2598 * The time zone file parser has been made more robust against crafted time
2599 zone files, avoiding heap buffer overflows related to the processing of
2600 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2601 large time zone data files. Overly long time zone specifiers in the TZ
2602 variable no longer result in stack overflows and crashes.
2603
2604 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2605 for LD and GD on x86 and x86-64, has been implemented. You will need
2606 binutils-2.24 or later to enable this optimization.
2607
2608 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2609 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2610 Hat). These updates cause user visible changes, such as the fix for bug
2611 17998.
2612
2613 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2614 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2615 condition in some applications.
2616
2617 * Added vector math library named libmvec with the following vectorized x86_64
2618 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2619 pow, powf.
2620 The library can be disabled with --disable-mathvec. Use of the functions is
2621 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2622 Shared library libmvec.so is linked in as needed when using -lm (no need to
2623 specify -lmvec explicitly for not static builds).
2624 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2625
2626 * A new fmemopen implementation has been added with the goal of POSIX
2627 compliance. The new implementation fixes the following long-standing
2628 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2629 old implementation is still present for use be by existing binaries.
2630
2631 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2632 releases. It has been fixed to match 2.19 and older, but binaries built
2633 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2634
2635 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2636 Contributed by Roland McGrath (Google).
2637
2638 * The header <regexp.h> is deprecated, and will be removed in a future
2639 release. Use of this header will trigger a deprecation warning.
2640 Application developers should update their code to use <regex.h> instead.
2641
2642 This header was formerly part of SUS, but was deprecated in 1994 and
2643 removed from the standard in 2001. Also, the glibc implementation
2644 leaks memory. See BZ#18681 for more details.
2645 \f
2646 Version 2.21
2647
2648 * The following bugs are resolved with this release:
2649
2650 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2651 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2652 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2653 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2654 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2655 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2656 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2657 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2658 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2659 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2660 17892.
2661
2662 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2663 too little memory for the to-be-scanned arguments and overflow the
2664 allocated buffer. The implementation now correctly computes the required
2665 buffer size when using malloc, and switches to malloc from alloca as
2666 intended.
2667
2668 * A new semaphore algorithm has been implemented in generic C code for all
2669 machines. Previous custom assembly implementations of semaphore were
2670 difficult to reason about or ensure that they were safe. The new version
2671 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2672 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2673 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2674
2675 * Port to Altera Nios II has been contributed by Mentor Graphics.
2676
2677 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2678 implementations for powerpc64/powerpc64le.
2679 Implemented by Adhemerval Zanella (IBM).
2680
2681 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2682 and powerpc64le. This may improve lock scaling of existing programs on
2683 HTM capable systems. The lock elision code is only enabled with
2684 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2685 powerpc will issue a transaction abort on every syscall to avoid side
2686 effects being visible outside transactions.
2687
2688 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2689 AArch64. Contributed by ARM Ltd.
2690
2691 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2692
2693 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2694 under certain input conditions resulting in the execution of a shell for
2695 command substitution when the application did not request it. The
2696 implementation now checks WRDE_NOCMD immediately before executing the
2697 shell and returns the error WRDE_CMDSUB as expected.
2698
2699 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2700 processing format strings with a large number of format specifiers.
2701
2702 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2703 infinite loop if the DNS response contained a PTR record of an unexpected
2704 format.
2705
2706 * The minimum GCC version that can be used to build this version of the GNU
2707 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2708 still be used to compile programs using the GNU C Library.
2709
2710 * The GNU C Library is now built with -Werror by default. This can be
2711 disabled by configuring with --disable-werror.
2712
2713 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2714
2715 * The obsolete sigvec function has been removed. This was the original
2716 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2717 programs have been using instead for about 25 years. Of course, ABI
2718 compatibility for old binaries using sigvec remains intact.
2719
2720 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2721 with newer versions of bison.
2722
2723 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2724 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2725 registers overlay two consecutive single-precision registers. MIPS32R2
2726 introduced a new FPU mode (FR=1) where double-precision registers extend the
2727 corresponding single-precision registers which is incompatible with the
2728 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2729 require the use of FR=1 making a transition necessary. New o32 ABI
2730 extensions enable users to migrate over time from the original o32 ABI
2731 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2732 tracks the ABI of any loaded object and verifies that new objects are
2733 compatible. Mode transitions will also be requested as required and
2734 unsupportable objects will be rejected. The ABI checks include both soft and
2735 hard float ABIs for o32, n32 and n64.
2736
2737 GCC 5 with GNU binutils 2.25 onwards:
2738 It is strongly recommended that all o32 system libraries are built using the
2739 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2740 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2741 compiler using --with-fp-32=xx to set this by default.
2742 \f
2743 Version 2.20
2744
2745 * The following bugs are resolved with this release:
2746
2747 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2748 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2749 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2750 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2751 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2752 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2753 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2754 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2755 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2756 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2757 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2758 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2759 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2760 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2761 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2762
2763 * Reverted change of ABI data structures for s390 and s390x:
2764 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2765 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2766 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2767 are preserved pointing straight to the same implementation as the old ones.
2768 Given that, new callers will simply provide a too-big buffer to these
2769 functions. Any applications/libraries out there that embed jmp_buf or
2770 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2771 against 2.19 headers will have to rebuilt again. This is necessary in any
2772 case to revert the breakage in their ABI caused by the glibc change.
2773
2774 * Support for file description locks is added to systems running the
2775 Linux kernel. The standard file locking interfaces are extended to
2776 operate on file descriptions, not file descriptors, via the use of
2777 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2778 are associated with an open file instead of a process.
2779
2780 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2781
2782 * The minimum Linux kernel version that this version of the GNU C Library
2783 can be used with is 2.6.32.
2784
2785 * Running the testsuite no longer terminates as soon as a test fails.
2786 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2787 with PASS or FAIL lines for individual tests. A summary of the results is
2788 printed, including a list of failing lists, and "make check" exits with
2789 error status if there were any unexpected failures. "make check
2790 stop-on-test-failure=y" may be used to keep the old behavior.
2791
2792 * The am33 port, which had not worked for several years, has been removed
2793 from ports.
2794
2795 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2796 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2797 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2798 conflicted with POSIX (support for which was removed in 2.19), the
2799 interfaces those macros enabled remain available when compiling with
2800 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2801 test macros defined.
2802
2803 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2804
2805 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2806 This may improve lock scaling of existing programs on TX capable systems.
2807 The lock elision code is only built with --enable-lock-elision=yes and
2808 then requires a GCC version supporting the TX builtins. With lock elision
2809 default mutexes are elided via __builtin_tbegin, if the cpu supports
2810 transactions. By default lock elision is not enabled and the elision code
2811 is not built.
2812
2813 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2814 copy the path argument. This allowed programs to cause posix_spawn to
2815 deference a dangling pointer, or use an unexpected pathname argument if
2816 the string was modified after the posix_spawn_file_actions_addopen
2817 invocation.
2818
2819 * All supported architectures now use the main glibc sysdeps directory
2820 instead of some being in a separate "ports" directory (which was
2821 distributed separately before glibc 2.17).
2822
2823 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2824 On configurations that support it (all Linux configurations), it's now
2825 used regardless of the --enable-add-ons switch to configure. It is no
2826 longer possible to build such configurations without pthreads support.
2827
2828 * Locale names, including those obtained from environment variables (LANG
2829 and the LC_* variables), are more tightly checked for proper syntax.
2830 setlocale will now fail (with EINVAL) for locale names that are overly
2831 long, contain slashes without starting with a slash, or contain ".." path
2832 components. (CVE-2014-0475) Previously, some valid locale names were
2833 silently replaced with the "C" locale when running in AT_SECURE mode
2834 (e.g., in a SUID program). This is no longer necessary because of the
2835 additional checks.
2836
2837 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2838 with application code using Intel MPX instructions. (With all previous
2839 versions, the MPX register state could be clobbered when making calls
2840 into or out of a shared library.) Note that while the new dynamic
2841 linker is compatible with all known x86 hardware whether or not it
2842 supports Intel MPX, some x86 instruction-set emulators might fail to
2843 handle the new instruction encodings. This is known to affect Valgrind
2844 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2845 release), and might affect other tools that do instruction emulation.
2846
2847 * Support for loadable gconv transliteration modules has been removed.
2848 The support for transliteration modules has been non-functional for
2849 over a decade, and the removal is prompted by security defects. The
2850 normal gconv conversion modules are still supported. Transliteration
2851 with //TRANSLIT is still possible, and the //IGNORE specifier
2852 continues to be supported. (CVE-2014-5119)
2853
2854 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2855 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2856 resulting a denial-of-service security vulnerability in applications which
2857 use functions related to iconv. (CVE-2014-6040)
2858 \f
2859 Version 2.19
2860
2861 * The following bugs are resolved with this release:
2862
2863 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2864 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2865 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2866 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2867 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2868 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2869 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2870 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2871 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2872 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2873 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2874 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2875 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2876 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2877 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2878 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2879 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2880 16431, 16453, 16474, 16506, 16510, 16529
2881
2882 * Slovenian translations for glibc messages have been contributed by the
2883 Translation Project's Slovenian team of translators.
2884
2885 * The public headers no longer use __unused nor __block. This change is to
2886 support compiling programs that are derived from BSD sources and use
2887 __unused internally, and to support compiling with Clang's -fblock
2888 extension which uses __block.
2889
2890 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2891 large collation sequences to optimize multiple passes. This cache
2892 computation may overflow for large collation sequences and may cause a
2893 stack or buffer overflow. This is now fixed to use a slower algorithm
2894 which does not use a cache if there is an integer overflow.
2895
2896 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2897 rules for large collation sequences to optimize multiple passes and falls
2898 back to alloca if malloc fails, resulting in a possible stack overflow.
2899 The implementation now falls back to an uncached collation sequence lookup
2900 if malloc fails.
2901
2902 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2903 initialized for static applications resulting in the security feature
2904 being disabled. The pointer guard is now correctly initialized to a
2905 random value for static applications. Existing static applications need
2906 to be recompiled to take advantage of the fix (bug 15754).
2907
2908 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2909 to the d_name member of struct dirent, or omit the terminating NUL
2910 character. (Bugzilla #14699).
2911
2912 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2913 aligned_alloc functions could allocate too few bytes or corrupt the
2914 heap when passed very large allocation size values (Bugzilla #15855,
2915 #15856, #15857).
2916
2917 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2918 for AF_INET6 has been fixed (Bugzilla #16072).
2919
2920 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2921 pap_CW, quz_PE, the_NP.
2922
2923 * Substantially revised locales: gd_GB, ht_HT
2924
2925 * The LC_ADDRESS field was updated to support country_car for almost all
2926 supported locales.
2927
2928 * ISO 1427 definitions were updated.
2929
2930 * ISO 3166 definitions were updated.
2931
2932 * The localedef utility now supports --big-endian and --little-endian
2933 command-line options to generate locales for a different system from that
2934 for which the C library was built.
2935
2936 * Binary locale files now only depend on the endianness of the system for
2937 which they are generated and not on other properties of that system. As a
2938 consequence, binary files generated with new localedef may be incompatible
2939 with old versions of the GNU C Library, and binary files generated with
2940 old localedef may be incompatible with this version of the GNU C Library,
2941 in the following circumstances:
2942
2943 + Locale files may be incompatible on m68k systems.
2944
2945 + Locale archive files (but not separate files for individual locales) may
2946 be incompatible on systems where plain "char" is signed.
2947
2948 * The configure option --disable-versioning has been removed. Builds with
2949 --disable-versioning had not worked for several years.
2950
2951 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2952 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2953
2954 * SystemTap probes for malloc have been introduced.
2955
2956 * SystemTap probes for slow multiple precision fallback paths of
2957 transcendental functions have been introduced.
2958
2959 * Support for powerpc64le has been added.
2960
2961 * The soft-float powerpc port now supports e500 processors.
2962
2963 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2964
2965 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2966 set of header declarations that are enabled by default, even when other
2967 feature test macros or compiler options such as -std=c99 would otherwise
2968 disable some of those declarations.
2969
2970 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2971 conflict with POSIX. The libbsd-compat library (which was a dummy library
2972 that did nothing) has also been removed.
2973
2974 * Preliminary documentation about Multi-Thread, Async-Signal and
2975 Async-Cancel Safety has been added.
2976
2977 * Change of ABI data structures for s390 and s390x:
2978 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2979 allow for future hardware extensions. All library functions that accept or
2980 return these structures were versioned in order to provide backwards
2981 compatibility with existing code. However, not all cases can be handled
2982 with interface versioning. If an application embeds either structure into
2983 another structure and that structure is passed to another compilation unit
2984 which expects the newer larger-sized object then any access to the new
2985 fields will result in undefined behaviour. Similarly any access to fields
2986 that were moved by the enlarging of the structures will cause undefined
2987 behaviour. To fix the undefined behaviour all objects that are part of the
2988 final in-memory image must agree on the size of structures, and this may
2989 require recompilation.
2990 \f
2991 Version 2.18
2992
2993 * The following bugs are resolved with this release:
2994
2995 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2996 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2997 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2998 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2999 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3000 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3001 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3002 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3003 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3004 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3005 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3006 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3007 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3008 15755, 15759.
3009
3010 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3011 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3012 Distributions can re-enable building and using pt_chown via the new configure
3013 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3014 considerable security risks and should only be used if the distribution
3015 understands and accepts the risks.
3016
3017 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3018 #15078).
3019
3020 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3021 fixed (Bugzilla #15330).
3022
3023 * Add support for calling C++11 thread_local object destructors on thread
3024 and program exit. This needs compiler support for offloading C++11
3025 destructor calls to glibc.
3026
3027 * Improved worst case performance of libm functions with double inputs and
3028 output.
3029
3030 * Support for priority inherited mutexes in pthread condition variables on
3031 non-x86 architectures.
3032
3033 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3034
3035 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3036
3037 * Optimized string functions for ARM. Implemented by Will Newton and
3038 Richard Henderson.
3039
3040 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3041
3042 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3043 Richard Henderson.
3044
3045 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3046 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3047
3048 * Added a benchmark framework to track performance of functions in glibc.
3049
3050 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3051 It is based on draft TS 18661 and currently enabled as a GNU extension.
3052
3053 * On Linux, the clock function now uses the clock_gettime system call
3054 for improved precision, rather than old times system call.
3055
3056 * Added support for version-3 tz format files. This is needed when using
3057 the tz database release 2013e or later, and affects a few unusual cases --
3058 currently only TZ='America/Godthab' for time stamps after 2037.
3059
3060 * Added new API functions pthread_getattr_default_np and
3061 pthread_setattr_default_np to get and set the default pthread
3062 attributes of a process.
3063
3064 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3065 This may improve lock scaling of existing programs on TSX capable systems.
3066 When the --enable-lock-elision=yes parameter is specified at configure
3067 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3068 mutexes.
3069
3070 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3071 availability). Implemented by Ryan S. Arnold.
3072
3073 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3074 \f
3075 Version 2.17
3076
3077 * The following bugs are resolved with this release:
3078
3079 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3080 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3081 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3082 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3083 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3084 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3085 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3086 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3087 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3088 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3089 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3090 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3091 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3092 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3093 14879, 14889, 14893, 14898, 14914.
3094
3095 * Optimization of memcpy for MIPS.
3096
3097 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3098 EMFILE has been fixed (Bugzilla #14889).
3099
3100 * The add-on ports collection is now distributed in the "ports" subdirectory
3101 of the main GNU C Library distribution, rather than separately.
3102
3103 * Port to ARM AArch64 contributed by Linaro.
3104
3105 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3106 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3107 zEnterprise z196.
3108 Implemented by Andreas Krebbel.
3109
3110 * The new function secure_getenv allows secure access to the environment,
3111 returning NULL if running in a SUID/SGID process. This function replaces
3112 the internal function __secure_getenv.
3113
3114 * SystemTap static probes have been added into the dynamic linker.
3115 Implemented by Gary Benson.
3116
3117 * Optimizations of string functions strstr, strcasestr and memmem.
3118 Implemented by Maxim Kuvyrkov.
3119
3120 * The minimum Linux kernel version that this version of the GNU C Library
3121 can be used with is 2.6.16.
3122
3123 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3124 powerpc POWER7. Implemented by Will Schmidt.
3125
3126 * New configure option --disable-nscd builds the C library such that it
3127 never attempts to contact the Name Service Caching Daemon (nscd).
3128 New configure option --disable-build-nscd avoids building nscd itself;
3129 this is the default if --disable-nscd is used.
3130
3131 * Improved support for cross-compilation, including cross-testing and
3132 bootstrap builds without a previously built glibc.
3133
3134 * Several testsuite tests are now able to test multiple IFUNC variants of an
3135 interface, rather than just testing the one that would be chooen by
3136 default.
3137
3138 * New configure options --with-bugurl and --with-pkgversion, for
3139 distributors to use to embed their bug-reporting and package version
3140 information in --help and --version output.
3141
3142 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3143 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3144 allows creation of chroots without the procfs mounted on /proc.
3145
3146 * The `crypt' function now fails if passed salt bytes that violate the
3147 specification for those values. On Linux, the `crypt' function will
3148 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3149 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3150 when the mode is enabled.
3151
3152 * The `clock_*' suite of functions (declared in <time.h>) is now available
3153 directly in the main C library. Previously it was necessary to link with
3154 -lrt to use these functions. This change has the effect that a
3155 single-threaded program that uses a function such as `clock_gettime' (and
3156 is not linked with -lrt) will no longer implicitly load the pthreads
3157 library at runtime and so will not suffer the overheads associated with
3158 multi-thread support in other code such as the C++ runtime library.
3159
3160 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3161 sat_IN, and szl_PL.
3162 \f
3163 Version 2.16
3164
3165 * The following bugs are resolved with this release:
3166
3167 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3168 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3169 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3170 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3171 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3172 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3173 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3174 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3175 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3176 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3177 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3178 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3179 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3180 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3181 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3182 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3183 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3184 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3185 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3186 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3187 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3188 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3189 14277, 14278.
3190
3191 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3192 configuring glibc with:
3193 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3194 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3195 Implemented by H.J. Lu.
3196
3197 * ISO C11 support:
3198
3199 + define static_assert
3200
3201 + do not declare gets
3202
3203 + declare at_quick_exit and quick_exit also for ISO C11
3204
3205 + aligned_alloc. NB: The code is deliberately allows the size parameter
3206 to not be a multiple of the alignment. This is a moronic requirement
3207 in the standard but it is only a requirement on the caller, not the
3208 implementation.
3209
3210 + timespec_get added
3211
3212 + uchar.h support added
3213
3214 + CMPLX, CMPLXF, CMPLXL added
3215
3216 Implemented by Ulrich Drepper.
3217
3218 * Support for the IA-64 has been moved to ports.
3219
3220 * Remove support for anything but ELF binary format
3221
3222 * Checking versions of poll, ppoll added.
3223 Implemented by Ulrich Drepper.
3224
3225 * More generic and 64-bit performance optimizations to math functions.
3226 Implemented by Ulrich Drepper.
3227
3228 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3229 headers and functions available at compile time as they were before
3230 version 2.14. This option will be removed at some time in the future
3231 after the TI-RPC library becomes fully sufficient for the needs of
3232 existing applications.
3233
3234 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3235 Note that glibc is not expected to work with any Linux kernel version
3236 before 2.6.
3237
3238 * New header <sys/auxv.h> and function getauxval allowing easy access to
3239 the AT_* key-value pairs passed from the Linux kernel. The header also
3240 defines the HWCAP_* bits associated with the AT_HWCAP key.
3241
3242 * New locales: mag_IN
3243
3244 * New configure option --enable-systemtap builds SystemTap static probes
3245 into libc for setjmp and longjmp and into libpthread for various operations.
3246 So far the setjmp/longjmp probes and some of the libpthread probes are
3247 provided only for i*86 and x86_64.
3248 Implemented by Roland McGrath and Rayson Ho.
3249
3250 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3251
3252 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3253 and Will Schmidt.
3254
3255 * More optimized functions for SPARC. Implemented by David S. Miller.
3256
3257 * Improved support for cross-compilation, especially bootstrap builds
3258 without a previously built glibc.
3259
3260 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3261 Chris Metcalf from Tilera.
3262
3263 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3264 now supported for ARM processors.
3265
3266 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3267 as the name of the dynamic linker, to distinguish it from the
3268 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3269
3270 * Support for CRIS has been removed from ports.
3271
3272 * A new class of installed header has been documented for low-level
3273 platform-specific functionality. PowerPC added the first instance with a
3274 function to provide time base register access. Contributed by Tulio
3275 Magno Quites Machado Filho.
3276
3277 * ix86 configurations now install header files that are consistent with
3278 what x86-64 configurations install. These same header files can be used
3279 for -m32, -m64, or -mx32 builds.
3280 Contributed by H.J. Lu.
3281
3282 * Math library bug fixes. A thorough audit of all open math library bugs was
3283 conducted by Joseph Myers. Significant progress was made on many math
3284 library bugs resulting in more accurate exceptions and function results.
3285 Many thanks to all those that contributed including Andreas Jaeger for his
3286 patch review and work on the x87 trigonometric instruction issues.
3287
3288 * Timezone data is no longer installed. Timezone-related binaries and scripts
3289 will continue to be installed. Users should obtain their timezone data from
3290 their distribution provider or from the tzdata package at
3291 <ftp://munnari.oz.au/pub/>.
3292 \f
3293 Version 2.15
3294
3295 * The following bugs are resolved with this release:
3296
3297 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3298 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3299 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3300 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3301 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3302 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3303 13484, 13506, 13515, 13523, 13524, 13538, 13540
3304
3305 * New program pldd to list loaded object of a process
3306 Implemented by Ulrich Drepper.
3307
3308 * Add nss_db support back to glibc. No more dependency on Berkeley db
3309 and support for initgroups lookups.
3310 Implemented by Ulrich Drepper.
3311
3312 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3313 Contributed by HJ Lu.
3314
3315 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3316 Contributed by HJ Lu.
3317
3318 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3319 on x86-32 and x86-64.
3320 Contributed by Liubov Dmitrieva.
3321
3322 * Optimized strchr and strrchr for SSE on x86-32.
3323 Contributed by Liubov Dmitrieva.
3324
3325 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3326 for x86-64 and x86-32.
3327 Contributed by Liubov Dmitrieva.
3328
3329 * New interfaces: scandirat, scandirat64
3330 Implemented by Ulrich Drepper.
3331
3332 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3333 Implemented by Ulrich Drepper.
3334
3335 * nscd now also caches the netgroup database.
3336 Implemented by Ulrich Drepper.
3337
3338 * Integrate libm with gcc's -ffinite-math-only option.
3339 Implemented by Ulrich Drepper.
3340
3341 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3342 to math functions. Implemented by Ulrich Drepper.
3343
3344 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3345 Implemented by Ulrich Drepper.
3346
3347 * New Linux interfaces: process_vm_readv, process_vm_writev
3348
3349 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3350 Implemented by Ulrich Drepper.
3351
3352 * Optimized nearbyint and strcasecmp for PPC.
3353 Implemented by Adhemerval Zanella.
3354
3355 * New locales: bho_IN, unm_US, es_CU, ta_LK
3356 \f
3357 Version 2.14
3358
3359 * The following bugs are resolved with this release:
3360
3361 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3362 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3363 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3364 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3365 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3366 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3367 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3368 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3369 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3370 12795, 12811, 12813, 12814, 12841
3371
3372 * The RPC implementation in libc is obsoleted. Old programs keep working
3373 but new programs cannot be linked with the routines in libc anymore.
3374 Programs in need of RPC functionality must be linked against TI-RPC.
3375 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3376
3377 Visible changes of this change include (obviously) the inability to link
3378 programs using RPC functions without referencing the TI-RPC library and the
3379 removal of the RPC headers from the glibc headers.
3380 Implemented by Ulrich Drepper.
3381
3382 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3383 syncfs, setns, sendmmsg
3384
3385 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3386 yue_HK, lij_IT, mhr_RU
3387
3388 * New encodings: CP770, CP771, CP772, CP773, CP774
3389
3390 * New program sotruss to trace calls through PLTs
3391 Implemented by Ulrich Drepper.
3392
3393 * The malloc hook implementation is marked deprecated and will be removed
3394 from the default implementation in the next version. The design never
3395 worked ever since the introduction of threads. Even programs which do
3396 not create threads themselves can use multiple threads created internally.
3397 \f
3398 Version 2.13
3399
3400 * The following bugs are resolved with this release:
3401
3402 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3403 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3404 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3405 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3406 12378, 12394, 12397
3407
3408 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3409
3410 * POWER7 optimizations: memset, memcmp, strncmp
3411
3412 * New optimized string functions for x86-64: strnlen (SSE2),
3413 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3414 Implemented by Ulrich Drepper.
3415 \f
3416 Version 2.12
3417
3418 * The following bugs are resolved with this release:
3419
3420 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3421 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3422 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3423 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3424 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3425 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3426 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3427 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3428
3429 * New interfaces: pthread_getname_np, pthread_setname_np
3430
3431 * New Linux interface: recvmmsg
3432
3433 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3434
3435 * The dynamic linker now recognizes supported ABI versions from the
3436 EI_ABIVERSION field in the ELF header.
3437 Implemented by Ulrich Drepper.
3438
3439 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3440 will not be used to fill in password fields in the passwd.byname replies.
3441 Instead it is used to synthesize the shadow.byname table, should it be
3442 missing. This is a useful mode in some installations involving Solaris.
3443 Implemented by Ulrich Drepper.
3444
3445 * New locales: kok_IN, sq_MK, cv_RU
3446 \f
3447 Version 2.11
3448
3449 * The following bugs are resolved with this release:
3450
3451 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3452 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3453 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3454 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3455 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3456 10789, 10817, 10823, 10840, 10847
3457
3458 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3459 mkostemps64
3460 Implemented by Ulrich Drepper.
3461
3462 * Checking version of longjmp added that fails if an uninitialized stack
3463 frame would be created. Implemented by Ulrich Drepper.
3464
3465 * STT_GNU_IFUNC is now supported in static executables.
3466 Implemented by H.J. Lu.
3467
3468 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3469
3470 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3471 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3472 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3473 Contributed by H.J. Lu.
3474
3475 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3476 Implemented by Ulrich Drepper.
3477
3478 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3479 strstr, strcasestr.
3480 Contributed by H.J. Lu.
3481
3482 * Support for fma instruction in AVX on x86-64.
3483 Implemented by H.J. Lu and Ulrich Drepper.
3484
3485 * AVX support in x86-64 auditing support in ld.so.
3486 Implemented by H.J. Lu.
3487
3488 * STB_GNU_UNIQUE support added.
3489 Implemented by Ulrich Drepper.
3490
3491 * Implement second fallback mode for optimized DNS lookup for even more
3492 broken environments. If two requests from the same port are not
3493 handled correctly close the socket and open a new one before sending
3494 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3495 can be used to select this mode right away, instead of rediscovering the
3496 necessity is every process again.
3497 Implemented by Ulrich Drepper.
3498
3499 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3500 Implemented by Adam Tkac.
3501
3502 * Optimized iconv conversions for S390x.
3503 Implemented by Andreas Krebbel.
3504
3505 * Using condvars with PI mutexes is now more efficient due to kernel
3506 support for requeueing to PI futexes. NPTL support added for x86-64.
3507 Implemented by Ulrich Drepper.
3508
3509 * New locales: ps_AF, my_MM
3510 \f
3511 Version 2.10
3512
3513 * The following bugs are resolved with this release:
3514
3515 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3516 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3517 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3518 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3519 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3520
3521 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3522 Implemented by Ulrich Drepper.
3523
3524 * New Linux interfaces: accept4, fallocate, fallocate64.
3525 Implemented by Ulrich Drepper.
3526
3527 * Correct declarations of string function when used in C++ code. This
3528 could lead to compile errors for invalid C++ code.
3529
3530 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3531 now in POSIX.
3532
3533 * New POSIX 2008 interface: psiginfo
3534 Implemented by Ulrich Drepper.
3535
3536 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3537 Implemented by Ulrich Drepper.
3538
3539 * Support for selecting between multiple function definitions at runtime
3540 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3541
3542 * The libcrypt library can now use the hash function implementations in
3543 NSS. Implemented by Ulrich Drepper.
3544
3545 * The malloc implementation can be compiled to be less memory efficient
3546 but higher performing in multi-threaded programs.
3547 Implemented by Ulrich Drepper.
3548
3549 * New locales: nan_TW@latin, ks_IN
3550
3551 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3552 Implemented by Ulrich Drepper.
3553
3554 * Extended printf hook support. It is possible to use user-defined types
3555 and extend existing format specifiers.
3556 Implemented by Ulrich Drepper.
3557
3558 * Handling for group shadow files has been added.
3559 Implemented by Ulrich Drepper.
3560
3561 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3562 servers (the case, e.g., for some people using the built-in DNS
3563 server in ADSL modems/routers). There is a once-per-process timeout
3564 in case of a broken server. To avoid it, users can run nscd or put
3565 'options single-request' in /etc/resolv.conf.
3566 Implemented by Ulrich Drepper.
3567 \f
3568 Version 2.9
3569
3570 * The following bugs are resolved with this release:
3571
3572 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3573 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3574 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3575 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3576 6968, 6974, 6980, 6995, 7008, 7009, 7029
3577
3578 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3579 up at the same time. Implemented by Ulrich Drepper.
3580
3581 * TLS descriptors for LD and GD on x86 and x86-64.
3582 Implemented by Alexandre Oliva.
3583
3584 * getaddrinfo now handles DCCP and UDPlite.
3585 Implemented by Ulrich Drepper.
3586
3587 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3588 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3589 Implemented by Ulrich Drepper.
3590
3591 * New implementation of memmem, strstr, and strcasestr which is O(n).
3592 Implemented by Eric Blake.
3593
3594 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3595
3596 * Implement "e" option for popen to open file descriptor with the
3597 close-on-exec flag set. Implemented by Ulrich Drepper.
3598
3599 * Implement "b" mode for fmemopen. In this mode writes writes don't
3600 implicitly add a NUL byte and seeks from the end of the buffer really
3601 use the buffer end, not the string length as the basis.
3602 Implemented by Ulrich Drepper.
3603
3604 * Many functions, exported and internal, now atomically set the close-on-exec
3605 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3606
3607 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3608 Sinhala)
3609 Implemented by Pravin Satpute.
3610
3611 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3612 \f
3613 Version 2.8
3614
3615 * The following bugs are resolved with this release:
3616
3617 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3618 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3619 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3620 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3621 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3622 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3623 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3624
3625 * New locales: bo_CN, bo_IN, shs_CA.
3626
3627 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3628
3629 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3630 Implemented by Pravin Satpute.
3631
3632 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3633
3634 * nscd caches now all timeouts for DNS entries
3635 Implemented by Ulrich Drepper.
3636
3637 * nscd is more efficient and wakes up less often.
3638 Implemented by Ulrich Drepper.
3639
3640 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3641 vdprintf, and obstack_vprintf.
3642 Implemented by Jakub Jelinek.
3643
3644 * Faster memset for x86-64.
3645 Implemented by Harsha Jagasia and H.J. Lu.
3646
3647 * Faster memcpy on x86.
3648 Implemented by Ulrich Drepper.
3649
3650 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3651 Implemented by Ulrich Drepper.
3652
3653 * Faster sqrt and sqrtf implemention for some PPC variants.
3654 Implemented by Steven Munroe.
3655 \f
3656 Version 2.7
3657
3658 * The following bugs are resolved with this release:
3659
3660 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3661 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3662 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3663 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3664 5071, 5103, 5104, 5112, 5113, 5184, 5186
3665
3666 * More checking functions: fread, fread_unlocked, open*, mq_open.
3667 Implemented by Jakub Jelinek and Ulrich Drepper.
3668
3669 * Extend fortification to C++. Implemented by Jakub Jelinek.
3670
3671 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3672 by not recognizing 'a' as a modifier when those specs are requested.
3673 Implemented by Jakub Jelinek.
3674
3675 * PPC optimizations to math and string functions.
3676 Implemented by Steven Munroe.
3677
3678 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3679 options to be passed. Implemented by Ulrich Drepper.
3680
3681 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3682
3683 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3684 Implemented by Ulrich Drepper.
3685
3686 * Handle private futexes in the NPTL implementation.
3687 Implemented by Jakub Jelinek and Ulrich Drepper.
3688
3689 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3690 Implemented by Roland McGrath and Ulrich Drepper.
3691
3692 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3693
3694 * SHA-256 and SHA-512 based password encryption.
3695 Implemented by Ulrich Drepper.
3696
3697 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3698 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3699 yo_NG.
3700
3701 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3702 Implemented by Ulrich Drepper.
3703 \f
3704 Version 2.6
3705
3706 * The following bugs are resolved with this release:
3707
3708 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3709 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3710 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3711 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3712 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3713 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3714 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3715 4702, 4858
3716
3717 * New Linux interfaces: epoll_pwait, sched_getcpu.
3718
3719 * New generic interfaces: strerror_l.
3720
3721 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3722 \f
3723 Version 2.5
3724
3725 * The following bugs are resolved with this release:
3726
3727 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3728 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3729 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3730 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3731 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3732 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3733 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3734 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3735 3225, 3251, 3252, 3253, 3273
3736
3737 * For Linux, the sorting of addresses returned by getaddrinfo now also
3738 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3739 Implemented by Ulrich Drepper.
3740
3741 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3742 Implemented by Ulrich Drepper.
3743
3744 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3745
3746 * New iconv module for MIK. Contributed by Alexander Shopov.
3747
3748 * For sites with broken group and/or passwd database, the auto-propagate
3749 option of nscd can prevent creating ID lookup entries from the results
3750 of a name lookup and vice versa. This usually is no problem but some
3751 site might have problems with the default behavior.
3752 Implemented by Ulrich Drepper.
3753
3754 * Iterating over entire database in NIS can be slow. With the
3755 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3756 to trade time for memory. The entire database will be read at once.
3757 Implemented by Ulrich Drepper.
3758
3759 * The interfaces introduced in RFC 3542 have been implemented by
3760 Ulrich Drepper.
3761
3762 * Support for the new ELF hash table format was added by Ulrich Drepper.
3763
3764 * Support for priority inheritance mutexes added by Jakub Jelinek and
3765 Ulrich Drepper.
3766
3767 * Support for priority protected mutexes added by Jakub Jelinek.
3768
3769 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3770 or_IN, csb_PL, fy_NL, sr_ME.
3771 \f
3772 Version 2.4
3773
3774 * More overflow detection functions.
3775
3776 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3777 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3778
3779 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3780 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3781 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3782 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3783 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3784 by Masahide Washizawa.
3785
3786 * It is now possible to install audit libraries for the dynamic linker using
3787 LD_AUDIT. Implemented by Ulrich Drepper.
3788
3789 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3790 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3791 For a libc and libpthread that works well on Linux 2.4 kernels, we
3792 recommend using the stable 2.3 branch.
3793
3794 * The new function open_wmemstream acts like open_memstream,
3795 but uses a wchar_t wide character buffer.
3796
3797 * The new function ppoll is an improved version of poll, similar to pselect.
3798
3799 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3800 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3801 renameat, unlinkat, linkat, symlinkat, readlinkat.
3802
3803 * New Linux kernel system calls: unshare,
3804 inotify_init, inotify_add_watch, inotify_rm_watch.
3805
3806 * The euidaccess function is now also known by the alias eaccess,
3807 for compatibility with some other systems.
3808
3809 * Timezone data updated to 2006b version.
3810 \f
3811 Version 2.3.6
3812
3813 * The following bugs are resolved with this release:
3814
3815 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3816 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3817 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3818 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3819 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3820 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3821
3822 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3823
3824 * As of this release, GCC 4 can be used to compile the C Library.
3825
3826 * Timezone data updated to 2005m version.
3827 \f
3828 Version 2.3.5
3829
3830 * The following bugs are resolved with this release:
3831
3832 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3833 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3834 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3835 777, 787, 821, 822, 823, 825
3836
3837 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3838 \f
3839 Version 2.3.4
3840
3841 * Support for RFC 3678. Real implementations exist only for Linux so far.
3842 Implemented by Ulrich Drepper.
3843
3844 * nscd can now cache entries persistently. Expiring entries are reloaded.
3845 For speedups the cache can be shared in memory with client processes.
3846 Implemented by Ulrich Drepper.
3847
3848 * nscd can now perform SELinux checks.
3849 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3850
3851 * getaddrinfo queries are now cached. Canonical name lookup is performed
3852 efficiently.
3853 Implemented by Ulrich Drepper.
3854
3855 * The nothrow function attribute is used when headers are used by gcc when
3856 compiling C code. This can avoid significant amounts of exception
3857 handling data.
3858
3859 * The malloc functions perform more error checking and are stricter when
3860 it comes to reacting on errors. The default action is to terminate
3861 the process after showing an error message. Implemented by Ulrich Drepper.
3862
3863 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3864 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3865
3866 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3867 namespaces using the new function dlmopen(). This feature is of course,
3868 like most other dynamic loading functionality, not available in statically
3869 linked applications. Implemented by Ulrich Drepper.
3870
3871 * Low-overhead boundary checking variants of string and some stdio functions
3872 were added. These are to be used in conjunction with a gcc patch by
3873 Jakub Jelinek which adds calls to these functions if possible.
3874 Implemented by Jakub Jelinek and Ulrich Drepper.
3875
3876 * Old code for several operating systems and machine architectures that
3877 have not been in working condition in a long time have been removed from
3878 the main source tree maintained by the GNU C Library's maintainers.
3879 These files are now reside in the separate `ports' source module
3880 that is usable as an add-on when building the library.
3881 \f
3882 Version 2.3.3
3883
3884 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3885 interrogate the dynamic linker, compatible with the Solaris interface.
3886
3887 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3888 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3889
3890 * getifaddrs now uses the netlink interface on Linux to get its information.
3891 Implemented by Thorsten Kukuk.
3892
3893 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3894 Implemented by Ulrich Drepper.
3895
3896 * support for non-executable stacks on x86 has been added. Changes mostly
3897 by Roland McGrath.
3898
3899 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3900 and Ulrich Drepper.
3901
3902 * getaddrinfo now performs destination address selection according to
3903 RFC 3484.
3904 \f
3905 Version 2.3.2
3906
3907 * Thread-safe interfaces for many functions that access locale data
3908 were added in version 2.3, but these features were omitted from NEWS.
3909 Many functions have variants with an `_l' suffix that take a `locale_t'
3910 object as a parameter rather than consulting the current locale.
3911 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3912 create and maintain `locale_t' objects. Additionally, the new function
3913 `uselocale' sets "the current locale" (as used by functions not so
3914 parameterized) set for an individual thread. These features were added
3915 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3916
3917 * The functions getresuid, getresgid, setresuid, and setresgid, which
3918 have long been available on Linux, are now declared in <unistd.h>
3919 and are now also available on the Hurd.
3920
3921 * ELF thread-local storage support (TLS) now works on x86-64.
3922
3923 * The new dynamic string token $LIB is expanded in shared library names.
3924 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3925
3926 * Aldy Hernandez contributed complete software floating point support for
3927 PowerPC machines with no FPU.
3928
3929 * fexecve is implemented on Linux.
3930
3931 * The `btowc' function should work at least twice as fast due to
3932 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3933
3934 * With approriate thread add-ons cancelable functions are now implemented
3935 in libc.so as well. No need to call the function in libpthread. This
3936 change allowed to finally disable the incorrect and expensive handling
3937 of weak definition in ld.so.
3938
3939 * Yet more PLT entries in libc.so have been removed. We finally arrived
3940 at the bare minimum. Startup times improved appropriately.
3941
3942 * Support for the new Linux/x86 system call interface was added. The
3943 AT_SYSINFO auxiliary vector entry is recognized and handled.
3944 \f
3945 Version 2.3
3946
3947 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3948 charsets.
3949
3950 * iconv (the program and the interface) now accepts empty names (excluding
3951 options like //TRANSLIT) to mean "use charset of current locale".
3952
3953 * localedef can now transliterate characters in strings which are not in
3954 the provided charmap. The information from the input locale is used.
3955
3956 * Prelinking support was added for ELF targets. This requires additional
3957 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3958
3959 * Read-only stdio streams now use mmap to speed up operation by eliminating
3960 copying and buffer underflows. To use add 'm' to the mode string of
3961 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3962
3963 * The malloc functions were completely rewritten by Wolfram Gloger based
3964 on Doug Lea's malloc-2.7.0.c.
3965
3966 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3967 implementation of regex.
3968
3969 * Bruno Haible upgraded the iconv and locale implementation to support
3970 Unicode 3.2.
3971
3972 * Contents of the LC_* and LANG environment variables in the CEN style are
3973 not recognized anymore. It never was used. Change by Ulrich Drepper.
3974
3975 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3976 thread-local storage (TLS) ABI on some platforms.
3977 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3978
3979 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3980 EUC-JISX0213, and TSCII.
3981
3982 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3983 BSD-compatible interface for getting all network interface addresses.
3984 Implementation for IPv4 by Roland McGrath.
3985
3986 * Loading of locale data is faster due to the introduction of a locale
3987 archive. Implemented by Roland McGrath and Ulrich Drepper.
3988
3989 * Startup times are significantly reduced by not using exported functions
3990 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3991 and Ulrich Drepper.
3992
3993 * Steven Munroe contributed a port to PowerPC64/Linux.
3994 \f
3995 Version 2.2.6
3996
3997 * The Hurd now uses the GNU libio implementation of stdio rather than the
3998 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3999
4000 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4001 with the same behavior as the Linux system call of the same name.
4002 \f
4003 Version 2.2.5
4004
4005 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4006 128-bit long double format.
4007
4008 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4009 IBM1161, and IBM1162 charsets.
4010
4011 * Andreas Jaeger contributed a port to x86-64/Linux.
4012
4013 * Peter Bruin contributed a port to PowerPC/Hurd.
4014
4015 * libc functions using I/O streams now can handle wide-oriented streams
4016 as well.
4017
4018 * optimizations in the dynamic linker. Binaries created by recent binutils
4019 versions start up quicker due to reduced time spent on relocations.
4020
4021 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4022 \f
4023 Version 2.2.4
4024
4025 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4026 asinh, atanh, j0 for the 128-bit long double format.
4027
4028 * Bruno Haible updated all the code handling Unicode in some form to
4029 support Unicode 3.1.
4030
4031 * Speed of regex for single-byte locales is back to previous levels.
4032 Patch by Isamu Hasegawa.
4033
4034 * Alpha, SPARC, and IA-64 now also using floating stacks.
4035
4036 * Startup time of internationalized applications greatly improved through
4037 iconv cache. Use iconvconfig to generate the cache file.
4038 Contributed by Ulrich Drepper.
4039
4040 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4041 severe performance problems. Patch by David Mosberger.
4042
4043 * The Hurd port got a lot more functionality like AIO, various stdio
4044 extensions, etc. Mainly done by Roland McGrath.
4045
4046 * mtrace can now lookup symbols in shared libraries.
4047 \f
4048 Version 2.2.3
4049
4050 * Intel's IA-64 math library is largely integrated. It provides fast and
4051 accurate implementatations for most basic and standard math functions
4052 in float, double, and long double format.
4053
4054 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4055 and asin for the 96-bit long double format and asin, log, tan for the
4056 128-bit long double format.
4057
4058 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4059 The basic double functions exist today. Contributed by Abraham Ziv
4060 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4061 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4062
4063 * An asynchronous name lookup library was added. The interface is designed
4064 after POSIX AIO. The proposal was circulated beforehand to get comments.
4065 No negative ones came in. Implemented by Ulrich Drepper.
4066
4067 * Port to S390/64bit contributed by Martin Schwidefsky
4068 <schwidefsky@de.ibm.com>.
4069
4070 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4071 of functions for Linux/IA-64.
4072
4073 * The RPC code is now thread safe. Threads can now use the same service
4074 of different services at the same time. Patch by Eric Norum
4075 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4076
4077 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4078 family of functions for Linux/S390.
4079
4080 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4081 of functions for Linux/x86.
4082
4083 * Port to Linux/CRIS contributed by Axis Communications.
4084 \f
4085 Version 2.2.2
4086
4087 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4088 we can now check for namespace violations and missing declarations. The
4089 result is that almost all headers are now Unix-compliant (as defined in
4090 the upcoming XPG6). The negative side is that some programs might need
4091 corrections, too, if they depend on the incorrect form of the headers in
4092 previous versions which defined too many symbols and included too many
4093 other headers.
4094
4095 * regex now handles multibyte character sets correctly.
4096 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4097
4098 * iconv (the program) does now conform to the upcoming XPG6 and handles
4099 charmaps. Instead of the charset names the path of charmaps can be
4100 provided and the conversion happens based on this data.
4101 Contributed by Ulrich Drepper.
4102
4103 * The locale program now provides detailed information about the installed
4104 locales. While
4105
4106 locale -a
4107
4108 only lists the names of the supported locales
4109
4110 locale -a --verbose
4111
4112 provides details such as country, language, and codeset name.
4113 Contributed by Ulrich Drepper.
4114 \f
4115 Version 2.2.1
4116
4117 * The gencat program now parses the input file according to the charset
4118 selected by the LC_CTYPE category. This is important for stateful
4119 character sets. To make generating catalogs easier there is a way
4120 to overwrite the charset selected by the locale: before the first
4121 message or $ quote line the catalog can contain a line like
4122
4123 $ codeset=ISO-8859-2
4124
4125 to select the charset (ISO-8859-2 in this case).
4126
4127 Implemented by Shinya Hanataka and Ulrich Drepper.
4128
4129 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4130 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4131 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4132
4133 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4134 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4135
4136 * The LANGUAGE environment variable is now ignored unless the locale is
4137 changed from the default "C" locale.
4138
4139 * The usual bug fixes.
4140 \f
4141 Version 2.2
4142
4143 * Greg McGary added runtime support for bounds checking using gcc's
4144 new -fbounded-pointers option. ix86 target is complete. PowerPC
4145 is in progress.
4146
4147 * Thorsten Kukuk added secure mode support to nscd.
4148
4149 * The Berkeley db libraries have been removed.
4150
4151 Related, the nss_db module is now in a separate package since it
4152 obviously requires a database library being available.
4153
4154 * Wide character I/O streams implemented by Ulrich Drepper.
4155
4156 * Functions from the extended socket API added by Ulrich Drepper.
4157
4158 * Functions feenableexcept and fedisableexcept to control the
4159 behaviour of individual exceptions have been added by Andreas Jaeger.
4160
4161 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4162
4163 * The resolver code has been updated from bind 8.2.3-T5B which supports
4164 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4165 and Mark Kettenis.
4166
4167 This change could in some situations effect backward compatibility. Since
4168 now `_res' is a thread-local instead of a global variable, modifying it
4169 in one thread does not have any effect in other threads.
4170
4171 The resolver library was also extended to allow IPv6 as the transport
4172 protocol for the requests. This work was done by Stig Venaas.
4173
4174 * Compatibility code for K&R C compilers has been removed from the
4175 header files. A ISO C compiler is needed to use the library
4176 (conforming to either C89 or C99 standard).
4177
4178 * Complete rewrite of the localedef program to support multibyte character
4179 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4180 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4181 Implemented by Ulrich Drepper.
4182
4183 Bruno Haible significantly improved the generation and use of the data
4184 structures for the wide character tables.
4185
4186 * Plural handling in gettext implemented by Ulrich Drepper.
4187
4188 * The utmp daemon has been removed.
4189
4190 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4191
4192 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4193 and Yutaka Niibe.
4194
4195 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4196
4197 * POSIX barriers implemented by Kaz Kylheku.
4198
4199 * POSIX spawn function family implemented by Ulrich Drepper.
4200
4201 * POSIX spinlocks are now available.
4202
4203 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4204
4205 * the configure option --enable-kernel=X.Y.Z allows to strip out
4206 compatibility for kernel versions before X.Y.Z. This is currently only
4207 implemented for Linux.
4208
4209 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4210 field sin6_scope_id. This means that all programs using IPv6 should be
4211 recompiled. Don't expect binary compatibility with previous glibc
4212 versions.
4213
4214 * various conversion modules for IBM character sets contributed by
4215 Masahide Washizawa.
4216
4217 * IA-64 port by Jes Sorensen and HJ Lu.
4218 \f
4219 Version 2.1.3
4220
4221 * bug fixes
4222
4223 \f
4224 Version 2.1.2
4225
4226 * bug fixes
4227
4228 \f
4229 Version 2.1.1
4230
4231 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4232
4233 * New xdr functions are added; some rpc functions are now 64bit clean.
4234
4235 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4236
4237 * Fixed known incompatibilities with glibc 2.0.
4238
4239 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4240
4241 * Optimized a number of functions (especially the ELF dynamic loader).
4242
4243 * Update timezone data files.
4244
4245 * lots of charmaps corrections
4246
4247 * some new locale definitions and charmaps
4248
4249 \f
4250 Version 2.1
4251
4252 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4253 conform to POSIX member type requirements. Symbol versions have been
4254 adjusted as needed within the library, and for direct use by applications,
4255 but there is potential for problems if third-party libraries use
4256 struct timeval as part of their interface. This does not present
4257 a problem for X and other "essential" system libraries.
4258
4259 * An additional locale model to support C++ Standard Library locale
4260 model and probably more was implemented by Ulrich Drepper.
4261
4262 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4263 symbol level.
4264
4265 * Miles Bader provided the `argp' function family to support hierachical
4266 command line argument parsing, layered on top of getopt.
4267
4268 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4269
4270 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4271 numbers.
4272
4273 * scanf recognizes the %a and %A format for scanning floating point numbers.
4274
4275 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4276 information and interfaces for the available integer types.
4277
4278 * about 130 new math functions were added to implement the ISO C9x math
4279 library.
4280
4281 * the new header <complex.h> contains definitions of the complex math
4282 functions from ISO C 9X.
4283
4284 * the new header <tgmath.h> defines generic macros to use complex or
4285 real valued functions.
4286
4287 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4288
4289 * Andreas Jaeger provided a test suite for the math library.
4290
4291 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4292
4293 * Ulrich Drepper added character set conversion functions (iconv).
4294
4295 * Optimized string functions have been added.
4296
4297 * The localedata addon is now part of glibc.
4298
4299 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4300
4301 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4302 daemon for NSS (nscd).
4303
4304 Missing a better place here are some numbers on improvements. Under
4305 Linux 2.1.125 un-tar-ing the kernel sources takes
4306
4307 user system wall
4308
4309 using local files 12.19s 6.88s 22.91s
4310
4311 using NIS 13.92s 8.91s 26.34s
4312
4313 using NIS & nscd 10.37s 7.34s 25.30s
4314
4315 using NIS+ 27.57s 30.37s 640.46s
4316
4317 using NIS+ & nscd 10.25s 7.83s 26.51s
4318
4319 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4320
4321 Keep in mind that non-namelookup related operations dominate above times.
4322 It was just a common complain that using NIS+ unpacking the kernel is
4323 horribly slow.
4324
4325 [1] The old nscd implementation is not available anymore since it was
4326 distributed with glibc up to version 2.0.98 and thus is now replaced.
4327
4328 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4329
4330 * Mark Kettenis provided a Hesiod NSS module.
4331
4332 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4333 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4334
4335 * rcmd can now handle netgroups (Dick Streefland).
4336
4337 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4338 Bambrough.
4339
4340 * Support for the IPv6 protocol has been added to the socket API, as per the
4341 latest draft standards.
4342
4343 * Support for Linux 2.2 has been added.
4344
4345 * Interface changes relative to the latest 2.0.x release:
4346 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4347 addseverity NEW: Unix98
4348 alphasort64 NEW: LFS
4349 argp_err_exit_status NEW: argp, GNU ext
4350 argp_error NEW: argp, GNU ext
4351 argp_failure NEW: argp, GNU ext
4352 argp_help NEW: argp, GNU ext
4353 argp_parse NEW: argp, GNU ext
4354 argp_program_bug_address NEW: argp, GNU ext
4355 argp_program_version NEW: argp, GNU ext
4356 argp_program_version_hook NEW: argp, GNU ext
4357 argp_state_help NEW: argp, GNU ext
4358 argp_usage NEW: argp, GNU ext
4359 authdes_create NEW: Secure RPC
4360 authdes_getucred NEW: Secure RPC
4361 authdes_pk_create NEW: Secure RPC
4362 backtrace NEW: GNU ext.
4363 backtrace_symbols NEW: GNU ext.
4364 backtrace_symbols_fd NEW: GNU ext.
4365 cacos NEW: ISO C 9x
4366 cacosf NEW: ISO C 9x
4367 cacosh NEW: ISO C 9x
4368 cacoshf NEW: ISO C 9x
4369 cacoshl NEW: ISO C 9x
4370 cacosl NEW: ISO C 9x
4371 capget NEW: kernel
4372 capset NEW: kernel
4373 carg NEW: ISO C 9x
4374 cargf NEW: ISO C 9x
4375 cargl NEW: ISO C 9x
4376 casin NEW: ISO C 9x
4377 casinf NEW: ISO C 9x
4378 casinh NEW: ISO C 9x
4379 casinhf NEW: ISO C 9x
4380 casinhl NEW: ISO C 9x
4381 casinl NEW: ISO C 9x
4382 catan NEW: ISO C 9x
4383 catanf NEW: ISO C 9x
4384 catanh NEW: ISO C 9x
4385 catanhf NEW: ISO C 9x
4386 catanhl NEW: ISO C 9x
4387 catanl NEW: ISO C 9x
4388 cbc_crypt NEW: Secure RPC
4389 ccos NEW: ISO C 9x
4390 ccosf NEW: ISO C 9x
4391 ccosh NEW: ISO C 9x
4392 ccoshf NEW: ISO C 9x
4393 ccoshl NEW: ISO C 9x
4394 ccosl NEW: ISO C 9x
4395 cexp NEW: ISO C 9x
4396 cexpf NEW: ISO C 9x
4397 cexpl NEW: ISO C 9x
4398 cimag NEW: ISO C 9x
4399 cimagf NEW: ISO C 9x
4400 cimagl NEW: ISO C 9x
4401 clearerr_locked REMOVED
4402 clntunix_create NEW: sunrpc ext
4403 clog NEW: ISO C 9x
4404 clog10 NEW: ISO C 9x
4405 clog10f NEW: ISO C 9x
4406 clog10l NEW: ISO C 9x
4407 clogf NEW: ISO C 9x
4408 clogl NEW: ISO C 9x
4409 conj NEW: ISO C 9x
4410 conjf NEW: ISO C 9x
4411 conjl NEW: ISO C 9x
4412 cpow NEW: ISO C 9x
4413 cpowf NEW: ISO C 9x
4414 cpowl NEW: ISO C 9x
4415 cproj NEW: ISO C 9x
4416 cprojf NEW: ISO C 9x
4417 cprojl NEW: ISO C 9x
4418 creal NEW: ISO C 9x
4419 crealf NEW: ISO C 9x
4420 creall NEW: ISO C 9x
4421 creat64 NEW: LFS
4422 csin NEW: ISO C 9x
4423 csinf NEW: ISO C 9x
4424 csinh NEW: ISO C 9x
4425 csinhf NEW: ISO C 9x
4426 csinhl NEW: ISO C 9x
4427 csinl NEW: ISO C 9x
4428 csqrt NEW: ISO C 9x
4429 csqrtf NEW: ISO C 9x
4430 csqrtl NEW: ISO C 9x
4431 ctan NEW: ISO C 9x
4432 ctanf NEW: ISO C 9x
4433 ctanh NEW: ISO C 9x
4434 ctanhf NEW: ISO C 9x
4435 ctanhl NEW: ISO C 9x
4436 ctanl NEW: ISO C 9x
4437 des_setparity NEW: Secure RPC
4438 ecb_crypt NEW: Secure RPC
4439 endutxent NEW: Unix98
4440 exp10 NEW: ISO C 9x
4441 exp10f NEW: ISO C 9x
4442 exp10l NEW: ISO C 9x
4443 exp2 NEW: ISO C 9x
4444 exp2f NEW: ISO C 9x
4445 exp2l NEW: ISO C 9x
4446 fattach NEW: STREAMS
4447 fdetach NEW: STREAMS
4448 fdim NEW: ISO C 9x
4449 fdimf NEW: ISO C 9x
4450 fdiml NEW: ISO C 9x
4451 feclearexcept NEW: ISO C 9x
4452 fegetenv NEW: ISO C 9x
4453 fegetexceptflag NEW: ISO C 9x
4454 fegetround NEW: ISO C 9x
4455 feholdexcept NEW: ISO C 9x
4456 feof_locked REMOVED
4457 feraiseexcept NEW: ISO C 9x
4458 ferror_locked REMOVED
4459 fesetenv NEW: ISO C 9x
4460 fesetexceptflag NEW: ISO C 9x
4461 fesetround NEW: ISO C 9x
4462 fetestexcept NEW: ISO C 9x
4463 feupdateenv NEW: ISO C 9x
4464 fflush_locked REMOVED
4465 ffsl NEW: GNU ext.
4466 ffsll NEW: GNU ext.
4467 fgetpos64 NEW: LFS
4468 fgets_unlocked NEW: GNU ext.
4469 fileno_locked REMOVED
4470 fma NEW: ISO C 9x
4471 fmaf NEW: ISO C 9x
4472 fmal NEW: ISO C 9x
4473 fmax NEW: ISO C 9x
4474 fmaxf NEW: ISO C 9x
4475 fmaxl NEW: ISO C 9x
4476 fmin NEW: ISO C 9x
4477 fminf NEW: ISO C 9x
4478 fminl NEW: ISO C 9x
4479 fmtmsg NEW: Unix98
4480 fopen64 NEW: LFS
4481 fputc_locked REMOVED
4482 fputs_unlocked NEW: GNU ext.
4483 fread_unlocked NEW: GNU ext.
4484 freopen64 NEW: LFS
4485 fseeko NEW: Unix98
4486 fsetpos64 NEW: LFS
4487 fstatfs64 NEW: LFS
4488 fstatvfs NEW: Unix98
4489 fstatvfs64 NEW: LFS
4490 ftello NEW: Unix98
4491 ftello64 NEW: LFS
4492 ftruncate64 NEW: LFS
4493 ftw64 NEW: LFS
4494 fwrite_unlocked NEW: GNU ext.
4495 gai_strerror NEW: IPv6
4496 gamma_r REMOVED
4497 gammaf_r REMOVED
4498 gammal_r REMOVED
4499 getchar_locked REMOVED
4500 getdate NEW: Unix98
4501 getdate_err NEW: Unix98
4502 getdate_r NEW: GNU ext.
4503 getmsg NEW: STREAMS
4504 getnameinfo NEW: IPv6
4505 getnetname NEW: Secure RPC
4506 getpmsg NEW: STREAMS
4507 getpt NEW: Unix98 PTY
4508 getrlimit64 NEW: LFS
4509 getutxent NEW: Unix98
4510 getutxid NEW: Unix98
4511 getutxline NEW: Unix98
4512 glob64 NEW: GNU ext.
4513 globfree64 NEW: GNU ext.
4514 gnu_get_libc_release NEW: GNU ext.
4515 gnu_get_libc_version NEW: GNU ext.
4516 grantpt NEW: Unix98 PTY
4517 host2netname NEW: Secure RPC
4518 iconv NEW: iconv
4519 iconv_close NEW: iconv
4520 iconv_open NEW: iconv
4521 if_freenameindex NEW: IPv6
4522 if_indextoname NEW: IPv6
4523 if_nameindex NEW: IPv6
4524 if_nametoindex NEW: IPv6
4525 in6addr_any NEW: IPv6
4526 in6addr_loopback NEW: IPv6
4527 inet6_isipv4mapped NEW: IPv6
4528 isastream NEW: STREAMS
4529 iswblank NEW: GNU ext.
4530 key_decryptsession NEW: Secure RPC
4531 key_decryptsession_pk NEW: Secure RPC
4532 key_encryptsession NEW: Secure RPC
4533 key_encryptsession_pk NEW: Secure RPC
4534 key_gendes NEW: Secure RPC
4535 key_get_conv NEW: Secure RPC
4536 key_secretkey_is_set NEW: Secure RPC
4537 key_setnet NEW: Secure RPC
4538 key_setsecret NEW: Secure RPC
4539 llrint NEW: ISO C 9x
4540 llrintf NEW: ISO C 9x
4541 llrintl NEW: ISO C 9x
4542 llround NEW: ISO C 9x
4543 llroundf NEW: ISO C 9x
4544 llroundl NEW: ISO C 9x
4545 log2 NEW: ISO C 9x
4546 log2f NEW: ISO C 9x
4547 log2l NEW: ISO C 9x
4548 lrint NEW: ISO C 9x
4549 lrintf NEW: ISO C 9x
4550 lrintl NEW: ISO C 9x
4551 lround NEW: ISO C 9x
4552 lroundf NEW: ISO C 9x
4553 lroundl NEW: ISO C 9x
4554 lseek64 NEW: LFS
4555 makecontext NEW: Unix98
4556 mempcpy NEW: GNU ext.
4557 mmap64 NEW: LFS
4558 moncontrol REMOVED
4559 modify_ldt NEW: kernel
4560 nan NEW: ISO C 9x
4561 nanf NEW: ISO C 9x
4562 nanl NEW: ISO C 9x
4563 nearbyint NEW: ISO C 9x
4564 nearbyintf NEW: ISO C 9x
4565 nearbyintl NEW: ISO C 9x
4566 netname2host NEW: Secure RPC
4567 netname2user NEW: Secure RPC
4568 nexttoward NEW: ISO C 9x
4569 nexttowardf NEW: ISO C 9x
4570 nexttowardl NEW: ISO C 9x
4571 nftw NEW: Unix98
4572 nftw64 NEW: LFS
4573 open64 NEW: LFS
4574 passwd2des NEW: Secure RPC
4575 pow10 NEW: GNU ext.
4576 pow10f NEW: GNU ext.
4577 pow10l NEW: GNU ext.
4578 pread NEW: Unix98
4579 pread64 NEW: LFS
4580 printf_size NEW: GNU ext.
4581 printf_size_info NEW: GNU ext.
4582 profil_counter REMOVED
4583 pthread_mutexattr_getkind_np REPLACED
4584 pthread_mutexattr_setkind_np REPLACED
4585 ptsname NEW: Unix98 PTY
4586 ptsname_r NEW: Unix98 PTY
4587 putc_locked REMOVED
4588 putchar_locked REMOVED
4589 putgrent NEW: GNU ext.
4590 putmsg NEW: STREAMS
4591 putpmsg NEW: STREAMS
4592 pututxline NEW: Unix98
4593 pwrite NEW: Unix98
4594 pwrite64 NEW: LFS
4595 readdir64 NEW: LFS
4596 readdir64_r NEW: LFS
4597 remquo NEW: ISO C 9x
4598 remquof NEW: ISO C 9x
4599 remquol NEW: ISO C 9x
4600 round NEW: ISO C 9x
4601 roundf NEW: ISO C 9x
4602 roundl NEW: ISO C 9x
4603 rtime NEW: GNU ext.
4604 scalbln NEW: ISO C 9x
4605 scalblnf NEW: ISO C 9x
4606 scalblnl NEW: ISO C 9x
4607 scandir64 NEW: LFS
4608 sendfile NEW: kernel
4609 setcontext NEW: Unix98
4610 setrlimit64 NEW: LFS
4611 setutxent NEW: Unix98
4612 sighold NEW: Unix98
4613 sigignore NEW: Unix98
4614 sigqueue NEW: POSIX.1b
4615 sigrelse NEW: Unix98
4616 sigset NEW: POSIX.1b
4617 sigtimedwait NEW: POSIX.1b
4618 sigwaitinfo NEW: POSIX.1b
4619 sincos NEW: GNU ext.
4620 sincosf NEW: GNU ext.
4621 sincosl NEW: GNU ext.
4622 statfs64 NEW: LFS
4623 statvfs NEW: Unix98
4624 statvfs64 NEW: LFS
4625 strcasestr NEW: GNU ext.
4626 strtoimax NEW: ISO C 9x
4627 strtoumax NEW: ISO C 9x
4628 strverscmp NEW: GNU ext.
4629 svcauthdes_stats NEW: Secure RPC
4630 svcunix_create NEW: sunrpc etx
4631 svcunixfd_create NEW: sunrpc ext
4632 swapcontext NEW: Unix98
4633 tcgetsid NEW: Unix98 PTY
4634 tdestroy NEW: GNU ext.
4635 tgamma NEW: ISO C 9x
4636 tgammaf NEW: ISO C 9x
4637 tgammal NEW: ISO C 9x
4638 tmpfile64 NEW: LFS
4639 trunc NEW: ISO C 9x
4640 truncate64 NEW: LFS
4641 truncf NEW: ISO C 9x
4642 truncl NEW: ISO C 9x
4643 umount2 NEW: kernel
4644 unlockpt NEW: Unix98 PTY
4645 updwtmpx NEW: Unix98
4646 user2netname NEW: Secure RPC
4647 utmpxname NEW: Unix98
4648 versionsort NEW: GNU ext.
4649 versionsort64 NEW: GNU ext.
4650 waitid NEW: Unix98
4651 wcscasecmp NEW: GNU ext.
4652 wcsncasecmp NEW: GNU ext.
4653 wcsnlen NEW: GNU ext.
4654 wcstoimax NEW: ISO C 9x
4655 wcstoll NEW: ISO C 9x
4656 wcstoull NEW: ISO C 9x
4657 wcstoumax NEW: ISO C 9x
4658 wcswcs NEW: Unix98
4659 wordexp NEW: POSIX.2
4660 wordfree NEW: POSIX.2
4661 write_profiling REMOVED
4662 xdecrypt NEW: Secure RPC
4663 xdr_authdes_cred NEW: Secure RPC
4664 xdr_authdes_verf NEW: Secure RPC
4665 xdr_cryptkeyarg NEW: Secure RPC
4666 xdr_cryptkeyarg2 NEW: Secure RPC
4667 xdr_cryptkeyres NEW: Secure RPC
4668 xdr_getcredres NEW: Secure RPC
4669 xdr_key_netstarg NEW: Secure RPC
4670 xdr_key_netstres NEW: Secure RPC
4671 xdr_keybuf NEW: Secure RPC
4672 xdr_keystatus NEW: Secure RPC
4673 xdr_netnamestr NEW: Secure RPC
4674 xdr_sizeof NEW: Secure RPC
4675 xdr_unixcred NEW: sunrpc ext
4676 xencrypt NEW: Secure RPC
4677 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4678 \f
4679 Version 2.0.6
4680
4681 * more bug fixes
4682
4683 \f
4684 Version 2.0.5
4685
4686 * more bug fixes
4687
4688 * inet_ntoa is thread-safe
4689
4690 * updwtmp is moved from libutil to libc
4691
4692 * rewrite of cbrt function
4693
4694 * update of timezone data
4695 \f
4696 Version 2.0.4
4697
4698 * more bug fixes
4699 \f
4700 Version 2.0.3
4701
4702 * more bug fixes
4703 \f
4704 Version 2.0.2
4705
4706 * more bug fixes
4707
4708 * add atoll function
4709
4710 * fix complex problems in Berkeley DB code
4711
4712 * fix math functions
4713 \f
4714 Version 2.0.1
4715
4716 * fixed lots of header problems (especially Linux/GNU specific)
4717
4718 * dynamic loader preserves all registers
4719
4720 * Roland McGrath provided support for handling of auxiliary objects in
4721 the ELF dynamic loader.
4722
4723 * support for parallel builds is improved
4724 \f
4725 Version 2.0
4726
4727 * GNU extensions are no longer declared by default. To enable them you
4728 must define the macro `_GNU_SOURCE' in your program or compile with
4729 `-D_GNU_SOURCE'.
4730
4731 * The library has changed from using GNU ld symbol aliases to using weak
4732 symbols where available. The ELF object file format supports weak
4733 symbols; GNU ld also supports weak symbols in the a.out format. (There
4734 is also now support for other GNU ld extensions in ELF. Use the
4735 `--with-elf' option to configure to indicate you have ELF, and
4736 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4737 of many files which contained only symbol aliases, reducing the size of
4738 the source and the compiled library; many other files were renamed to
4739 less cryptic names previously occupied by the symbol alias files.
4740 There is a new header file <elf.h> for programs which operate on
4741 files in the ELF format.
4742
4743 * Converted to Autoconf version 2, so `configure' has more options.
4744 Run `configure --help' to see the details.
4745
4746 * The library can now be configured to build profiling, highly-optimized
4747 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4748 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4749 `configure' enable building these extra libraries. The shared library is
4750 built by default when using both ELF and GNU ld. When shared libraries
4751 are enabled, the new library `-ldl' is available for arbitrary run-time
4752 loading of shared objects; its interface is defined in <dlfcn.h>. The
4753 new header file <link.h> gives access to the internals of the run-time
4754 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4755 application of same name on other systems and it provides information
4756 about dynamically linked binaries.
4757
4758 * The C library now provides the run-time support code for profiling
4759 executables compiled with `-pg'. Programs can control the profiling code
4760 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4761 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4762 the support for this file format was contributed by David Mosberger-Tang.
4763
4764 * The math code has been replaced with a math library based on fdlibm from
4765 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4766 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4767 functions. The math functions now reside in a separate library, so
4768 programs using them will need to use `-lm' their linking commands.
4769
4770 * John C. Bowman contributed optimized ix87 assembler inline functions.
4771
4772 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4773 mechanism similar to that found in Solaris 2. This is now used for the
4774 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4775 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4776 file controls what services are used for each individual database. This
4777 works by loading shared libraries with names specified in `nsswitch.conf',
4778 so service modules can be changed or added at any time without even
4779 relinking any program. Currently there are the file, db, and NIS based
4780 NSS services available.
4781
4782 * The new functions `strtoq' and `strtouq' parse integer values from
4783 strings, like `strtol' and `strtoul', but they return `long long int' and
4784 `unsigned long long int' values, respectively (64-bit quantities).
4785
4786 * The new functions `strtof' and `strtold' parse floating-point values from
4787 strings, like `strtod', but they return `float' and `long double' values,
4788 respectively (on some machines `double' and `long double' are the same).
4789
4790 * Ulrich Drepper has contributed new implementations of the floating-point
4791 printing and reading code used in the `printf' family of functions and
4792 `strtod', `strtof', and `strtold'. These new functions are perfectly
4793 accurate, and much faster than the old ones.
4794
4795 * The implementation of the POSIX locale model was completely rewritten by
4796 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4797 to compile the POSIX locale definition.
4798
4799 * The former dummy implementations of the strcoll and strxfrm function are
4800 now replaced by fully functional code contributed by Ulrich Drepper. The
4801 collation information comes from the POSIX locale definitions.
4802
4803 * The new header <langinfo.h> defines an interface for accessing
4804 various locale-dependent data (using the locale chosen with `setlocale').
4805
4806 * Ulrich Drepper has contributed a new suite of functions for operation on
4807 wide-character and multibyte-character strings, in <wchar.h>;
4808 and classification and case conversion of wide characters, in <wctype.h>.
4809 These new functions are conforming to the ISO C, Amendement 1 specification.
4810
4811 * There is now a second implementation of the standard I/O library available.
4812 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4813 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4814
4815 * You can now use positional parameter specifications in format strings
4816 for the `printf' and `scanf' families of functions. For example,
4817 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4818 ``Number 6, Mr Jones''. This is mainly useful when providing different
4819 format strings for different languages, whose grammars may dictate
4820 different orderings of the values being printed. To support this
4821 feature, the interface for `register_printf_handler' has changed; see
4822 the header file <printf.h> for details.
4823
4824 * The `printf' and `scanf' families of functions now understand a new
4825 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4826 to group numbers as indicated by the locale; for `scanf' and friends, this
4827 says to accept as valid only a number with all the proper grouping
4828 separators in the right places. In the default "C" locale, numbers are
4829 not grouped; but locales for specific countries will define the usual
4830 conventions (i.e. separate thousands with `,' in the US locale).
4831
4832 * The pgrp functions have been regularized, slightly incompatibly but much
4833 less confusingly. The core functions are now `getpgid' and `setpgid',
4834 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4835 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4836 provided for compatibility. There is no longer an incompatible `getpgrp'
4837 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4838
4839 * The new header file <fts.h> and suite of functions simplify programs that
4840 operate on directory trees. This code comes from 4.4 BSD.
4841
4842 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4843 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4844 NSS scheme used in glibc.
4845
4846 * There is a new malloc debugging hook `__memalign_hook'.
4847
4848 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4849 `unsigned int' in <sys/types.h>. These are for compatibility only and
4850 their use is discouraged.
4851
4852 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4853 done differently, so that it works even without GNU ld.
4854
4855 * New function `euidaccess' checks allowed access to a file like `access',
4856 but using the effective IDs instead of the real IDs.
4857
4858 * The time zone data files have been updated for the latest and greatest
4859 local time conventions of the countries of the world.
4860
4861 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4862 see <dirent.h>.
4863
4864 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4865 for formatting floating-point numbers. They are provided only for
4866 compatibility; new programs should use `sprintf' instead. There are
4867 also equivalent function for the `long double' floating-point type and
4868 all functions also exist in a reentrant form.
4869
4870 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4871 functions for maintaining the login-record files (primarily of use to
4872 system programs such as `login'), and convenient functions for
4873 allocating and initializing a pseudo-terminal (pty) device.
4874
4875 * Ulrich Drepper has contributed new support for System V style
4876 shared memory and IPC on systems that support it.
4877
4878 * Ulrich Drepper has contributed several miscellaneous new functions found
4879 in System V: The `hsearch' family of functions provide an effective
4880 implementation of hash tables; `a64l' and `l64a' provide a very simple
4881 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4882 number generator.
4883
4884 * Ulrich Drepper has contributed new reentrant counterparts for the
4885 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4886
4887 * Ulrich Drepper has contributed new, highly-optimized versions of several
4888 string functions for the i486/Pentium family of processors.
4889
4890 * Ulrich Drepper has updated the Linux-specific code, based largely
4891 on work done in Hongjiu Lu's version of GNU libc for Linux.
4892 The GNU library now supports Linux versions 2.0.10 and later,
4893 using the ELF object file format (i[3456]86-*-linux).
4894
4895 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4896
4897 * David Mosberger-Tang and Richard Henderson have ported the C library
4898 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4899 dynamic linking support for ELF/Alpha.
4900
4901 * Richard Henderson contributed several Alpha optimized assembler function
4902 for arithmetic and string handling.
4903
4904 * Ulrich Drepper has contributed a new set of message catalog functions to
4905 support multiple languages using the <libintl.h> interface, for use with
4906 his new package GNU gettext. Translation volunteers have contributed
4907 catalogs of the library's messages in Spanish, German, and Korean.
4908
4909 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4910 program and the `catgets' function for reading the catalog files it
4911 creates. (The <libintl.h> interface is preferred; we include the
4912 <nl_types.h> interface using `catgets' only for source compatibility with
4913 programs already written to use it.)
4914
4915 * New header file <values.h> gives SVID-compatible names for <limits.h>
4916 constants.
4917
4918 * Various new macros, declarations, and small header files for compatibility
4919 with 4.4 BSD.
4920
4921 * New function `group_member' is a convenient way to check if a process has
4922 a given effective group ID.
4923
4924 * When using GCC 2.7 and later, the socket functions are now declared in a
4925 special way so that passing an argument of type `struct sockaddr_in *',
4926 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4927 `struct sockaddr *' type, does not generate a type-clash warning.
4928
4929 * New function `error' declared in header file <error.h> is a convenient
4930 function for printing error messages and optionally exiting; this is the
4931 canonical function used in GNU programs. The new functions `err', `warn',
4932 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4933 doing the same thing.
4934
4935 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4936 POSIX.2 `glob' function to do ~ and {...} expansion.
4937
4938 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4939 `clearenv' which is used in POSIX.9 is also available.
4940
4941 * New function `getsid' returns session ID number on systems that support it.
4942
4943 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4944 files <db.h> and <mpool.h> provide a rich set of functions for several
4945 types of simple databases stored in memory and in files, and <ndbm.h> is
4946 an old `ndbm'-compatible interface using the `db' functions. Link with
4947 `-ldb' to get these functions.
4948
4949 * New macro `strdupa' copies a string like `strdup', but uses local stack
4950 space from `alloca' instead of dynamic heap space from `malloc'.
4951
4952 * New function `strnlen' is like `strlen' but searches only a given maximum
4953 number of characters for the null terminator. `stpncpy', `strndup' and
4954 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4955 function.
4956
4957 * New function `statfs' in header <sys/statfs.h>.
4958
4959 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4960 provide convenient functions for operating on blocks of null-terminated
4961 strings.
4962
4963 * A new suite of functions in <utmp.h> handle all the details of reading
4964 and writing the utmp file.
4965
4966 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4967 Thorsten Kukuk.
4968
4969 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4970 completely POSIX compliant and also implemented the extended functionality
4971 to handle alternate digit representation and alternate era date formats.
4972
4973 * Ulrich Drepper provided an implementation of the `strptime' function
4974 defined in XPG4.2 which transforms a string into a `struct tm' value.
4975
4976 * Paul Eggert provided the tzselect shell script as part of the timezone
4977 code. The shell script makes it easy to select the correct timezone
4978 specification.
4979
4980 * The implementation of the malloc family of functions is completely replaced
4981 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4982 The implementation uses the mmap function (if available) and it is
4983 optimized for the use in multi threaded programs.
4984
4985 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4986 functions. This new functionality is usable by specifying a special
4987 salt string and it is compatible with implementation on *BSD systems.
4988
4989 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4990 `getsubopt' to handle second level command line options, `bsd_signal'
4991 to access BSD style `signal' functionality, the obsolete `regexp' style
4992 expression matcher.
4993
4994 * the `lchown' function is available on system which support this
4995 functionality.
4996
4997 * The implementation of the shadow password handling function was contributed
4998 by Ulrich Drepper.
4999
5000 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5001
5002 * POSIX.1g support was added. The <sys/select.h> header is available,
5003 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5004 implementation of `getaddrinfo'.
5005 \f
5006 Version 1.09
5007
5008 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5009
5010 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5011 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5012
5013 * The new function `daemon' from 4.4 BSD is useful for server programs that
5014 want to put themselves in the background.
5015
5016 * Joel Sherrill has contributed support for several standalone boards that
5017 run without an operating system.
5018
5019 * `printf', `scanf' and friends now accept a `q' type modifier for long
5020 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5021
5022 * All of the code taken from BSD (notably most of the math and networking
5023 routines) has been updated from the BSD 4.4-Lite release.
5024
5025 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5026
5027 * The new functions `getdomainname' and `setdomainname' fetch or change the
5028 YP/NIS domain name. These are system calls which exist on systems which
5029 have YP (aka NIS).
5030
5031 * The time zone data files have been updated for the latest international
5032 conventions.
5033
5034 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5035 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5036 \f
5037 Version 1.08
5038
5039 * The C library now includes support for Sun RPC, from Sun's free
5040 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5041 are included. (There is still no support for YP.)
5042
5043 * Tom Quinn has contributed a port of the C library to SGI machines running
5044 Irix 4 (mips-sgi-irix4).
5045
5046 * The new `lockf' function is a simplified interface to the locking
5047 facilities of `fcntl', included for compatibility.
5048
5049 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5050
5051 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5052
5053 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5054 compatibility.
5055
5056 * The error message printed by `assert' for a failed assertion now includes
5057 the name of the program (if using GNU ld) and the name of the calling
5058 function (with versions of GCC that support this).
5059
5060 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5061
5062 * The library now includes the <sys/mman.h> header file and memory
5063 management functions `mmap', `munmap', `mprotect', `msync', and
5064 `madvise', on systems that support those facilities.
5065
5066 * The interface for `mcheck' has changed slightly: the function called to
5067 abort the program when an allocation inconsistency is detected now takes
5068 an argument that indicates the type of failure. The new function
5069 `mprobe' lets you request a consistency check for a particular block at
5070 any time (checks are normally done only when you call `free' or `realloc'
5071 on a block).
5072
5073 * It is now possible to easily cross-compile the C library, building on one
5074 system a library to run on another machine and/or operating system. All
5075 you need to do is set the variable `HOST_CC' in `configparms' to the
5076 native compiler for programs to run on the machine you are building on (a
5077 few generator programs are used on Unix systems); set `CC' to the
5078 cross-compiler.
5079
5080 * The new function `fexecve' (only implemented on the GNU system) executes
5081 a program file given a file descriptor already open on the file.
5082 \f
5083 Version 1.07
5084
5085 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5086 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5087
5088 * You can set the variable `libprefix' in `configparms' to specify a prefix
5089 to be prepended to installed library files; this makes it easy to install
5090 the GNU C library to be linked as `-lgnuc' or whatever.
5091
5092 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5093 copies a limited number of characters from a string, and returns the
5094 address of the last character written.
5095
5096 * You no longer need to check for whether the installed `stddef.h' is
5097 compatible with the GNU C library. configure now checks for you.
5098
5099 * You can now define a per-stream `fileno' function to convert the
5100 stream's cookie into an integral file descriptor.
5101
5102 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5103 allocates zero bytes of storage, and returns a unique pointer which
5104 you can pass to `realloc' or `free'. The behavior is undefined if
5105 you dereference this pointer.
5106
5107 * The C library now runs on Sony NEWS m68k machines running either
5108 NewsOS 3 or NewsOS 4.
5109
5110 * The new `syscall' function is a system-dependent primitive function
5111 for invoking system calls. It has the canonical behavior on Unix
5112 systems, including unreliable return values for some calls (such as
5113 `pipe', `fork' and `getppid').
5114
5115 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5116 to `EAGAIN', which is the preferred name. On systems whose kernels
5117 use two distinct codes, the C library now translates EWOULDBLOCK to
5118 EAGAIN in every system call function.
5119 \f
5120 Version 1.06
5121
5122 * The GNU C Library Reference Manual is now distributed with the library.
5123 `make dvi' will produce a DVI file of the printed manual.
5124 `make info' will produce Info files that you can read on line using C-h i
5125 in Emacs or the `info' program.
5126 Please send comments on the manual to bug-glibc-manual@gnu.org.
5127
5128 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5129
5130 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5131
5132 * Jason Merrill has contributed a port to the Sequent Symmetry running
5133 Dynix version 3 (i386-sequent-dynix).
5134
5135 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5136 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5137
5138 * New function `memory_warnings' lets you arrange to get warnings when
5139 malloc is running out of memory to allocate, like Emacs gives you.
5140
5141 * The C library now contains the relocating allocator used in Emacs 19 for
5142 its editing buffers. This allocator (ralloc) minimizes allocation
5143 overhead and fragmentation by moving allocated regions around whenever it
5144 needs to. You always refer to a ralloc'd region with a "handle" (a
5145 pointer to a pointer--an object of type `void **').
5146
5147 * There is a new `printf' format: `%m' gives you the string corresponding
5148 to the error code in `errno'.
5149
5150 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5151 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5152 pass, the `a' modifier says to fill in a `char **' you pass with a
5153 malloc'd string.
5154
5155 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5156 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5157 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5158
5159 * `mkstemp' is a traditional Unix function to atomically create and open a
5160 uniquely-named temporary file.
5161 \f
5162 Version 1.05
5163
5164 * The standard location for the file that says what the local timezone is
5165 has changed again. It is now `/usr/local/etc/localtime' (or more
5166 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5167
5168 * The distribution no longer contains any files with names longer than 14
5169 characters.
5170
5171 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5172 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5173
5174 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5175 \f
5176 Version 1.04
5177
5178 * The configuration process has changed quite a bit. The `configure'
5179 script is now used just like the configuration scripts for other GNU
5180 packages. The `sysdeps' directory hierarchy is much rearranged.
5181 The file `INSTALL' explains the new scheme in detail.
5182
5183 * The header files no longer need to be processed into ANSI C and
5184 traditional C versions. There is just one set of files to install, and
5185 it will work with ANSI or old C compilers (including `gcc -traditional').
5186
5187 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5188 MIPS DECStation running Ultrix 4.
5189
5190 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5191 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5192 made itself into a shared library.
5193
5194 * Yet further improved support for the i386, running 4.3 BSD-like systems
5195 (such as Mach 3 with the Unix single-server), or System V.
5196
5197 * New function `strncasecmp' to do case-insensitive string comparison
5198 with limited length.
5199
5200 * New function `strsep' is a reentrant alternative to `strtok'.
5201
5202 * New functions `scandir' and `alphasort' for searching directories.
5203
5204 * New function `setenv' is a better interface to `putenv'.
5205
5206 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5207 function for traversing a directory tree.
5208
5209 * The GNU obstack package is now also part of the C library.
5210 The new function `open_obstack_stream' creates a stdio stream that
5211 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5212 formatted output directly to an obstack.
5213
5214 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5215 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5216
5217 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5218
5219 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5220 things to your strings.
5221
5222 * There are some new test programs: test-fseek, testmb, and testrand.
5223
5224 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5225 These ports are not finished, but are a good starting place for really
5226 supporting those systems.
5227
5228 * `/etc/localtime' is now the standard location for the file that says what
5229 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5230 This follows the general principle that `/etc' is the place for all local
5231 configuration files.
5232
5233 * The C library header files now use `extern "C"' when used by the C++
5234 compiler, so the C library should now work with C++ code.
5235
5236 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5237 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5238 in <strings.h>.)
5239
5240 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5241 function which uses the `merge sort' algorithm, and is said to be
5242 significantly faster than the old GNU `qsort' function. Merge sort is
5243 now the standard `qsort' function. The new algorithm can require a lot
5244 of temporary storage; so, the old sorting function is called when the
5245 required storage is not available.
5246
5247 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5248 provides the Unix `crypt' function, plus some other entry points.
5249
5250 * The code and header files taken from 4.4 BSD have been updated with the
5251 latest files released from Berkeley.
5252 \f
5253 ----------------------------------------------------------------------
5254 Copyright information:
5255
5256 Copyright (C) 1992-2019 Free Software Foundation, Inc.
5257
5258 Permission is granted to anyone to make or distribute verbatim copies
5259 of this document as received, in any medium, provided that the
5260 copyright notice and this permission notice are preserved,
5261 thus giving the recipient permission to redistribute in turn.
5262
5263 Permission is granted to distribute modified versions
5264 of this document, or of portions of it,
5265 under the above conditions, provided also that they
5266 carry prominent notices stating who last changed them.
5267 \f
5268 Local variables:
5269 version-control: never
5270 fill-column: 76
5271 End: