]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
x86: Check IFUNC definition in unrelocated executable [BZ #20019]
[thirdparty/glibc.git] / NEWS
1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2018 Free Software Foundation, Inc.
3 See the end for copying conditions.
4
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
7 \f
8 Version 2.28.1
9
10 Major new features:
11
12 * The entry for the new Japanese era has been added for ja_JP locale.
13
14 Deprecated and removed features, and other changes affecting compatibility:
15
16 * For powercp64le ABI, Transactional Lock Elision is now enabled iff kernel
17 indicates that it will abort the transaction prior to entering the kernel
18 (PPC_FEATURE2_HTM_NOSC on hwcap2). On older kernels the transaction is
19 suspended, and this caused some undefined side-effects issues by aborting
20 transactions manually. Glibc avoided it by abort transactions manually on
21 each syscall, but it lead to performance issues on newer kernels where the
22 HTM state is saved and restore lazily (the state being saved even when the
23 process actually does not use HTM).
24
25 * The copy_file_range function fails with ENOSYS if the kernel does not
26 support the system call of the same name. Previously, user space
27 emulation was performed, but its behavior did not match the kernel
28 behavior, which was deemed too confusing. Applications which use the
29 copy_file_range function will have to be run on kernels which implement
30 the copy_file_range system call. Support for most architectures was added
31 in version 4.5 of the mainline Linux kernel.
32
33 The following bugs are resolved with this release:
34
35 [18035] Fix pldd hang
36 [19444] build failures with -O1 due to -Wmaybe-uninitialized
37 [20018] getaddrinfo should reject IP addresses with trailing characters
38 [20019] NULL pointer dereference in libc.so.6 IFUNC due to uninitialized GOT
39 [20209] localedata: Spelling mistake for Sunday in Greenlandic kl_GL
40 [20568] Fix crash in _IO_wfile_sync
41 [22927] libanl: properly cleanup if first helper thread creation failed
42 [23400] stdlib/test-bz22786.c creates temporary files in glibc source tree
43 [23497] readdir64@GLIBC_2.1 cannot parse the kernel directory stream
44 [23509] CET enabled glibc is incompatible with the older linker
45 [23521] nss_files aliases database file stream leak
46 [23538] pthread_cond_broadcast: Fix waiters-after-spinning case
47 [23562] signal: Use correct type for si_band in siginfo_t
48 [23578] regex: Fix memory overread in re_compile_pattern
49 [23579] libc: Errors misreported in preadv2
50 [23606] Missing ENDBR32 in sysdeps/i386/start.S
51 [23614] powerpc: missing CFI register information in __mpn_* functions
52 [23679] gethostid: Missing NULL check for gethostbyname_r result
53 [23709] Fix CPU string flags for Haswell-type CPUs
54 [23717] Fix stack overflow in stdlib/tst-setcontext9
55 [23821] si_band in siginfo_t has wrong type long int on sparc64
56 [23822] ia64 static libm.a is missing exp2f, log2f and powf symbols
57 [23864] libc: [riscv] missing kernel-features.h undefines
58 [23844] pthread_rwlock_trywrlock results in hang
59 [23927] Linux if_nametoindex() does not close descriptor (CVE-2018-19591)
60 [23972] __old_getdents64 uses wrong d_off value on overflow
61 [24018] gettext may return NULL
62 [24022] riscv may lack <asm/syscalls.h>
63 [24024] strerror() might set errno to ENOMEM due to -fno-math-error
64 [24027] malloc: Integer overflow in realloc
65 [24034] tst-cancel21-static fails with SIGBUS on pre-ARMv7 when using GCC 8
66 [24040] riscv64: unterminated call chain in __thread_start
67 [24097] Can't use 64-bit register for size_t in assembly codes for x32 (CVE-2019-6488)
68 [24155] x32 memcmp can treat positive length as 0 (if sign bit in RDX is set) (CVE-2019-7309)
69 [24161] __run_fork_handlers self-deadlocks in malloc/tst-mallocfork2
70 [24228] old x86 applications that use legacy libio crash on exit
71 [24476] dlfcn: Guard __dlerror_main_freeres with __libc_once_get (once)
72 [24744] io: Remove the copy_file_range emulation.
73 [25203] libio: Disable vtable validation for pre-2.1 interposed handles
74 [25204] Ignore LD_PREFER_MAP_32BIT_EXEC for SUID programs
75 [25225] ld.so fails to link on x86 if GCC defaults to -fcf-protection
76 [25232] No const correctness for strchr et al. for Clang++
77 [25414] 'glob' use-after-free bug (CVE-2020-1752)
78 [25423] Array overflow in backtrace on powerpc
79 [25933] Off by one error in __strncmp_avx2
80 [27130] "rep movsb" performance issue
81 [27177] GLIBC_TUNABLES=glibc.cpu.x86_ibt=on:glibc.cpu.x86_shstk=on doesn't work
82
83 Security related changes:
84
85 CVE-2018-19591: A file descriptor leak in if_nametoindex can lead to a
86 denial of service due to resource exhaustion when processing getaddrinfo
87 calls with crafted host names. Reported by Guido Vranken.
88
89 CVE-2019-6488: On x32, the size_t parameter may be passed in the lower
90 32 bits of a 64-bit register with with non-zero upper 32 bit. When it
91 happened, accessing the 32-bit size_t value as the full 64-bit register
92 in the assembly string/memory functions would cause a buffer overflow.
93 Reported by H.J. Lu.
94
95 CVE-2019-7309: x86-64 memcmp used signed Jcc instructions to check
96 size. For x86-64, memcmp on an object size larger than SSIZE_MAX
97 has undefined behavior. On x32, the size_t argument may be passed
98 in the lower 32 bits of the 64-bit RDX register with non-zero upper
99 32 bits. When it happened with the sign bit of RDX register set,
100 memcmp gave the wrong result since it treated the size argument as
101 zero. Reported by H.J. Lu.
102
103 CVE-2016-10739: The getaddrinfo function could successfully parse IPv4
104 addresses with arbitrary trailing characters, potentially leading to data
105 or command injection issues in applications.
106
107 CVE-2019-9169: Attempted case-insensitive regular-expression match
108 via proceed_next_node in posix/regexec.c leads to heap-based buffer
109 over-read. Reported by Hongxu Chen.
110
111 CVE-2019-19126: ld.so failed to ignore the LD_PREFER_MAP_32BIT_EXEC
112 environment variable during program execution after a security
113 transition, allowing local attackers to restrict the possible mapping
114 addresses for loaded libraries and thus bypass ASLR for a setuid
115 program. Reported by Marcin Koƛcielnicki.
116
117 CVE-2020-1752: A use-after-free vulnerability in the glob function when
118 expanding ~user has been fixed.
119
120 \f
121 Version 2.28
122
123 Major new features:
124
125 * The localization data for ISO 14651 is updated to match the 2016
126 Edition 4 release of the standard, this matches data provided by
127 Unicode 9.0.0. This update introduces significant improvements to the
128 collation of Unicode characters. This release deviates slightly from
129 the standard in that the collation element ordering for lowercase and
130 uppercase LATIN script characters is adjusted to ensure that regular
131 expressions with ranges like [a-z] and [A-Z] don't interleave e.g. A
132 is not matched by [a-z]. With the update many locales have been
133 updated to take advantage of the new collation information. The new
134 collation information has increased the size of the compiled locale
135 archive or binary locales.
136
137 * The GNU C Library can now be compiled with support for Intel CET, AKA
138 Intel Control-flow Enforcement Technology. When the library is built
139 with --enable-cet, the resulting glibc is protected with indirect
140 branch tracking (IBT) and shadow stack (SHSTK). CET-enabled glibc is
141 compatible with all existing executables and shared libraries. This
142 feature is currently supported on i386, x86_64 and x32 with GCC 8 and
143 binutils 2.29 or later. Note that CET-enabled glibc requires CPUs
144 capable of multi-byte NOPs, like x86-64 processors as well as Intel
145 Pentium Pro or newer. NOTE: --enable-cet has been tested for i686,
146 x86_64 and x32 on non-CET processors. --enable-cet has been tested
147 for x86_64 and x32 on CET SDVs, but Intel CET support hasn't been
148 validated for i686.
149
150 * The GNU C Library now has correct support for ABSOLUTE symbols
151 (SHN_ABS-relative symbols). Previously such ABSOLUTE symbols were
152 relocated incorrectly or in some cases discarded. The GNU linker can
153 make use of the newer semantics, but it must communicate it to the
154 dynamic loader by setting the ELF file's identification (EI_ABIVERSION
155 field) to indicate such support is required.
156
157 * Unicode 11.0.0 Support: Character encoding, character type info, and
158 transliteration tables are all updated to Unicode 11.0.0, using
159 generator scripts contributed by Mike FABIAN (Red Hat).
160
161 * <math.h> functions that round their results to a narrower type are added
162 from TS 18661-1:2014 and TS 18661-3:2015:
163
164 - fadd, faddl, daddl and corresponding fMaddfN, fMaddfNx, fMxaddfN and
165 fMxaddfNx functions.
166
167 - fsub, fsubl, dsubl and corresponding fMsubfN, fMsubfNx, fMxsubfN and
168 fMxsubfNx functions.
169
170 - fmul, fmull, dmull and corresponding fMmulfN, fMmulfNx, fMxmulfN and
171 fMxmulfNx functions.
172
173 - fdiv, fdivl, ddivl and corresponding fMdivfN, fMdivfNx, fMxdivfN and
174 fMxdivfNx functions.
175
176 * Two grammatical forms of month names are now supported for the following
177 languages: Armenian, Asturian, Catalan, Czech, Kashubian, Occitan, Ossetian,
178 Scottish Gaelic, Upper Sorbian, and Walloon. The following languages now
179 support two grammatical forms in abbreviated month names: Catalan, Greek,
180 and Kashubian.
181
182 * Newly added locales: Lower Sorbian (dsb_DE) and Yakut (sah_RU) also
183 include the support for two grammatical forms of month names.
184
185 * Building and running on GNU/Hurd systems now works without out-of-tree
186 patches.
187
188 * The renameat2 function has been added, a variant of the renameat function
189 which has a flags argument. If the flags are zero, the renameat2 function
190 acts like renameat. If the flag is not zero and there is no kernel
191 support for renameat2, the function will fail with an errno value of
192 EINVAL. This is different from the existing gnulib function renameatu,
193 which performs a plain rename operation in case of a RENAME_NOREPLACE
194 flags and a non-existing destination (and therefore has a race condition
195 that can clobber the destination inadvertently).
196
197 * The statx function has been added, a variant of the fstatat64
198 function with an additional flags argument. If there is no direct
199 kernel support for statx, glibc provides basic stat support based on
200 the fstatat64 function.
201
202 * IDN domain names in getaddrinfo and getnameinfo now use the system libidn2
203 library if installed. libidn2 version 2.0.5 or later is recommended. If
204 libidn2 is not available, internationalized domain names are not encoded
205 or decoded even if the AI_IDN or NI_IDN flags are passed to getaddrinfo or
206 getnameinfo. (getaddrinfo calls with non-ASCII names and AI_IDN will fail
207 with an encoding error.) Flags which used to change the IDN encoding and
208 decoding behavior (AI_IDN_ALLOW_UNASSIGNED, AI_IDN_USE_STD3_ASCII_RULES,
209 NI_IDN_ALLOW_UNASSIGNED, NI_IDN_USE_STD3_ASCII_RULES) have been
210 deprecated. They no longer have any effect.
211
212 * Parsing of dynamic string tokens in DT_RPATH, DT_RUNPATH, DT_NEEDED,
213 DT_AUXILIARY, and DT_FILTER has been expanded to support the full
214 range of ELF gABI expressions including such constructs as
215 '$ORIGIN$ORIGIN' (if valid). For SUID/GUID applications the rules
216 have been further restricted, and where in the past a dynamic string
217 token sequence may have been interpreted as a literal string it will
218 now cause a load failure. These load failures were always considered
219 unspecified behaviour from the perspective of the dynamic loader, and
220 for safety are now load errors e.g. /foo/${ORIGIN}.so in DT_NEEDED
221 results in a load failure now.
222
223 * Support for ISO C threads (ISO/IEC 9899:2011) has been added. The
224 implementation includes all the standard functions provided by
225 <threads.h>:
226
227 - thrd_current, thrd_equal, thrd_sleep, thrd_yield, thrd_create,
228 thrd_detach, thrd_exit, and thrd_join for thread management.
229
230 - mtx_init, mtx_lock, mtx_timedlock, mtx_trylock, mtx_unlock, and
231 mtx_destroy for mutual exclusion.
232
233 - call_once for function call synchronization.
234
235 - cnd_broadcast, cnd_destroy, cnd_init, cnd_signal, cnd_timedwait, and
236 cnd_wait for conditional variables.
237
238 - tss_create, tss_delete, tss_get, and tss_set for thread-local storage.
239
240 Application developers must link against libpthread to use ISO C threads.
241
242 Deprecated and removed features, and other changes affecting compatibility:
243
244 * The nonstandard header files <libio.h> and <_G_config.h> are no longer
245 installed. Software that was using either header should be updated to
246 use standard <stdio.h> interfaces instead.
247
248 * The stdio functions 'getc' and 'putc' are no longer defined as macros.
249 This was never required by the C standard, and the macros just expanded
250 to call alternative names for the same functions. If you hoped getc and
251 putc would provide performance improvements over fgetc and fputc, instead
252 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
253 necessary, flockfile and funlockfile.
254
255 * All stdio functions now treat end-of-file as a sticky condition. If you
256 read from a file until EOF, and then the file is enlarged by another
257 process, you must call clearerr or another function with the same effect
258 (e.g. fseek, rewind) before you can read the additional data. This
259 corrects a longstanding C99 conformance bug. It is most likely to affect
260 programs that use stdio to read interactive input from a terminal.
261 (Bug #1190.)
262
263 * The macros 'major', 'minor', and 'makedev' are now only available from
264 the header <sys/sysmacros.h>; not from <sys/types.h> or various other
265 headers that happen to include <sys/types.h>. These macros are rarely
266 used, not part of POSIX nor XSI, and their names frequently collide with
267 user code; see https://sourceware.org/bugzilla/show_bug.cgi?id=19239 for
268 further explanation.
269
270 <sys/sysmacros.h> is a GNU extension. Portable programs that require
271 these macros should first include <sys/types.h>, and then include
272 <sys/sysmacros.h> if __GNU_LIBRARY__ is defined.
273
274 * The tilegx*-*-linux-gnu configurations are no longer supported.
275
276 * The obsolete function ustat is no longer available to newly linked
277 binaries; the headers <ustat.h> and <sys/ustat.h> have been removed. This
278 function has been deprecated in favor of fstatfs and statfs.
279
280 * The obsolete function nfsservctl is no longer available to newly linked
281 binaries. This function was specific to systems using the Linux kernel
282 and could not usefully be used with the GNU C Library on systems with
283 version 3.1 or later of the Linux kernel.
284
285 * The obsolete function name llseek is no longer available to newly linked
286 binaries. This function was specific to systems using the Linux kernel
287 and was not declared in a header. Programs should use the lseek64 name
288 for this function instead.
289
290 * The AI_IDN_ALLOW_UNASSIGNED and NI_IDN_ALLOW_UNASSIGNED flags for the
291 getaddrinfo and getnameinfo functions have been deprecated. The behavior
292 previously selected by them is now always enabled.
293
294 * The AI_IDN_USE_STD3_ASCII_RULES and NI_IDN_USE_STD3_ASCII_RULES flags for
295 the getaddrinfo and getnameinfo functions have been deprecated. The STD3
296 restriction (rejecting '_' in host names, among other things) has been
297 removed, for increased compatibility with non-IDN name resolution.
298
299 * The fcntl function now have a Long File Support variant named fcntl64. It
300 is added to fix some Linux Open File Description (OFD) locks usage on non
301 LFS mode. As for others *64 functions, fcntl64 semantics are analogous with
302 fcntl and LFS support is handled transparently. Also for Linux, the OFD
303 locks act as a cancellation entrypoint.
304
305 * The obsolete functions encrypt, encrypt_r, setkey, setkey_r, cbc_crypt,
306 ecb_crypt, and des_setparity are no longer available to newly linked
307 binaries, and the headers <rpc/des_crypt.h> and <rpc/rpc_des.h> are no
308 longer installed. These functions encrypted and decrypted data with the
309 DES block cipher, which is no longer considered secure. Software that
310 still uses these functions should switch to a modern cryptography library,
311 such as libgcrypt.
312
313 * Reflecting the removal of the encrypt and setkey functions above, the
314 macro _XOPEN_CRYPT is no longer defined. As a consequence, the crypt
315 function is no longer declared unless _DEFAULT_SOURCE or _GNU_SOURCE is
316 enabled.
317
318 * The obsolete function fcrypt is no longer available to newly linked
319 binaries. It was just another name for the standard function crypt,
320 and it has not appeared in any header file in many years.
321
322 * We have tentative plans to hand off maintenance of the passphrase-hashing
323 library, libcrypt, to a separate development project that will, we hope,
324 keep up better with new passphrase-hashing algorithms. We will continue
325 to declare 'crypt' in <unistd.h>, and programs that use 'crypt' or
326 'crypt_r' should not need to change at all; however, distributions will
327 need to install <crypt.h> and libcrypt from a separate project.
328
329 In this release, if the configure option --disable-crypt is used, glibc
330 will not install <crypt.h> or libcrypt, making room for the separate
331 project's versions of these files. The plan is to make this the default
332 behavior in a future release.
333
334 Changes to build and runtime requirements:
335
336 GNU make 4.0 or later is now required to build glibc.
337
338 Security related changes:
339
340 CVE-2016-6261, CVE-2016-6263, CVE-2017-14062: Various vulnerabilities have
341 been fixed by removing the glibc-internal IDNA implementation and using
342 the system-provided libidn2 library instead. Originally reported by Hanno
343 Böck and Christian Weisgerber.
344
345 CVE-2017-18269: An SSE2-based memmove implementation for the i386
346 architecture could corrupt memory. Reported by Max Horn.
347
348 CVE-2018-11236: Very long pathname arguments to realpath function could
349 result in an integer overflow and buffer overflow. Reported by Alexey
350 Izbyshev.
351
352 CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
353 architecture could write beyond the target buffer, resulting in a buffer
354 overflow. Reported by Andreas Schwab.
355
356 The following bugs are resolved with this release:
357
358 [1190] stdio: fgetc()/fread() behaviour is not POSIX compliant
359 [6889] manual: 'PWD' mentioned but not specified
360 [13575] libc: SSIZE_MAX defined as LONG_MAX is inconsistent with ssize_t,
361 when __WORDSIZE != 64
362 [13762] regex: re_search etc. should return -2 on memory exhaustion
363 [13888] build: /tmp usage during testing
364 [13932] math: dbl-64 pow unexpectedly slow for some inputs
365 [14092] nptl: Support C11 threads
366 [14095] localedata: Review / update collation data from Unicode / ISO
367 14651
368 [14508] libc: -Wformat warnings
369 [14553] libc: Namespace pollution loff_t in sys/types.h
370 [14890] libc: Make NT_PRFPREG canonical.
371 [15105] libc: Extra PLT references with -Os
372 [15512] libc: __bswap_constant_16 not compiled when -Werror -Wsign-
373 conversion is given
374 [16335] manual: Feature test macro documentation incomplete and out of
375 date
376 [16552] libc: Unify umount implementations in terms of umount2
377 [17082] libc: htons et al.: statement-expressions prevent use on global
378 scope with -O1 and higher
379 [17343] libc: Signed integer overflow in /stdlib/random_r.c
380 [17438] localedata: pt_BR: wrong d_fmt delimiter
381 [17662] libc: please implement binding for the new renameat2 syscall
382 [17721] libc: __restrict defined as /* Ignore */ even in c11
383 [17979] libc: inconsistency between uchar.h and stdint.h
384 [18018] dynamic-link: Additional $ORIGIN handling issues (CVE-2011-0536)
385 [18023] libc: extend_alloca is broken (questionable pointer comparison,
386 horrible machine code)
387 [18124] libc: hppa: setcontext erroneously returns -1 as exit code for
388 last constant.
389 [18471] libc: llseek should be a compat symbol
390 [18473] soft-fp: [powerpc-nofpu] __sqrtsf2, __sqrtdf2 should be compat
391 symbols
392 [18991] nss: nss_files skips large entry in database
393 [19239] libc: Including stdlib.h ends up with macros major and minor being
394 defined
395 [19463] libc: linknamespace failures when compiled with -Os
396 [19485] localedata: csb_PL: Update month translations + add yesstr/nostr
397 [19527] locale: Normalized charset name not recognized by setlocale
398 [19667] string: Missing Sanity Check for malloc calls in file 'testcopy.c'
399 [19668] libc: Missing Sanity Check for malloc() in file 'tst-setcontext-
400 fpscr.c'
401 [19728] network: out of bounds stack read in libidn function
402 idna_to_ascii_4i (CVE-2016-6261)
403 [19729] network: out of bounds heap read on invalid utf-8 inputs in
404 stringprep_utf8_nfkc_normalize (CVE-2016-6263)
405 [19818] dynamic-link: Absolute (SHN_ABS) symbols incorrectly relocated by
406 the base address
407 [20079] libc: Add SHT_X86_64_UNWIND to elf.h
408 [20251] libc: 32bit programs pass garbage in struct flock for OFD locks
409 [20419] dynamic-link: files with large allocated notes crash in
410 open_verify
411 [20530] libc: bswap_16 should use __builtin_bswap16() when available
412 [20890] dynamic-link: ldconfig: fsync the files before atomic rename
413 [20980] manual: CFLAGS environment variable replaces vital options
414 [21163] regex: Assertion failure in pop_fail_stack when executing a
415 malformed regexp (CVE-2015-8985)
416 [21234] manual: use of CFLAGS makes glibc detect no optimization
417 [21269] dynamic-link: i386 sigaction sa_restorer handling is wrong
418 [21313] build: Compile Error GCC 5.4.0 MIPS with -0S
419 [21314] build: Compile Error GCC 5.2.0 MIPS with -0s
420 [21508] locale: intl/tst-gettext failure with latest msgfmt
421 [21547] localedata: Tibetan script collation broken (Dzongkha and Tibetan)
422 [21812] network: getifaddrs() returns entries with ifa_name == NULL
423 [21895] libc: ppc64 setjmp/longjmp not fully interoperable with static
424 dlopen
425 [21942] dynamic-link: _dl_dst_substitute incorrectly handles $ORIGIN: with
426 AT_SECURE=1
427 [22241] localedata: New locale: Yakut (Sakha) locale for Russia (sah_RU)
428 [22247] network: Integer overflow in the decode_digit function in
429 puny_decode.c in libidn (CVE-2017-14062)
430 [22342] nscd: NSCD not properly caching netgroup
431 [22391] nptl: Signal function clear NPTL internal symbols inconsistently
432 [22550] localedata: es_ES locale (and other es_* locales): collation
433 should treat ñ as a primary different character, sync the collation
434 for Spanish with CLDR
435 [22638] dynamic-link: sparc: static binaries are broken if glibc is built
436 by gcc configured with --enable-default-pie
437 [22639] time: year 2039 bug for localtime etc. on 64-bit platforms
438 [22644] string: memmove-sse2-unaligned on 32bit x86 produces garbage when
439 crossing 2GB threshold (CVE-2017-18269)
440 [22646] localedata: redundant data (LC_TIME) for es_CL, es_CU, es_EC and
441 es_BO
442 [22735] time: Misleading typo in time.h source comment regarding
443 CLOCKS_PER_SECOND
444 [22753] libc: preadv2/pwritev2 fallback code should handle offset=-1
445 [22761] libc: No trailing `%n' conversion specifier in FMT passed from
446 `__assert_perror_fail ()' to `__assert_fail_base ()'
447 [22766] libc: all glibc internal dlopen should use RTLD_NOW for robust
448 dlopen failures
449 [22786] libc: Stack buffer overflow in realpath() if input size is close
450 to SSIZE_MAX (CVE-2018-11236)
451 [22787] dynamic-link: _dl_check_caller returns false when libc is linked
452 through an absolute DT_NEEDED path
453 [22792] build: tcb-offsets.h dependency dropped
454 [22797] libc: pkey_get() uses non-reserved name of argument
455 [22807] libc: PTRACE_* constants missing for powerpc
456 [22818] glob: posix/tst-glob_lstat_compat failure on alpha
457 [22827] dynamic-link: RISC-V ELF64 parser mis-reads flag in ldconfig
458 [22830] malloc: malloc_stats doesn't restore cancellation state on stderr
459 [22848] localedata: ca_ES: update date definitions from CLDR
460 [22862] build: _DEFAULT_SOURCE is defined even when _ISOC11_SOURCE is
461 [22884] math: RISCV fmax/fmin handle signalling NANs incorrectly
462 [22896] localedata: Update locale data for an_ES
463 [22902] math: float128 test failures with GCC 8
464 [22918] libc: multiple common of `__nss_shadow_database'
465 [22919] libc: sparc32: backtrace yields infinite backtrace with
466 makecontext
467 [22926] libc: FTBFS on powerpcspe
468 [22932] localedata: lt_LT: Update of abbreviated month names from CLDR
469 required
470 [22937] localedata: Greek (el_GR, el_CY) locales actually need ab_alt_mon
471 [22947] libc: FAIL: misc/tst-preadvwritev2
472 [22963] localedata: cs_CZ: Add alternative month names
473 [22987] math: [powerpc/sparc] fdim inlines errno, exceptions handling
474 [22996] localedata: change LC_PAPER to en_US in es_BO locale
475 [22998] dynamic-link: execstack tests are disabled when SELinux is
476 disabled
477 [23005] network: Crash in __res_context_send after memory allocation
478 failure
479 [23007] math: strtod cannot handle -nan
480 [23024] nss: getlogin_r is performing NSS lookups when loginid isn't set
481 [23036] regex: regex equivalence class regression
482 [23037] libc: initialize msg_flags to zero for sendmmsg() calls
483 [23069] libc: sigaction broken on riscv64-linux-gnu
484 [23094] localedata: hr_HR: wrong thousands_sep and mon_thousands_sep
485 [23102] dynamic-link: Incorrect parsing of multiple consecutive $variable
486 patterns in runpath entries (e.g. $ORIGIN$ORIGIN)
487 [23137] nptl: s390: pthread_join sometimes block indefinitely (on 31bit
488 and libc build with -Os)
489 [23140] localedata: More languages need two forms of month names
490 [23145] libc: _init/_fini aren't marked as hidden
491 [23152] localedata: gd_GB: Fix typo in "May" (abbreviated)
492 [23171] math: C++ iseqsig for long double converts arguments to double
493 [23178] nscd: sudo will fail when it is run in concurrent with commands
494 that changes /etc/passwd
495 [23196] string: __mempcpy_avx512_no_vzeroupper mishandles large copies
496 (CVE-2018-11237)
497 [23206] dynamic-link: static-pie + dlopen breaks debugger interaction
498 [23208] localedata: New locale - Lower Sorbian (dsb)
499 [23233] regex: Memory leak in build_charclass_op function in file
500 posix/regcomp.c
501 [23236] stdio: Harden function pointers in _IO_str_fields
502 [23250] nptl: Offset of __private_ss differs from GCC
503 [23253] math: tgamma test suite failures on i686 with -march=x86-64
504 -mtune=generic -mfpmath=sse
505 [23259] dynamic-link: Unsubstituted ${ORIGIN} remains in DT_NEEDED for
506 AT_SECURE
507 [23264] libc: posix_spawnp wrongly executes ENOEXEC in non compat mode
508 [23266] nis: stringop-truncation warning with new gcc8.1 in nisplus-
509 parser.c
510 [23272] math: fma(INFINITY,INFIITY,0.0) should be INFINITY
511 [23277] math: nan function should not have const attribute
512 [23279] math: scanf and strtod wrong for some hex floating-point
513 [23280] math: wscanf rounds wrong; wcstod is ok for negative numbers and
514 directed rounding
515 [23290] localedata: IBM273 is not equivalent to ISO-8859-1
516 [23303] build: undefined reference to symbol
517 '__parse_hwcap_and_convert_at_platform@@GLIBC_2.23'
518 [23307] dynamic-link: Absolute symbols whose value is zero ignored in
519 lookup
520 [23313] stdio: libio vtables validation and standard file object
521 interposition
522 [23329] libc: The __libc_freeres infrastructure is not properly run across
523 DSO boundaries.
524 [23349] libc: Various glibc headers no longer compatible with
525 <linux/time.h>
526 [23351] malloc: Remove unused code related to heap dumps and malloc
527 checking
528 [23363] stdio: stdio-common/tst-printf.c has non-free license
529 [23396] regex: Regex equivalence regression in single-byte locales
530 [23422] localedata: oc_FR: More updates of locale data
531 [23442] build: New warning with GCC 8
532 [23448] libc: Out of bounds access in IBM-1390 converter
533 [23456] libc: Wrong index_cpu_LZCNT
534 [23458] build: tst-get-cpu-features-static isn't added to tests
535 [23459] libc: COMMON_CPUID_INDEX_80000001 isn't populated for Intel
536 processors
537 [23467] dynamic-link: x86/CET: A property note parser bug
538 [24112] network: Do not send DNS queries for non-host names (where all
539 answers will be rejected)
540
541 \f
542 Version 2.27
543
544 Major new features:
545
546 * The GNU C Library can now be compiled with support for building static
547 PIE executables (See --enable-static-pie in INSTALL). These static PIE
548 executables are like static executables but can be loaded at any address
549 and provide additional security hardening benefits at the cost of some
550 memory and performance. When the library is built with --enable-static-pie
551 the resulting libc.a is usable with GCC 8 and above to create static PIE
552 executables using the GCC option '-static-pie'. This feature is currently
553 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
554 aarch64 with binutils 2.30 or later.
555
556 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
557 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
558 H.J. Lu from Intel.
559
560 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
561
562 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
563
564 * In order to support faster and safer process termination the malloc API
565 family of functions will no longer print a failure address and stack
566 backtrace after detecting heap corruption. The goal is to minimize the
567 amount of work done after corruption is detected and to avoid potential
568 security issues in continued process execution. Reducing shutdown time
569 leads to lower overall process restart latency, so there is benefit both
570 from a security and performance perspective.
571
572 * The abort function terminates the process immediately, without flushing
573 stdio streams. Previous glibc versions used to flush streams, resulting
574 in deadlocks and further data corruption. This change also affects
575 process aborts as the result of assertion failures.
576
577 * On platforms where long double has the IEEE binary128 format (aarch64,
578 alpha, mips64, riscv, s390 and sparc), the math library now implements
579 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
580 These are the same interfaces added in version 2.26 for some platforms where
581 this format is supported but is not the format of long double.
582
583 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
584 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
585 implements interfaces for that type, as defined by ISO/IEC TS
586 18661-3:2015. These are corresponding interfaces to those supported for
587 _Float128.
588
589 * The math library now implements interfaces for the _Float32, _Float64 and
590 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
591 corresponding interfaces to those supported for _Float128.
592
593 * glibc now implements the memfd_create and mlock2 functions on Linux.
594
595 * Support for memory protection keys was added. The <sys/mman.h> header now
596 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
597 pkey_get.
598
599 * The copy_file_range function was added.
600
601 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
602
603 * The ldconfig utility now processes `include' directives using the C/POSIX
604 collation ordering. Previous glibc versions used locale-specific
605 ordering, the change might break systems that relied on that.
606
607 * Support for two grammatical forms of month names has been added.
608 In a call to strftime, the "%B" and "%b" format specifiers will now
609 produce the grammatical form required when the month is used as part
610 of a complete date. New "%OB" and "%Ob" specifiers produce the form
611 required when the month is named by itself. For instance, in Greek
612 and in many Slavic and Baltic languages, "%B" will produce the month
613 in genitive case, and "%OB" will produce the month in nominative case.
614
615 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
616 are all valid and will all accept any known form of month
617 name---standalone or complete, abbreviated or full. In a call to
618 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
619 the strings used by "%B" and "%b", respectively. New query
620 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
621 used by "%OB" and "%Ob", respectively.
622
623 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
624 define the strings for %OB and %Ob, respectively; these have the
625 same syntax as "mon" and "abmon". These arrays are optional; if they
626 are not provided then they have the same content as "mon" and "abmon",
627 respectively.
628
629 These features are provided for locales which define "alt_mon" and/or
630 "ab_alt_mon" in their locale source data. This release includes such
631 alternative month name data for the following languages: Belarusian,
632 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
633
634 This feature is currently a GNU extension, but it is expected to
635 be added to the next revision of POSIX, and it is also already
636 available on some BSD-derived operating systems.
637
638 This feature will cause existing statically compiled applications
639 to fail to load locales and fall back to the builtin C/POSIX locales.
640 See notes below for other changes affecting compatibility.
641
642 * Support for the RISC-V ISA running on Linux has been added. This port
643 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
644 for the following ISA and ABI pairs:
645
646 - rv64imac lp64
647 - rv64imafdc lp64
648 - rv64imafdc lp64d
649
650 Deprecated and removed features, and other changes affecting compatibility:
651
652 * Statically compiled applications attempting to load locales compiled for the
653 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
654 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
655 support for two grammatical forms of the month names, also extends the locale
656 data binary format. Static applications needing locale support must be
657 recompiled to match the runtime and data they are deployed with. In some
658 distributions there is an upgrade window where dynamically linked applications
659 may use a new library but the old locale data and also fall back to the
660 builtin C/POSIX locales; restarting the application process is sufficient to
661 fix this.
662
663 * Support for statically linked applications which call dlopen is deprecated
664 and will be removed in a future version of glibc. Applications which call
665 dlopen need to be linked dynamically instead.
666
667 * Support for old programs which use internal stdio data structures and
668 functions is deprecated. This includes programs which use the C++ streams
669 provided by libstdc++ in GCC 2.95. Programs which use the internal
670 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
671 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
672 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
673 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
674 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
675 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
676 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
677 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
678 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
679 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
680 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
681 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
682 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
683 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
684 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
685 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
686 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
687 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
688 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
689 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
690 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
691 future version of glibc. Unlike other symbol removals, these old
692 applications will not be supported using compatibility symbols.
693
694 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
695 defined by <sys/ptrace.h>.
696
697 * libm no longer supports SVID error handling (calling a user-provided
698 matherr function on error) or the _LIB_VERSION variable to control error
699 handling. (SVID error handling and the _LIB_VERSION variable still work
700 for binaries linked against older versions of the GNU C Library.) The
701 libieee.a library is no longer provided. math.h no longer defines struct
702 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
703 TLOSS, PLOSS and HUGE.
704
705 * The libm functions pow10, pow10f and pow10l are no longer supported for
706 new programs. Programs should use the standard names exp10, exp10f and
707 exp10l for these functions instead.
708
709 * The mcontext_t type is no longer the same as struct sigcontext. On
710 platforms where it was previously the same, this changes the C++ name
711 mangling for interfaces involving this type.
712
713 * The add-ons mechanism for building additional packages at the same time as
714 glibc has been removed. The --enable-add-ons configure option is now
715 ignored.
716
717 * The --without-fp configure option is now ignored. Whether hardware
718 floating-point instructions are used is now configured based on whether
719 the compiler used at configure time (without any options implied by a
720 --with-cpu= configure option) uses such instructions.
721
722 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
723 the specified string can be parsed as a domain name.
724
725 * In the malloc_info output, the <heap> element may contain another <aspace>
726 element, "subheaps", which contains the number of sub-heaps.
727
728 * The libresolv function p_secstodate is no longer supported for new
729 programs.
730
731 * The tilepro-*-linux-gnu configuration is no longer supported.
732
733 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
734 and will be removed in a future release. Software that is still using
735 either header should be updated to use standard <stdio.h> interfaces
736 instead.
737
738 libio.h was originally the header for a set of supported GNU extensions,
739 but they have not been maintained as such in many years, they are now
740 standing in the way of improvements to stdio, and we don't think there are
741 any remaining external users. _G_config.h was never intended for public
742 use, but predates the bits convention.
743
744 Changes to build and runtime requirements:
745
746 * bison version 2.7 or later is required to generate code in the 'intl'
747 subdirectory.
748
749 Security related changes:
750
751 CVE-2009-5064: The ldd script would sometimes run the program under
752 examination directly, without preventing code execution through the
753 dynamic linker. (The glibc project disputes that this is a security
754 vulnerability; only trusted binaries must be examined using the ldd
755 script.)
756
757 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
758 suffered from a one-byte overflow during ~ operator processing (either
759 on the stack or the heap, depending on the length of the user name).
760 Reported by Tim RĂŒhsen.
761
762 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
763 would sometimes fail to free memory allocated during ~ operator
764 processing, leading to a memory leak and, potentially, to a denial
765 of service.
766
767 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
768 without GLOB_NOESCAPE, could write past the end of a buffer while
769 unescaping user names. Reported by Tim RĂŒhsen.
770
771 CVE-2017-17426: The malloc function, when called with an object size near
772 the value SIZE_MAX, would return a pointer to a buffer which is too small,
773 instead of NULL. This was a regression introduced with the new malloc
774 thread cache in glibc 2.26. Reported by Iain Buclaw.
775
776 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
777 to the allocation of too much memory. (This is not a security bug per se,
778 it is mentioned here only because of the CVE assignment.) Reported by
779 Qualys.
780
781 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
782 of the number of search path components. (This is not a security
783 vulnerability per se because no trust boundary is crossed if the fix for
784 CVE-2017-1000366 has been applied, but it is mentioned here only because
785 of the CVE assignment.) Reported by Qualys.
786
787 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
788 for AT_SECURE or SUID binaries could be used to load libraries from the
789 current directory.
790
791 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
792 succeeds without returning an absolute path due to unexpected behaviour
793 of the Linux kernel getcwd syscall. Reported by halfdog.
794
795 CVE-2018-6485: The posix_memalign and memalign functions, when called with
796 an object size near the value of SIZE_MAX, would return a pointer to a
797 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
798
799 CVE-2018-6551: The malloc function, when called with an object size near
800 the value of SIZE_MAX, would return a pointer to a buffer which is too
801 small, instead of NULL.
802
803 The following bugs are resolved with this release:
804
805 [866] glob: glob should match dangling symlinks
806 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
807 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
808 locale
809 [5997] math: Very slow execution of sinf function
810 [10580] localedata: hr_HR: updated locale
811 [10871] locale: 'mon' array should contain both nominative and genitive
812 cases
813 [12349] localedata: eu_ES: incorrect thousands separator
814 [13605] localedata: shn_MM: new Shan locale
815 [13805] localedata: ru_RU: currency should use ',' as radix point
816 [13953] localedata: km_KH: locale update
817 [13994] localedata: mjw_IN: new locale
818 [14121] build: make writes .mo files in po directory
819 [14333] libc: Fix the race between atexit() and exit()
820 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
821 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
822 "Bangla"
823 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
824 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
825 width Latin characters
826 [15332] localedata: es_CU: locale update
827 [15436] stdio: Don't close or flush stdio streams on abort
828 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
829 letters
830 [16148] localedata: ca_ES: incorrect thousands separator
831 [16750] dynamic-link: ldd should not try to execute the binaries
832 (CVE-2009-5064)
833 [16777] localedata: pl_PL: incorrect thousands separator in locale
834 [16905] localedata: hanzi: new collation
835 [17563] localedata: cmn_TW: add hanzi collation
836 [17750] localedata: wrong collation order of diacritics in most locales
837 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
838 if malloc succeeded
839 [17956] build: Build fails on missing definitions from header file
840 nss/nss.h when Mozilla NSS is used for cryptography
841 [18203] libc: realpath() does not handle unreachable paths correctly
842 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
843 [18812] localedata: kab_DZ: new Kabyle Algeria locale
844 [18822] libc: Internal functions are called via PLT
845 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
846 [19170] libc: __gmon_start__ defined in hppa in crtn.S
847 [19574] libc: glibc should support building static PIE binaries
848 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
849 U+309A
850 [19971] glob: glob: Do not skip entries with zero d_ino values
851 [19982] localedata: fr.po: spelling mistake for error code EXDEV
852 [20008] localedata: km_KH: convert to translit_neutral
853 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
854 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
855 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
856 [20482] localedata: de_CH: abbreviated weekdays should be two letters
857 [20498] localedata: miq_NI: new MĂ­skitu / Miskito (miq) language locale
858 for Nicaragua
859 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
860 they're set, wrong errors returned by gaih_inet when lookup functions
861 are not found.
862 [20756] localedata: [PATCH] Use Unicode wise thousands separator
863 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
864 access
865 [20952] localedata: yuw_PG: new locale
866 [21084] localedata: charmaps/IBM858: new codepage
867 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
868 [21242] libc: assert gives pedantic warning in old gcc versions
869 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
870 __regcall calling convention
871 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
872 [21326] libc: C99 functions are not declared for C++11 and later if
873 _GNU_SOURCE is not predefined by g++
874 [21457] libc: sys/ucontext.h namespace
875 [21530] libc: tmpfile() should be implemented using O_TMPFILE
876 [21660] math: GCC fails to compile a formula with tgmath.h
877 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
878 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
879 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
880 [21684] math: tgmath.h handling of complex integers
881 [21685] math: tgmath.h handling of bit-fields
882 [21686] math: tgmath.h handling of __int128
883 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
884 locale
885 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
886 protector=all
887 [21750] localedata: column width of characters incompatible with classical
888 wcwidth
889 [21754] malloc: malloc: Perform as little work as possible after heap
890 consistency check failures
891 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
892 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
893 [21791] string: Unused XXX_chk_XXX functions in libc.a
894 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
895 PIE
896 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
897 locales
898 [21845] localedata: Added new Locale bho_NP
899 [21853] localedata: Fix abday Which looks same as day in zh_SG
900 [21854] localedata: Added New Locale en_SC
901 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
902 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
903 _dl_runtime_resolve_avx_slow
904 [21885] network: getaddrinfo: gethosts does not release resolver context
905 on memory allocation failure
906 [21899] libc: XPG4.2 sigaction namespace
907 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
908 the suspect)
909 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
910 default (i386)
911 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
912 errno value, causing getaddrinfo to fail
913 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
914 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
915 instead of EAI_NODATA
916 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
917 constant PTRACE_SEIZE_DEVEL
918 [21930] math: C-only gcc builtins used in <math.h> isinf
919 [21932] network: Unpaired __resolv_context_get in generic get*_r
920 implementation
921 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
922 [21944] libc: sigval namespace
923 [21951] localedata: Update hanzi collation by stroke
924 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
925 sysdeps/x86_64/fpu/e_expf.S
926 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
927 allocation in syscall loops)
928 [21959] localedata: Fix Country name for xh_ZA
929 [21960] localedata: Fix abmon for bem_ZM
930 [21966] math: AVX2 mathvec functions use FMA without checking
931 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
932 [21971] localedata: Added New Locale for mfe_MU
933 [21972] libc: assert macro requires operator== (int) for its argument type
934 [21973] math: [sparc] libm missing sqrtl compat symbol
935 [21974] libc: Remove __bb_init_func and __bb_exit_func
936 [21982] string: stratcliff.c: error: assuming signed overflow does not
937 occur with -O3
938 [21986] stdio: __guess_grouping is called incorrectly
939 [21987] math: [sparc32] wrong bits/long-double.h installed
940 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
941 amounts)
942 [22022] localedata: Missing country_name for mni_IN
943 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
944 niu_NZ
945 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
946 checks
947 [22026] locale: iconv_open: heap overflow on gconv_init failure
948 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
949 [22035] math: [m68k] bits/math-inline.h macro namespace
950 [22038] localedata: Fix abbreviated weeks and months for Somali
951 [22044] localedata: Remove redundant data for Limburgish Language
952 [22050] malloc: Linking with -lmcheck does not hook
953 __malloc_initialize_hook correctly
954 [22051] libc: zero terminator in the middle of glibc's .eh_frame
955 [22052] malloc: malloc failed to compile with GCC 7 and -O3
956 [22070] localedata: charmaps/UTF-8: wcwidth for
957 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
958 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
959 Jungseong and Jongseong) should be 0
960 [22078] nss: nss_files performance issue in multi mode
961 [22082] math: bits/math-finite.h exp10 condition
962 [22086] libc: pcprofiledump incorrect cross-endian condition
963 [22093] dynamic-link: ld.so no longer searches in .../x86_64
964 [22095] network: Name server address allocation memory leak in resolv.conf
965 parsing after OOM
966 [22096] network: __resolv_conf_attach can incorrectly free passed conf
967 object
968 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
969 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
970 e.g. ET_GNU_DEBUG_*
971 [22111] malloc: malloc: per thread cache is not returned when thread exits
972 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
973 [22134] libc: [linux] implement fexecve with execveat
974 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
975 ppc64le
976 [22145] libc: ttyname() gives up too early in the face of namespaces
977 [22146] math: C++ build issue with float128 on x86_64
978 [22153] nptl: nptl: save error code before process termination
979 [22156] libc: [hppa,ia64,microblaze] Executable stack default
980 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
981 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
982 [22165] libc: [hppa] Text relocations in libc.so
983 [22180] libc: destructor registered via __cxa_atexit is called twice
984 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
985 symlinks") cause "make" segfaults
986 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
987 math_force_eval
988 [22207] libc: FAIL: stdlib/test-atexit-race
989 [22225] math: nearbyint arithmetic moved before feholdexcept
990 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
991 [22235] math: iscanonical in C++ and float128
992 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
993 the svid compat wrapper
994 [22244] math: ynf and yn are wrong without the svid compat wrapper
995 [22273] libc: Improper assert in Linux posix_spawn implementation
996 [22284] libc: -pg -pie doesn't work
997 [22292] locale: localedef exits with error 4 when it should be error 1
998 [22294] locale: Allow "" for int_currency_symbol definition in locales.
999 [22295] locale: Don't warn on non-symbolic characters in locale sources in
1000 --verbose.
1001 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
1002 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
1003 to 2.26
1004 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
1005 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
1006 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
1007 [22322] libc: [mips64] wrong bits/long-double.h installed
1008 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
1009 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
1010 (CVE-2017-15804)
1011 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
1012 [22343] malloc: Integer overflow in posix_memalign
1013 [22347] libc: getrandom() returns the number of bytes that were copied to
1014 the buffer even though the comments say "Return 0 on success and -1 on
1015 failure."
1016 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
1017 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
1018 -m32
1019 [22370] dynamic-link: Incorrect note padding check
1020 [22375] libc: malloc returns pointer from tcache_get when should return
1021 NULL (CVE-2017-17426)
1022 [22377] math: iseqsig, float128 and C++
1023 [22382] localedata: Error in tpi_PG locale
1024 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
1025 inside the ASCII printable range
1026 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
1027 -mlong-double-64
1028 [22403] localedata: Slash needs escaping in some locales
1029 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
1030 [22409] network: res_hnok does not accept some host names used on the
1031 Internet
1032 [22412] network: res_dnok, res_hnok should perform syntax checks
1033 [22413] network: ns_name_pton ignores syntactically invalid trailing
1034 backslash
1035 [22415] stdio: setvbuf can lead to invalid free/segfault
1036 [22432] build: Non-deterministic build
1037 [22439] malloc: malloc_info should compute summary statistics for all sub-
1038 heaps in an arena
1039 [22442] network: if_nametoindex could report index for the wrong
1040 networking interface
1041 [22446] build: aliasing violation calling readlink in handle_request
1042 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
1043 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
1044 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
1045 __nscd_hash/__nss_hash
1046 [22463] network: p_secstodate overflow handling
1047 [22469] localedata: pl_PL LC_COLLATE does not use i18n
1048 [22478] libc: sigwait can fail with EINTR
1049 [22505] libc: ldconfig processes include directive in locale-specific
1050 order
1051 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
1052 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
1053 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
1054 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
1055 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
1056 [22534] localedata: Collation rules for Serbian and Bosnian should be the
1057 same as for Croatian
1058 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
1059 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
1060 [22577] libc: missing newline after "cannot allocate TLS data structures
1061 for initial thread"
1062 [22588] manual: manual/conf.texi: missing underscore in front of
1063 SC_SSIZE_MAX
1064 [22593] math: nextafter and nexttoward are declared with const attribute
1065 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
1066 [22603] string: ia64 memchr overflows internal pointer check
1067 [22605] libc: SH clone does not set the exit code correctly
1068 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
1069 (CVE-2017-1000408)
1070 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
1071 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
1072 modified in case of success
1073 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
1074 [22615] manual: manual: ambiguous wording about errno value in case of
1075 success
1076 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
1077 stack-protector=all
1078 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
1079 binaries or if /proc is not mounted (CVE-2017-16997)
1080 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
1081 [22630] build: $(no-pie-ldflag) is no longer effective
1082 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
1083 [22635] nptl: pthread_self returns NULL before libpthread is loaded
1084 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
1085 [22637] nptl: guard size is subtracted from thread stack size instead of
1086 adding it on top
1087 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
1088 [22657] localedata: hu_HU: Avoid double space in date
1089 [22660] math: fmax, fmin sNaN handling on alpha
1090 [22664] libc: New warning of GCC8
1091 [22665] math: alpha: ceil and floor raise inexact exceptions
1092 [22666] math: alpha: trunc raise inexact exceptions
1093 [22667] libc: makecontext lacks stack alignment on i386
1094 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
1095 [22679] libc: getcwd(3) can succeed without returning an absolute path
1096 (CVE-2018-1000001)
1097 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
1098 elision and tunables
1099 [22687] math: [powerpc-nofpu] complex long double functions spurious
1100 "invalid" exception
1101 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
1102 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
1103 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
1104 exception
1105 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
1106 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
1107 power4
1108 [22701] nis: Incomplete removal of libnsl
1109 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
1110 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
1111 [22715] dynamic-link: FAIL: elf/tst-audit10
1112 [22719] libc: Backtrace tests fail on hppa
1113 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
1114 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
1115 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
1116 before the first call to crypt_r () in crypt/badsalttest.c
1117
1118 \f
1119 Version 2.26
1120
1121 Major new features:
1122
1123 * A per-thread cache has been added to malloc. Access to the cache requires
1124 no locks and therefore significantly accelerates the fast path to allocate
1125 and free small amounts of memory. Refilling an empty cache requires locking
1126 the underlying arena. Performance measurements show significant gains in a
1127 wide variety of user workloads. Workloads were captured using a special
1128 instrumented malloc and analyzed with a malloc simulator. Contributed by
1129 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
1130
1131 * Unicode 10.0.0 Support: Character encoding, character type info, and
1132 transliteration tables are all updated to Unicode 10.0.0, using
1133 generator scripts contributed by Mike FABIAN (Red Hat).
1134 These updates cause user visible changes, especially the changes in
1135 wcwidth for many emoji characters cause problems when emoji sequences
1136 are rendered with pango, see for example:
1137 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
1138
1139 * Collation of Hungarian has been overhauled and is now consistent with "The
1140 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
1141 Egmont Koblinger.
1142
1143 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
1144
1145 - The GNU C Library will now detect when /etc/resolv.conf has been
1146 modified and reload the changed configuration. The new resolver option
1147 “no-reload” (RES_NORELOAD) disables this behavior.
1148
1149 - The GNU C Library now supports an arbitrary number of search domains
1150 (configured using the “search” directive in /etc/resolv.conf);
1151 previously, there was a hard limit of six domains. For backward
1152 compatibility, applications that directly modify the ‘_res’ global
1153 object are still limited to six search domains.
1154
1155 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
1156 Library will now randomly pick a name server from the configuration as a
1157 starting point. (Previously, the second name server was always used.)
1158
1159 * The tunables feature is now enabled by default. This allows users to tweak
1160 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
1161
1162 * New function reallocarray, which resizes an allocated block (like realloc)
1163 to the product of two sizes, with a guaranteed clean failure upon integer
1164 overflow in the multiplication. Originally from OpenBSD, contributed by
1165 Dennis Wölfing and RĂŒdiger Sonderfeld.
1166
1167 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
1168 These are extended versions of preadv and pwritev, respectively, taking an
1169 additional flags argument. The set of supported flags depends on the
1170 running kernel; full support currently requires kernel 4.7 or later.
1171
1172 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
1173 create a new session ID for the spawned process. This feature is
1174 scheduled to be added to the next major revision of POSIX; for the time
1175 being, it is available under _GNU_SOURCE.
1176
1177 * errno.h is now safe to use from C-preprocessed assembly language on all
1178 supported operating systems. In this context, it will only define the
1179 Exxxx constants, as preprocessor macros expanding to integer literals.
1180
1181 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
1182 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
1183 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
1184 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
1185
1186 To compile programs that use this feature, the compiler must support
1187 128-bit floating point with the type name _Float128 (as defined by TS
1188 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
1189 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
1190 must be defined to make the new interfaces visible.
1191
1192 The new functions and macros correspond to those present for other
1193 floating-point types (except for a few obsolescent interfaces not
1194 supported for the new type), with F128 or f128 suffixes; for example,
1195 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
1196 printf or scanf formats for the new type; the strfromf128 and strtof128
1197 interfaces should be used instead.
1198
1199 Deprecated and removed features, and other changes affecting compatibility:
1200
1201 * The synchronization that pthread_spin_unlock performs has been changed to
1202 now be equivalent to a C11 atomic store with release memory order to the
1203 spin lock's memory location. Previously, several (but not all)
1204 architectures used stronger synchronization (e.g., containing what is
1205 often called a full barrier). This change can improve performance, but
1206 may affect odd fringe uses of spin locks that depend on the previous
1207 behavior (e.g., using spin locks as atomic variables to try to implement
1208 Dekker's mutual exclusion algorithm).
1209
1210 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
1211 removed.
1212
1213 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
1214 will only be built and installed when the GNU C Library is configured with
1215 --enable-obsolete-rpc. This allows alternative RPC implementations, such
1216 as TIRPC or rpcsvc-proto, to be used.
1217
1218 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
1219 libnss_compat, are deprecated, and will not be built or installed by
1220 default.
1221
1222 The NIS(+) support library, libnsl, is also deprecated. By default, a
1223 compatibility shared library will be built and installed, but not headers
1224 or development libraries. Only a few NIS-related programs require this
1225 library. (In particular, the GNU C Library has never required programs
1226 that use 'gethostbyname' to be linked with libnsl.)
1227
1228 Replacement implementations based on TIRPC, which additionally support
1229 IPv6, are available from <https://github.com/thkukuk/>. The configure
1230 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
1231 name service modules, to be built and installed.
1232
1233 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
1234 support is enabled, the configured recursive resolver must support EDNS.
1235 (Responding to EDNS-enabled queries with responses which are not
1236 EDNS-enabled is fine, but FORMERR responses are not.)
1237
1238 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
1239 servers have not supported this opcode for a long time.
1240
1241 * The _res_opcodes variable has been removed from libresolv. It had been
1242 exported by accident.
1243
1244 * <string.h> no longer includes inline versions of any string functions,
1245 as this kind of optimization is better done by the compiler. The macros
1246 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
1247
1248 * The nonstandard header <xlocale.h> has been removed. Most programs should
1249 use <locale.h> instead. If you have a specific need for the definition of
1250 locale_t with no other declarations, please contact
1251 libc-alpha@sourceware.org and explain.
1252
1253 * The obsolete header <sys/ultrasound.h> has been removed.
1254
1255 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
1256
1257 * The obsolete function cfree has been removed. Applications should use
1258 free instead.
1259
1260 * The stack_t type no longer has the name struct sigaltstack. This changes
1261 the C++ name mangling for interfaces involving this type.
1262
1263 * The ucontext_t type no longer has the name struct ucontext. This changes
1264 the C++ name mangling for interfaces involving this type.
1265
1266 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
1267 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
1268 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
1269 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
1270 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
1271 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
1272 fpregset_t type no longer has the name struct fpu. This changes the C++
1273 name mangling for interfaces involving those types.
1274
1275 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
1276 synced with the kernel:
1277
1278 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
1279 are not supported on this architecture and have been removed.
1280
1281 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
1282 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
1283 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
1284
1285 Programs that assume the GET/SETREGS ptrace requests are universally
1286 available will now fail to build, instead of malfunctioning at runtime.
1287
1288 Changes to build and runtime requirements:
1289
1290 * Linux kernel 3.2 or later is required at runtime, on all architectures
1291 supported by that kernel. (This is a change from version 2.25 only for
1292 x86-32 and x86-64.)
1293
1294 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
1295
1296 * On most architectures, GCC 4.9 or later is required to build the GNU C
1297 Library. On powerpc64le, GCC 6.2 or later is required.
1298
1299 Older GCC versions and non-GNU compilers are still supported when
1300 compiling programs that use the GNU C Library. (We do not know exactly
1301 how old, and some GNU extensions to C may be _de facto_ required. If you
1302 are interested in helping us make this statement less vague, please
1303 contact libc-alpha@sourceware.org.)
1304
1305 Security related changes:
1306
1307 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
1308 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
1309
1310 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
1311 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
1312
1313 * Avoid printing a backtrace from the __stack_chk_fail function since it is
1314 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
1315 (CVE-2010-3192).
1316
1317 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
1318 fixed (CVE-2017-12133).
1319
1320 The following bugs are resolved with this release:
1321
1322 [984] network: Respond to changed resolv.conf in gethostbyname
1323 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
1324 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
1325 fields
1326 [12189] libc: __stack_chk_fail should not attempt a backtrace
1327 (CVE-2010-3192)
1328 [14096] time: Race condition on timezone/tst-timezone.out
1329 [14172] localedata: az_IR: new locale
1330 [14995] build: glibc fails to build if gold is the default linker, even if
1331 ld.bfd is available
1332 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
1333 [16637] network: inet_pton function is accepting IPv6 with bad format
1334 [16640] string: string/strtok.c: undefined behaviour inconsistent between
1335 x86 and other generic code
1336 [16875] localedata: ko_KR: fix lang_name
1337 [17225] localedata: ar_SY: localized month names for May and June are
1338 incorrect
1339 [17297] localedata: da_DK: wrong date_fmt string
1340 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
1341 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
1342 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
1343 [19066] localedata: ar_SA abbreviated day and month names are in English
1344 [19569] network: resolv: Support an arbitrary number of search domains
1345 [19570] network: Implement random DNS server selection in the stub
1346 resolver
1347 [19838] locale: localedef fails on PA-RISC
1348 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
1349 order of 0D36 and 0D37
1350 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
1351 chillu characters
1352 [20098] libc: FAIL: debug/backtrace-tst on hppa
1353 [20257] network: sunrpc: clntudp_call does not enforce timeout when
1354 receiving data
1355 [20275] localedata: locale day/abday/mon/abmon should not have trailing
1356 whitespace
1357 [20313] localedata: Update locale data to Unicode 9.0
1358 [20424] manual: Document how to provide a malloc replacement
1359 [20496] localedata: agr_PE: new language locale AwajĂșn / Aguaruna (agr)
1360 for Peru
1361 [20686] locale: Add el_GR@euro to SUPPORTED.
1362 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
1363 failures consistently
1364 [21015] dynamic-link: Document and fix --enable-bind-now
1365 [21016] nptl: pthread_cond support is broken on hppa
1366 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
1367 frame-pointer on i386
1368 [21049] libc: segfault in longjmp_chk() due to clobbered processor
1369 register
1370 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
1371 [21088] libc: Build fails with --enable-static-nss
1372 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
1373 generic c code is used
1374 [21109] libc: Tunables broken on big-endian
1375 [21112] math: powf has large ulp errors with base close to 1 and exponent
1376 around 4000
1377 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
1378 (CVE-2017-12133)
1379 [21120] malloc: glibc malloc is incompatible with GCC 7
1380 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
1381 with -lieee
1382 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
1383 (-0.0) when linking with -lieee
1384 [21171] math: log10, log2 and lgamma return incorrect results
1385 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
1386 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
1387 [21207] localedata: ce_RU: update weekdays from CLDR
1388 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
1389 [21217] localedata: Update months from CLDR-31
1390 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
1391 [21243] libc: support_delete_temp_file should issue warning for failed
1392 remove()
1393 [21244] libc: support resolv_test_start() socket fd close should be
1394 checked for errors.
1395 [21253] libc: localedef randomly segfaults when using -fstack-check due to
1396 new posix_spawn implementation
1397 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
1398 leads to lower CPU frequency
1399 [21259] libc: [alpha] termios.h missing IXANY for POSIX
1400 [21261] libc: [sparc64] bits/setjmp.h namespace
1401 [21267] network: [mips] bits/socket.h IOC* namespace
1402 [21268] libc: [alpha] termios.h NL2, NL3 namespace
1403 [21270] libc: mmap64 silently truncates large offset values
1404 [21275] libc: posix_spawn always crashes on ia64 now
1405 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
1406 [21280] math: [powerpc] logbl for POWER7 return incorrect results
1407 [21289] libc: Incorrect declaration for 32-bit platforms with
1408 _FILE_OFFSET_BITS=64 causes build error
1409 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
1410 not support gethostbyname4_r
1411 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
1412 switching
1413 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
1414 arenas
1415 [21340] libc: Support POSIX_SPAWN_SETSID
1416 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
1417 [21359] network: ns_name_pack needs additional byte in destination buffer
1418 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
1419 against fragmentation attacks (CVE-2017-12132)
1420 [21369] network: resolv: Remove EDNS fallback
1421 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
1422 and _POSIX_C_SOURCE
1423 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
1424 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
1425 [21393] stdio: Missing dup3 error check in freopen, freopen64
1426 [21396] libc: Use AVX2 memcpy/memset on Skylake server
1427 [21399] localedata: Bad description for U00EC in
1428 localedata/charmaps/CP1254
1429 [21411] malloc: realloc documentation error
1430 [21426] network: sys/socket.h uio.h namespace
1431 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
1432 [21445] libc: signal.h bsd_signal namespace
1433 [21455] network: Network headers stdint.h namespace
1434 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
1435 retry value
1436 [21475] network: resolv: Overlong search path is truncated mid-label
1437 [21511] libc: sigstack namespace
1438 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
1439 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
1440 bits/syscall.h: No such file or directory
1441 [21517] libc: struct sigaltstack namespace
1442 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
1443 [21533] localedata: Update locale data to Unicode 10.0
1444 [21537] libc:
1445 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
1446 junk at end of line, first unrecognized character is `@'
1447 [21538] libc: SIG_HOLD missing for XPG4
1448 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
1449 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
1450 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
1451 [21543] libc: sigevent namespace
1452 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
1453 size for general purpose registers in mcontext_t structure
1454 [21550] libc: sigwait namespace
1455 [21552] libc: XPG4 bsd_signal namespace
1456 [21554] libc: sigpause namespace
1457 [21560] libc: sys/wait.h signal.h namespace
1458 [21561] libc: waitid namespace
1459 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
1460 '/usr/include/stdlib.h': Permission denied
1461 [21575] libc: sys/wait.h missing struct rusage definition
1462 [21584] libc: sigaltstack etc namespace
1463 [21597] libc: siginterrupt namespace
1464 [21607] math: hppa: FAIL: math/test-tgmath
1465 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
1466 issue on x86-64
1467 [21622] libc: [tile] missing SA_* for POSIX.1:2008
1468 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
1469 stack and heap (CVE-2017-1000366)
1470 [21625] libc: wait3 namespace
1471 [21654] nss: Incorrect pointer alignment in NSS group merge result
1472 construction
1473 [21657] network: Parse interface zone id for node-local multicast
1474 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
1475 between_2_3
1476 [21666] libc: .symver is used on common symbol
1477 [21668] network: resolv: res_init cross-thread broadcast introduces race
1478 conditions
1479 [21687] math: tgmath.h totalorder, totalordermag return type
1480 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
1481 Hindi Locale
1482 [21696] libc: Incorrect assumption of of __cpu_mask in
1483 posix/sched_cpucount.c
1484 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
1485 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
1486 locale
1487 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
1488 leading to relocation crash
1489 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
1490 `resolv_conf_matches (resp, conf)' failed.
1491 [21710] localedata: Added Samoan language locale for Samoa
1492 [21711] localedata: Pashto yesstr/nostr locale are missing
1493 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
1494 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
1495 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
1496 [21723] localedata: yesstr/nostr missing for Chinese language locale
1497 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
1498 locale
1499 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
1500 locale
1501 [21728] localedata: New Locale for Tongan language
1502 [21729] localedata: incorrect LC_NAME fields for hi_IN
1503 [21733] localedata: yesstr and nostr are missing for zh_HK
1504 [21734] localedata: Missing yesstr and nostr are for kw_GB
1505 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
1506 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
1507 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
1508 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1509 strings
1510 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1511 protector=all
1512 [21749] localedata: Wrong abbreviated day name (“abday”) for
1513 ar_JO/ar_LB/ar_SY
1514 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1515 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1516 [21759] localedata: missing yesstr and nostr for Tigrinya
1517 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1518 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1519 [21767] localedata: Missing Bislama locales
1520 [21768] localedata: Missing yesstr and nostr for aa_ET
1521 [21770] localedata: Missing Field in li_NL
1522 [21778] nptl: Robust mutex may deadlock
1523 [21779] libc: MicroBlaze segfaults when loading libpthread
1524 [21783] localedata: Fix int_select international_call_prefixes
1525 [21784] localedata: Inconsistency in country_isbn
1526 [21788] localedata: Missing Country Postal Abbreviations
1527 [21794] localedata: Added-country_isbn-for-Italy
1528 [21795] localedata: Add/Fix country_isbn for France
1529 [21796] localedata: Added country_isbn for Republic of Korea
1530 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1531 [21799] localedata: Added int_select international_call_prefixes
1532 [21801] localedata: Added int_select international_call_prefixes
1533 [21804] nptl: Double semicolon in thread-shared-types.h
1534 [21807] localedata: LC_ADDRESS fix for pap_CW
1535 [21808] localedata: Fix LC_ADDRESS for pap_AW
1536 [21821] localedata: Added country_name in mai_IN
1537 [21822] localedata: Fix LC_TIME for mai_IN
1538 [21823] localedata: missing yesstr, nostr for sa_IN
1539 [21825] localedata: Fix name_mrs for mag_IN
1540 [21828] localedata: 2.26 changelog should mention user visible changes
1541 with unicode 9.0
1542 [21835] localedata: Added Maithili language locale for Nepal
1543 [21838] localedata: Removed redundant data for the_NP
1544 [21839] localedata: Fix LC_MONETARY for ta_LK
1545 [21844] localedata: Fix Latin characters and Months Sequence.
1546 [21848] localedata: Fix mai_NP Title Name
1547
1548 \f
1549 Version 2.25
1550
1551 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1552 24731-2:2010, is supported to enable declarations of functions from that
1553 TR. Note that not all functions from that TR are supported by the GNU C
1554 Library.
1555
1556 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1557 18661-1:2014, is supported to enable declarations of functions and macros
1558 from that TS. Note that not all features from that TS are supported by
1559 the GNU C Library.
1560
1561 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1562 18661-4:2015, is supported to enable declarations of functions and macros
1563 from that TS. Note that most features from that TS are not supported by
1564 the GNU C Library.
1565
1566 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1567 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1568 Since the GNU C Library defaults to a much newer revision of POSIX, this
1569 will only affect programs that specifically request an old conformance
1570 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1571 see a change in the visible declarations, but a program compiled with
1572 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1573 will not.
1574
1575 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1576 defined by all multithreaded code, but glibc has not required this for
1577 many years.
1578
1579 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1580 means that in a future release, the macros “major”, “minor”, and “makedev”
1581 will only be available from <sys/sysmacros.h>.
1582
1583 These macros are not part of POSIX nor XSI, and their names frequently
1584 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1585 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1586 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1587 problem.
1588
1589 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1590 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1591 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1592
1593 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1594 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1595 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1596 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1597 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1598 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1599 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1600 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1601 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1602 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1603 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1604 WINT_WIDTH.
1605
1606 * New <math.h> features are added from TS 18661-1:2014:
1607
1608 - Signaling NaN macros: SNANF, SNAN, SNANL.
1609
1610 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1611 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1612 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1613
1614 - llogb functions: the llogb, llogbf and llogbl functions, and the
1615 FP_LLOGB0 and FP_LLOGBNAN macros.
1616
1617 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1618 fminmagf, fminmagl.
1619
1620 - Comparison macros: iseqsig.
1621
1622 - Classification macros: iscanonical, issubnormal, iszero.
1623
1624 - Total order functions: totalorder, totalorderf, totalorderl,
1625 totalordermag, totalordermagf, totalordermagl.
1626
1627 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1628
1629 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1630 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1631
1632 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1633 are added to libc. They convert a floating-point number into string.
1634
1635 * Most of glibc can now be built with the stack smashing protector enabled.
1636 It is recommended to build glibc with --enable-stack-protector=strong.
1637 Implemented by Nick Alcock (Oracle).
1638
1639 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1640 intended to be used instead of memset() to erase sensitive data after use;
1641 the compiler will not optimize out calls to explicit_bzero even if they
1642 are "unnecessary" (in the sense that no _correct_ program can observe the
1643 effects of the memory clear).
1644
1645 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1646 to float instead of double. This does not affect the ABI of any libraries
1647 that are part of the GNU C Library, but may affect the ABI of other
1648 libraries that use this type in their interfaces.
1649
1650 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1651 float_t and double_t types are now defined to long double instead of float
1652 and double. These options are not the default, and this does not affect
1653 the ABI of any libraries that are part of the GNU C Library, but it may
1654 affect the ABI of other libraries that use this type in their interfaces,
1655 if they are compiled or used with those options.
1656
1657 * The getentropy and getrandom functions, and the <sys/random.h> header file
1658 have been added.
1659
1660 * The buffer size for byte-oriented stdio streams is now limited to 8192
1661 bytes by default. Previously, on Linux, the default buffer size on most
1662 file systems was 4096 bytes (and thus remains unchanged), except on
1663 network file systems, where the buffer size was unpredictable and could be
1664 as large as several megabytes.
1665
1666 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1667 for the Linux quota interface which predates kernel version 2.4.22 has
1668 been removed.
1669
1670 * The malloc_get_state and malloc_set_state functions have been removed.
1671 Already-existing binaries that dynamically link to these functions will
1672 get a hidden implementation in which malloc_get_state is a stub. As far
1673 as we know, these functions are used only by GNU Emacs and this change
1674 will not adversely affect already-built Emacs executables. Any undumped
1675 Emacs executables, which normally exist only during an Emacs build, should
1676 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1677
1678 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1679 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1680 “no-ip6-dotint” had already been the default, and support for the
1681 “ip6-dotint” option was removed from the Internet in 2006.
1682
1683 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1684 flag from <resolv.h> have been removed. The option relied on a
1685 backwards-incompatible DNS extension which was never deployed on the
1686 Internet.
1687
1688 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1689 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1690 They were already unimplemented.
1691
1692 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1693 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1694 removed again from the IETF name lookup interface specification in RFC
1695 2553. Applications should use getaddrinfo instead.
1696
1697 * DNSSEC-related declarations and definitions have been removed from the
1698 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1699 decode the data part of DNSSEC record types. Previous versions of glibc
1700 only implemented minimal support for the previous version of DNSSEC, which
1701 is incompatible with the currently deployed version.
1702
1703 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1704 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1705 <arpa/nameser.h> header file because the distinction between RR types and
1706 meta-RR types is not officially standardized, subject to revision, and
1707 thus not suitable for encoding in a macro.
1708
1709 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1710 and rhook members of the res_state type in <resolv.h> have been removed.
1711 The glibc stub resolver did not support these hooks, but the header file
1712 did not reflect that.
1713
1714 * For multi-arch support it is recommended to use a GCC which has
1715 been built with support for GNU indirect functions. This ensures
1716 that correct debugging information is generated for functions
1717 selected by IFUNC resolvers. This support can either be enabled by
1718 configuring GCC with '--enable-gnu-indirect-function', or by
1719 enabling it by default by setting 'default_gnu_indirect_function'
1720 variable for a particular architecture in the GCC source file
1721 'gcc/config.gcc'.
1722
1723 * GDB pretty printers have been added for mutex and condition variable
1724 structures in POSIX Threads. When installed and loaded in gdb these pretty
1725 printers show various pthread variables in human-readable form when read
1726 using the 'print' or 'display' commands in gdb.
1727
1728 * Tunables feature added to allow tweaking of the runtime for an application
1729 program. This feature can be enabled with the '--enable-tunables' configure
1730 flag. The GNU C Library manual has details on usage and README.tunables has
1731 instructions on adding new tunables to the library.
1732
1733 * A new version of condition variables functions have been implemented in
1734 the NPTL implementation of POSIX Threads to provide stronger ordering
1735 guarantees.
1736
1737 * A new version of pthread_rwlock functions have been implemented to use a more
1738 scalable algorithm primarily through not using a critical section anymore to
1739 make state changes.
1740
1741 Security related changes:
1742
1743 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1744 have been created with makecontext could fail to terminate due to a
1745 missing .cantunwind annotation. This has been observed to lead to a hang
1746 (denial of service) in some Go applications compiled with gccgo. Reported
1747 by Andreas Schwab. (CVE-2016-6323)
1748
1749 * The DNS stub resolver functions would crash due to a NULL pointer
1750 dereference when processing a query with a valid DNS question type which
1751 was used internally in the implementation. The stub resolver now uses a
1752 question type which is outside the range of valid question type values.
1753 (CVE-2015-5180)
1754
1755 The following bugs are resolved with this release:
1756
1757 [4099] stdio: Overly agressive caching by stream i/o functions.
1758 [7065] build: Support building glibc with -fstack-protector or -fstack-
1759 protector-all
1760 [9842] localedata: en_CA: incorrect date format
1761 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1762 before it started waiting
1763 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1764 [15765] nptl: sem_open is wrongly a cancellation point
1765 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1766 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1767 wrong type
1768 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1769 library linked with pthread
1770 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1771 "FIXME: Ingo" issue)
1772 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1773 [17252] libc: getrandom and getentropy syscall
1774 [17730] malloc: thread-local storage is sometimes improperly free()'d
1775 after being __libc_memalign()'d
1776 [18241] stdio: failed fseek on memstream does not set errno and can fail
1777 when it shouldnt
1778 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1779 not
1780 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1781 PTHREAD_PRIO_INHERIT mutex on ARM
1782 [18784] network: res_query and related function crash for special record
1783 type queries (CVE-2015-5180)
1784 [19380] math: strtod does not raise "inexact"
1785 [19387] string: Integer overflow in memchr
1786 [19390] string: Integer overflow in strncat
1787 [19398] build: linknamespace tests fail with massively parallel build
1788 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1789 termination
1790 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1791 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1792 [19514] libc: [PATCH] Fix spelling errors in spelling
1793 "implement"/"implementation" in several places
1794 [19582] network: Deprecate RES_USE_INET6
1795 [19673] manual: clog10 docs appear to be erroneous
1796 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1797 causes a segmentation fault
1798 [19826] libc: invalid pointer returned from __tls_get_addr with static
1799 linking
1800 [20016] network: resolv: Remove hooks support from the API
1801 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1802 uninitialized GOT
1803 [20033] math: [x86_64] vectorized math function don't call the __finite
1804 versions
1805 [20116] nptl: use after free in pthread_create
1806 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1807 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1808 is always true.
1809 [20311] nptl: please install proc_service.h
1810 [20366] libc: Compilation errors in installed headers in strict-compliance
1811 modes
1812 [20370] malloc: malloc: Arena free list management is still racy
1813 (incorrect fix in bug 19243)
1814 [20386] libc: assert (X = 0) does not result in GCC warning
1815 [20432] malloc: malloc: Minimize interface required for interposition
1816 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1817 in _Unwind_Backtrace (CVE-2016-6323)
1818 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1819 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1820 libpthread.a
1821 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1822 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1823 [20477] network: resolv: incorrect double-checked locking related to
1824 _res_hconf
1825 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1826 information.
1827 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1828 penalty
1829 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1830 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1831 cause transition penalty
1832 [20517] math: sparcv9 missing fdiml compat symbol
1833 [20524] manual: strverscmp is inconsistent
1834 [20525] libc: <sys/quota.h> should be based on kernel headers
1835 [20539] math: GCC 7 -static -lm fails to link at -O3
1836 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1837 mcount.oS)
1838 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1839 Checking
1840 [20591] network: Remove obsolete DNSSEC support
1841 [20592] network: DNS resource record type classification macros in
1842 <arpa/nameser.h> are incorrect
1843 [20593] network: Update DNS RR type definitions
1844 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1845 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1846 multi-arch
1847 [20629] network: libresolv: Remove support for bitstring labels
1848 (RES_USEBSTRING)
1849 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1850 assertion: "maxidx >= 2"
1851 [20660] math: [arm] Use VSQRT
1852 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1853 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1854 [20689] libc: Test for FMA should also check for AVX.
1855 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1856 [20715] math: iszero macro breaks existing code
1857 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1858 [20728] libc: powerpc: Missing TOC stub in clone
1859 [20729] build: glibc-2.24 fails to build for i486 with -Os
1860 [20750] build: Build fails with default PIE enabled gcc-6.x
1861 [20768] math: [alpha] sqrt fegetenv namespace
1862 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1863 glibc
1864 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1865 __FLT_EVAL_METHOD__ is 2
1866 [20790] network: rpcgen buffer overrun in get_prog_declaration
1867 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1868 [20829] libc: crypt snprintf namespace
1869 [20847] libc: tst-vfork3 failure
1870 [20855] math: Default bits/mathdef.h has inappropriate float_t
1871 [20859] math: [sh4] FP_ILOGB0 invalid
1872 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1873 U+20AC), not same as GBK
1874 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1875 [20916] math: pow handling of sNaN arguments
1876 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1877 [20919] math: Bad pow (qNaN, 0) result with -lieee
1878 [20940] math: hypot sNaN handling
1879 [20947] math: fmax, fmin sNaN handling
1880 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1881 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1882 RES_USE_INET6
1883 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1884 [20973] nptl: robust mutexes: Lost wake-ups
1885 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1886 [20978] nis: strlen on null pointer in nss_nisplus
1887 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1888 wrong condition
1889 [21014] string: i686 memchr overflows internal pointer check
1890 [21019] libc: [mips] n32 lseek incorrect on overflow
1891 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1892 [21026] libc: [MIPS] readahead syscall is broken on n64
1893 [21028] math: Fallback fesetexceptflag should always succeed
1894 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1895 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1896 clang
1897 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1898 [21061] librt: [microblaze] librt lost clock_* exports
1899 [21073] libc: tunables: insecure environment variables passed to
1900 subprocesses with AT_SECURE
1901 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1902 \f
1903 Version 2.24
1904
1905 * The minimum Linux kernel version that this version of the GNU C Library
1906 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1907 kernel version 2.6.32 or later suffices (on architectures that already
1908 required kernel versions more recent than 3.2, those requirements remain
1909 unchanged). Linux 3.2 or later kernel headers are required on all
1910 architectures.
1911
1912 * The pap_AN locale has been deleted. This has been deprecated for a long
1913 time. It has been replaced by pap_AW & pap_CW, both of which have long
1914 been included in previous releases.
1915
1916 * The readdir_r and readdir64_r functions have been deprecated. It is
1917 recommended to use readdir and readdir64 instead.
1918
1919 * The type “union wait” has been removed. It was deprecated in the early
1920 1990s and never part of POSIX. Application code should use the int type
1921 instead of “union wait”.
1922
1923 * A new NSS action is added to facilitate large distributed system
1924 administration. The action, MERGE, allows remote user stores like LDAP
1925 to be merged into local user stores like /etc/groups in order to provide
1926 easy to use, updated, and managed sets of merged credentials. The new
1927 action can be used by configuring it in /etc/nsswitch.conf:
1928 group: files [SUCCESS=merge] nis
1929 Implemented by Stephen Gallagher (Red Hat).
1930
1931 * The deprecated __malloc_initialize_hook variable has been removed from the
1932 API.
1933
1934 * The long unused localedef --old-style option has been removed. It hasn't
1935 done anything in over 16 years. Scripts using this option can safely
1936 drop it.
1937
1938 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1939 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1940 return the next representable value in the direction of positive infinity
1941 and the nextdown functions return the next representable value in the
1942 direction of negative infinity. These are currently enabled as GNU
1943 extensions.
1944
1945 Security related changes:
1946
1947 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1948 could result in a stack overflow when getnetbyname was called with an
1949 overly long name. (CVE-2016-3075)
1950
1951 * Previously, getaddrinfo copied large amounts of address data to the stack,
1952 even after the fix for CVE-2013-4458 has been applied, potentially
1953 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1954 instead. Reported by Michael Petlan. (CVE-2016-3706)
1955
1956 * The glob function suffered from a stack-based buffer overflow when it was
1957 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1958 Reported by Alexander Cherepanov. (CVE-2016-1234)
1959
1960 * The Sun RPC UDP client could exhaust all available stack space when
1961 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1962 alloca plugin for GCC. (CVE-2016-4429)
1963
1964 * The IPv6 name server management code in libresolv could result in a memory
1965 leak for each thread which is created, performs a failing naming lookup,
1966 and exits. Over time, this could result in a denial of service due to
1967 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1968
1969 The following bugs are resolved with this release:
1970
1971 [1170] localedata: ne_NP: update Nepali locale definition file
1972 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1973 of MS-DOS.
1974 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1975 [6796] math: fdim() does not set errno on overflow
1976 [10354] libc: posix_spawn should use vfork() in more cases than presently
1977 [11213] localedata: localedata: add copyright disclaimer to locale files
1978 [12143] localedata: chr_US: new Cherokee locale
1979 [12450] localedata: sgs_LT: new locale
1980 [12676] localedata: ln_CD: new locale
1981 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1982 CLDR data
1983 [13304] math: fma, fmaf, fmal produce wrong results
1984 [14259] build: --localedir arg to configure is ignored
1985 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1986 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1987 [14934] localedata: es_CL: wrong first weekday chilean locale
1988 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1989 romanisation
1990 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1991 and +/-
1992 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1993 [15368] nptl: raise() is not async-signal-safe
1994 [15479] math: ceil, floor, round and trunc raise inexact exception
1995 [15578] localedata: kk_KZ: various updates
1996 [16003] localedata: pap_AN: punt old locale
1997 [16137] localedata: iw_IL: punt old locale
1998 [16190] localedata: eo: new esperanto locale
1999 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
2000 [16742] malloc: race condition: pthread_atfork() called before first
2001 malloc() results in unexpected locking behaviour/deadlocks
2002 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
2003 all locales
2004 [16983] localedata: postal_fmt does not allow %l and %n modifiers
2005 [17565] localedata: pt_PT: wrong (work-)week start
2006 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
2007 [17950] build: Build fails with -msse
2008 [18205] localedata: be_BY*: wrong first_weekday and first_workday
2009 [18433] libc: posix_spawn does not return correctly upon failure to
2010 execute
2011 [18453] localedata: charmaps/IBM875: incorrect codes
2012 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
2013 -Wsystem-headers
2014 [18896] localedata: he_IL: improvements for currency
2015 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
2016 Romanian locale data
2017 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
2018 symbol
2019 [19056] libc: Deprecate readdir_r
2020 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
2021 language
2022 [19198] localedata: nl_NL: small improvements for Dutch locales
2023 [19257] network: Per-thread memory leak in __res_vinit with IPv6
2024 nameservers (CVE-2016-5417)
2025 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
2026 machine
2027 [19400] locale: Language missing in "iso-639.def", trivial fix in
2028 description
2029 [19431] malloc: Deadlock between fflush, getdelim, and fork
2030 [19505] libc: Incorrect file descriptor validity checks in
2031 posix_spawn_file_actions_add{open,close,dup2}
2032 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
2033 when using RTLD_NEXT
2034 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
2035 `intl/{gettextP,loadinfo}.h'
2036 [19534] libc: execle, execlp may use malloc
2037 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
2038 [19573] network: res_nclose and __res_maybe_init disagree about name
2039 server initialization, breaking Hesiod
2040 [19575] localedata: Status of GB18030 tables
2041 [19581] localedata: sr_* date_fmt string contains additional newline
2042 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
2043 Excavator core
2044 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
2045 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
2046 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
2047 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
2048 double range
2049 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
2050 part zero incorrect
2051 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
2052 equality tests
2053 [19610] dynamic-link: ldconfig -X removes stale symbolic links
2054 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
2055 [19633] locale: strfmon_l applies global locale to number formatting
2056 [19642] network: Memory leak in getnameinfo
2057 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
2058 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
2059 glibc-2.22
2060 [19654] math: [x86_64] Need testcase for BZ #19590 fix
2061 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
2062 'tst-numeric.c'
2063 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
2064 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
2065 low part
2066 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
2067 result
2068 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
2069 unaligned stack
2070 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
2071 pointers and lengths in error-case.
2072 [19727] locale: Converting from/to UTF-xx with iconv() does not always
2073 report errors on UTF-16 surrogates values.
2074 [19755] nscd: nscd assertion failure in gc
2075 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
2076 [19759] libc: mempcpy shouldn't be inlined
2077 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
2078 [19765] libc: s390 needs an optimized mempcpy
2079 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
2080 NAME_MAX limit assumption (CVE-2016-1234)
2081 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
2082 [19787] network: Missing and incorrect truncation checks in getnameinfo
2083 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
2084 modes
2085 [19791] network: Assertion failure in res_query.c with un-connectable name
2086 server addresses
2087 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
2088 [19822] math: libm.so install clobbers old version
2089 [19825] network: resolv: send_vc can return uninitialized data in second
2090 response to getaddrinfo
2091 [19830] network: nss_dns: should check RDATA length against buffer length
2092 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
2093 confronted with A/AAAA records of invalid size
2094 [19837] nss: nss_db: No retries for some long lines with a larger buffer
2095 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
2096 [19853] stdio: Printing IBM long double in decimal with high precision is
2097 sometimes incorrect
2098 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
2099 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
2100 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
2101 record types
2102 [19865] network: Assertion failure or memory leak in
2103 _nss_dns_getcanonname_r
2104 [19868] network: nss_dns: netent code does not skip over non-PTR records
2105 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
2106 (CVE-2016-3075)
2107 [19881] string: Improve x86-64 memset
2108 [19907] string: Incorrect memcpy tests
2109 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
2110 resolving symbols
2111 [19925] libc: termios.h XCASE namespace
2112 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
2113 [19929] libc: limits.h NL_NMAX namespace
2114 [19931] stdio: Memory leak in vfprintf
2115 [19957] libc: clone(CLONE_VM) access invalid parent memory
2116 [19963] localedata: en_IL: New locale
2117 [19989] stdio: stdio.h cuserid namespace
2118 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
2119 gethosts
2120 [19996] locale: langinfo.h nl_langinfo_l namespace
2121 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
2122 [20010] network: getaddrinfo: Stack overflow in hostent translation
2123 (CVE-2016-3706)
2124 [20012] stdio: libio: fmemopen append mode failure
2125 [20014] stdio: stdio.h namespace for pre-threads POSIX
2126 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
2127 [20023] libc: fcntl.h timespec namespace
2128 [20024] math: [x86_64] vectorized sincos trashes the stack
2129 [20031] network: nss_hesiod: Heap overflow in get_txt_records
2130 [20041] time: sys/time.h timespec namespace
2131 [20043] libc: unistd.h missing cuserid for UNIX98 and before
2132 [20044] libc: unistd.h missing pthread_atfork for UNIX98
2133 [20051] libc: ttyslot in wrong header under wrong conditions
2134 [20054] libc: gethostname not declared for XPG4
2135 [20055] libc: termios.h missing tcgetsid for XPG4
2136 [20072] dynamic-link: x86 init_cpu_features is called twice in static
2137 executable
2138 [20073] libc: sys/stat.h fchmod namespace
2139 [20074] libc: stdlib.h rand_r namespace
2140 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
2141 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
2142 XPG3
2143 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
2144 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
2145 (CVE-2016-4429)
2146 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
2147 [20119] libc: Wrong mask for processors level type from CPUID
2148 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
2149 AS not supporting AVX512
2150 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
2151 sNaN argument
2152 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
2153 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
2154 argument
2155 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
2156 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
2157 [20178] libc: posix_spawn{p} should not call exit
2158 [20191] stdio: libio: vtables hardening
2159 [20195] string: FMA4 detection requires CPUID execution with register
2160 eax=0x80000001
2161 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
2162 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
2163 subnormals
2164 [20212] math: acos (sNaN) returns sNaN
2165 [20213] math: asin (sNaN) returns sNaN
2166 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
2167 [20218] math: [i386] asinhl (sNaN) returns sNaN
2168 [20219] math: [i386] atanhl (sNaN) returns sNaN
2169 [20222] stdio: fopencookie: Mangle function pointers
2170 [20224] math: [i386] cbrtl (sNaN) returns sNaN
2171 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
2172 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
2173 input
2174 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
2175 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
2176 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
2177 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
2178 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
2179 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
2180 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
2181 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
2182 [20240] math: modf (sNaN) returns sNaN
2183 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
2184 [20250] math: frexp (sNaN) returns sNaN
2185 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
2186 [20255] math: [i386] fdim, fdimf return with excess range and precision /
2187 double rounding
2188 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
2189 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
2190 subscript is below array bounds [-Werror=array-bounds]
2191 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
2192 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
2193 (Only arm/linux)
2194 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
2195 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
2196 fallbacks
2197 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
2198 "invalid" exceptions
2199 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
2200 [20316] localedata: id_ID: Februari instead of Pebruari
2201 [20327] string: POWER8 strcasecmp returns incorrect result
2202 [20347] math: Failure: Test: j0_downward (0xap+0)
2203 [20348] libc: FAIL: misc/tst-preadvwritev64
2204 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
2205 [20350] libc: There is no test for p{read,write}64
2206 [20357] math: Incorrect cos result for 1.5174239687223976
2207 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
2208 \f
2209 Version 2.23
2210
2211 * Unicode 8.0.0 Support: Character encoding, character type info, and
2212 transliteration tables are all updated to Unicode 8.0.0, using new
2213 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
2214 These updates cause user visible changes, such as the fixes for bugs
2215 89, 16061, and 18568.
2216
2217 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
2218 kernel-internal CPU set size. This means that requests that change the
2219 CPU affinity which failed before (for example, an all-ones CPU mask) will
2220 now succeed. Applications that need to determine the effective CPU
2221 affinities need to call sched_getaffinity or pthread_getaffinity_np after
2222 setting it because the kernel can adjust it (and the previous size check
2223 would not detect this in the majority of cases).
2224
2225 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
2226 the following new symbols are used: fts64_children, fts64_close,
2227 fts64_open, fts64_read and fts64_set.
2228
2229 * getaddrinfo now detects certain invalid responses on an internal netlink
2230 socket. If such responses are received, an affected process will
2231 terminate with an error message of "Unexpected error <number> on netlink
2232 descriptor <number>" or "Unexpected netlink response of size <number> on
2233 descriptor <number>". The most likely cause for these errors is a
2234 multi-threaded application which erroneously closes and reuses the netlink
2235 file descriptor while it is used by getaddrinfo.
2236
2237 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
2238 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
2239 unnecessary serialization of memory allocation requests across threads.
2240 The defect is now corrected. Users should see a substantial increase in
2241 the concurent throughput of allocation requests for applications which
2242 trigger this bug. Affected applications typically create and
2243 destroy threads frequently. (Bug 19048 was reported and analyzed by
2244 Ericsson.)
2245
2246 * There is now a --disable-timezone-tools configure option for disabling the
2247 building and installing of the timezone related utilities (zic, zdump, and
2248 tzselect). This is useful for people who build the timezone data and code
2249 independent of the GNU C Library.
2250
2251 * The obsolete header <regexp.h> has been removed. Programs that require
2252 this header must be updated to use <regex.h> instead.
2253
2254 * The obsolete functions bdflush, create_module, get_kernel_syms,
2255 query_module and uselib are no longer available to newly linked binaries;
2256 the header <sys/kdaemon.h> has been removed. These functions and header
2257 were specific to systems using the Linux kernel and could not usefully be
2258 used with the GNU C Library on systems with version 2.6 or later of the
2259 Linux kernel.
2260
2261 * Optimized string, wcsmbs and memory functions for IBM z13.
2262 Implemented by Stefan Liebler.
2263
2264 * Newly linked programs that define a variable called signgam will no longer
2265 have it set by the lgamma, lgammaf and lgammal functions. Programs that
2266 require signgam to be set by those functions must ensure that they use the
2267 variable provided by the GNU C Library and declared in <math.h>, without
2268 defining their own copy.
2269
2270 * The minimum GCC version that can be used to build this version of the GNU
2271 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
2272 still be used to compile programs using the GNU C Library.
2273
2274 Security related changes:
2275
2276 * An out-of-bounds value in a broken-out struct tm argument to strftime no
2277 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
2278
2279 * The LD_POINTER_GUARD environment variable can no longer be used to disable
2280 the pointer guard feature. It is always enabled. Previously,
2281 LD_POINTER_GUARD could be used to disable security hardening in binaries
2282 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
2283 (CVE-2015-8777)
2284
2285 * An integer overflow in hcreate and hcreate_r could lead to an
2286 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
2287
2288 * The catopen function no longer has unbounded stack usage. Reported by
2289 Max. (CVE-2015-8779)
2290
2291 * The nan, nanf and nanl functions no longer have unbounded stack usage
2292 depending on the length of the string passed as an argument to the
2293 functions. Reported by Joseph Myers. (CVE-2014-9761)
2294
2295 * A stack-based buffer overflow was found in libresolv when invoked from
2296 libnss_dns, allowing specially crafted DNS responses to seize control
2297 of execution flow in the DNS client. The buffer overflow occurs in
2298 the functions send_dg (send datagram) and send_vc (send TCP) for the
2299 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
2300 family. The use of AF_UNSPEC triggers the low-level resolver code to
2301 send out two parallel queries for A and AAAA. A mismanagement of the
2302 buffers used for those queries could result in the response of a query
2303 writing beyond the alloca allocated buffer created by
2304 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
2305 the overflow. Thanks to the Google Security Team and Red Hat for
2306 reporting the security impact of this issue, and Robert Holiday of
2307 Ciena for reporting the related bug 18665. (CVE-2015-7547)
2308
2309 The following bugs are resolved with this release:
2310
2311 [89] localedata: Locales nb_NO and nn_NO should transliterate ÊÞÄ
2312 [887] math: Math library function "logb" and "nextafter" inconsistent
2313 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
2314 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
2315 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
2316 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
2317 use `mkstemp'
2318 [4404] localedata: German translation of "Alarm clock" is misleading
2319 [6799] math: nextafter() and nexttoward() doen't set errno on
2320 overflow/underflow errors
2321 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
2322 overflow/underflow
2323 [10432] nis: _nss_nis_setnetgrent assertion failure
2324 [11460] libc: fts has no LFS support
2325 [12926] network: getaddrinfo()/make_request() may spin forever
2326 [13065] nptl: Race condition in pthread barriers
2327 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
2328 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
2329 are not contiguous
2330 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
2331 [14912] libc: Rename non-installed bits/*.h headers
2332 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
2333 [15367] math: Let gcc use __builtin_isinf
2334 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
2335 [15421] math: lgamma wrongly sets signgam for ISO C
2336 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
2337 FE_INVALID with argument out of range
2338 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
2339 all exceptions
2340 [15786] dynamic-link: ifunc resolver functions can smash function
2341 arguments
2342 [15918] math: Unnecessary check for equality in hypotf()
2343 [16061] localedata: Review / update transliteration data
2344 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
2345 should include
2346 [16141] time: strptime %z offset restriction
2347 [16171] math: drem should be alias of remainder
2348 [16296] math: fegetround is pure?
2349 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
2350 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
2351 GNU/Linux
2352 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
2353 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
2354 arguments
2355 [16422] math: [powerpc] math-float, math-double failing llrint tests with
2356 "Exception "Inexact" set" on ppc32
2357 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
2358 [16517] math: Missing underflow exception from tanf/tan/tanl
2359 [16519] math: Missing underflow exception from sinhf
2360 [16520] math: Missing underflow exception from tanhf
2361 [16521] math: Missing underflow exception from exp2
2362 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
2363 rounding results
2364 [16734] stdio: fopen calls mmap to allocate its buffer
2365 [16961] math: nan function incorrect handling of bad sequences
2366 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
2367 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
2368 [16985] locale: localedef: confusing error message when opening output
2369 fails
2370 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
2371 [17197] locale: Redundant shift character in iconv conversion output at
2372 block boundary
2373 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
2374 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
2375 [17250] dynamic-link: static linking breaks nss loading
2376 (getaddrinfo/getpwnam/etc...)
2377 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
2378 4.7?
2379 [17441] math: isnan() should use __builtin_isnan() in GCC
2380 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
2381 (related to lock elision)
2382 [17787] manual: Exponent on page 324 of the PDF ends prematurely
2383 [17886] time: strptime should be able to parse "Z" as a timezone with %z
2384 [17887] time: strptime should be able to parse "+01:00" style timezones
2385 [17905] libc: catopen() Multiple unbounded stack allocations
2386 (CVE-2015-8779)
2387 [18084] libc: backtrace (..., 0) dumps core on x86
2388 [18086] libc: nice() sets errno to 0 on success
2389 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
2390 is too large (CVE-2015-8778)
2391 [18251] dynamic-link: SONAME missing when audit modules provides path
2392 [18265] libc: add attributes for wchar string and memory functions
2393 [18370] math: csqrt missing underflows
2394 [18421] libc: [hppa] read-only segment has dynamic relocations
2395 [18472] libc: Obsolete syscall wrappers should be compat symbols
2396 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
2397 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
2398 [18525] localedata: Remove locale timezone information
2399 [18560] libc: [powerpc] spurious bits/ipc.h definitions
2400 [18568] localedata: Update locale data to Unicode 8.0
2401 [18589] locale: sort-test.sh fails at random
2402 [18595] math: ctan, ctanh missing underflows
2403 [18604] libc: assert macro-expands its argument
2404 [18610] math: S390: fetestexcept() reports any exception if DXC-code
2405 contains a vector instruction exception.
2406 [18611] math: j1, jn missing errno setting on underflow
2407 [18618] localedata: sync Chechen locale definitions with other *_RU
2408 locales
2409 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
2410 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
2411 [18665] network: In send_dg, the recvfrom function is NOT always using the
2412 buffer size of a newly created buffer (CVE-2015-7547)
2413 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
2414 missing break ?
2415 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
2416 32bit processes
2417 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
2418 [18699] math: tilegx cproj() for various complex infinities does not yield
2419 infinity
2420 [18724] libc: Harden put*ent functions against data injection
2421 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
2422 [18755] build: build errors with -DNDEBUG
2423 [18757] stdio: fmemopen fails to set errno on failure
2424 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
2425 be forced unloaded
2426 [18781] libc: openat64 lacks O_LARGEFILE
2427 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
2428 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
2429 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
2430 [18790] math: [ldbl-128ibm] tanhl inaccurate
2431 [18795] libc: stpncpy fortification misses buffer lengths that are
2432 statically too large
2433 [18796] build: build fails for --disable-mathvec
2434 [18803] math: hypot missing underflows
2435 [18820] stdio: fmemopen may leak memory on failure
2436 [18823] math: csqrt spurious underflows
2437 [18824] math: fma spurious underflows
2438 [18825] math: pow missing underflows
2439 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
2440 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
2441 [18870] build: sem_open.c fails to compile with missing symbol
2442 FUTEX_SHARED
2443 [18872] stdio: Fix memory leak in printf_positional
2444 [18873] libc: posix_fallocate overflow check ineffective
2445 [18875] math: Excess precision leads incorrect libm
2446 [18877] libc: arm: mmap offset regression
2447 [18887] libc: memory corruption when using getmntent on blank lines
2448 [18918] localedata: hu_HU: change time to HH:MM:SS format
2449 [18921] libc: Regression: extraneous stat() and fstat() performed by
2450 opendir()
2451 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
2452 binaries (CVE-2015-8777)
2453 [18951] math: tgamma missing underflows
2454 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
2455 signgam
2456 [18953] localedata: lt_LT: change currency symbol to the euro
2457 [18956] math: powf inaccuracy
2458 [18961] math: [i386] exp missing underflows
2459 [18966] math: [i386] exp10 missing underflows
2460 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
2461 [18969] build: multiple string test failures due to missing locale
2462 dependencies
2463 [18970] libc: Reference of pthread_setcancelstate in libc.a
2464 [18977] math: float / long double Bessel functions not in XSI POSIX
2465 [18980] math: i386 libm functions return with excess range and precision
2466 [18981] math: i386 scalb*, ldexp return with excess range and precision
2467 [18982] stdio: va_list and vprintf
2468 [18985] time: Passing out of range data to strftime() causes a segfault
2469 (CVE-2015-8776)
2470 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
2471 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
2472 [19012] locale: iconv_open leaks memory on error path
2473 [19016] math: clog, clog10 inaccuracy
2474 [19018] nptl: Mangle function pointers in tls_dtor_list
2475 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
2476 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
2477 [19048] malloc: malloc: arena free list can become cyclic, increasing
2478 contention
2479 [19049] math: [powerpc] erfc incorrect zero sign
2480 [19050] math: [powerpc] log* incorrect zero sign
2481 [19058] math: [x86_64] Link fail with -fopenmp and -flto
2482 [19059] math: nexttoward overflow incorrect in non-default rounding modes
2483 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
2484 [19074] network: Data race in _res_hconf_reorder_addrs
2485 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
2486 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
2487 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
2488 rounding modes
2489 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
2490 ILP32
2491 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
2492 [19086] manual: posix_fallocate64 documented argument order is wrong.
2493 [19088] math: lround, llround missing exceptions close to overflow
2494 threshold
2495 [19094] math: lrint, llrint missing exceptions close to overflow threshold
2496 [19095] math: dbl-64 lrint incorrect for 64-bit long
2497 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
2498 [19124] dynamic-link: ld.so failed to build with older assmebler
2499 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
2500 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
2501 [19134] math: [powerpc32] lround, lroundf spurious exceptions
2502 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
2503 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
2504 pthread_setaffinity_np
2505 [19156] math: [ldbl-128] j0l spurious underflows
2506 [19164] nptl: tst-getcpu fails with many possible CPUs
2507 [19168] math: math/test-ildoubl and math/test-ldouble failure
2508 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2509 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2510 prelink
2511 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2512 (FE_NOMASK_ENV) do not clear SSE exceptions
2513 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2514 _int_new_arena/reused_arena
2515 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2516 [19201] math: dbl-64 remainder incorrect sign of zero result
2517 [19205] math: bits/math-finite.h conditions do not match math.h and
2518 bits/mathcalls.h
2519 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2520 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2521 for C99-based standards
2522 [19212] libc: features.h not -Wundef clean
2523 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2524 math-only
2525 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2526 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2527 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2528 disabled
2529 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2530 "inexact" exceptions
2531 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2532 arguments
2533 [19242] libc: strtol incorrect in Turkish locales
2534 [19243] malloc: reused_arena can pick an arena on the free list, leading
2535 to an assertion failure and reference count corruption
2536 [19253] time: tzset() ineffective when temporary TZ did not include DST
2537 rules
2538 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2539 [19270] math: [hppa] Shared libm missing __isnanl
2540 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2541 MAP_HUGETLB and MAP_STACK defines
2542 [19313] nptl: Wrong __cpu_mask for x32
2543 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2544 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2545 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2546 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2547 [19363] time: x32: times() return value wrongly truncates/sign extends
2548 from 32bit
2549 [19367] dynamic-link: Improve branch prediction on Silvermont
2550 [19369] network: Default domain name not reset by res_ninit when "search"
2551 / "domain" entry is removed from resolv.conf
2552 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2553 subnormals
2554 [19385] localedata: bg_BG: time separator should be colon, not comma
2555 [19408] libc: linux personality syscall wrapper may erroneously return an
2556 error on 32-bit architectures
2557 [19415] libc: dladdr returns wrong names on hppa
2558 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2559 IBM905, IBM907, and IBM909
2560 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2561 [19443] build: build failures with -DDEBUG
2562 [19451] build: Make check fails on test-double-vlen2
2563 [19462] libc: Glibc failed to build with -Os
2564 [19465] math: Wrong code with -Os
2565 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2566 -Os
2567 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2568 CPU's.
2569 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2570 [19486] math: S390: Math tests fail with "Exception Inexact set".
2571 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2572 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2573 architectures
2574 [19590] math: Fail to build shared objects that use libmvec.so functions.
2575 \f
2576 Version 2.22
2577
2578 * The following bugs are resolved with this release:
2579
2580 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2581 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2582 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2583 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2584 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2585 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2586 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2587 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2588 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2589 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2590 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2591 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2592 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2593 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2594 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2595 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2596 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2597 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2598 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2599 18657, 18676, 18694, 18696, 18887.
2600
2601 * Cache information can be queried via sysconf() function on s390 e.g. with
2602 _SC_LEVEL1_ICACHE_SIZE as argument.
2603
2604 * A buffer overflow in gethostbyname_r and related functions performing DNS
2605 requests has been fixed. If the NSS functions were called with a
2606 misaligned buffer, the buffer length change due to pointer alignment was
2607 not taken into account. This could result in application crashes or,
2608 potentially arbitrary code execution, using crafted, but syntactically
2609 valid DNS responses. (CVE-2015-1781)
2610
2611 * The time zone file parser has been made more robust against crafted time
2612 zone files, avoiding heap buffer overflows related to the processing of
2613 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2614 large time zone data files. Overly long time zone specifiers in the TZ
2615 variable no longer result in stack overflows and crashes.
2616
2617 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2618 for LD and GD on x86 and x86-64, has been implemented. You will need
2619 binutils-2.24 or later to enable this optimization.
2620
2621 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2622 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2623 Hat). These updates cause user visible changes, such as the fix for bug
2624 17998.
2625
2626 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2627 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2628 condition in some applications.
2629
2630 * Added vector math library named libmvec with the following vectorized x86_64
2631 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2632 pow, powf.
2633 The library can be disabled with --disable-mathvec. Use of the functions is
2634 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2635 Shared library libmvec.so is linked in as needed when using -lm (no need to
2636 specify -lmvec explicitly for not static builds).
2637 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2638
2639 * A new fmemopen implementation has been added with the goal of POSIX
2640 compliance. The new implementation fixes the following long-standing
2641 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2642 old implementation is still present for use be by existing binaries.
2643
2644 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2645 releases. It has been fixed to match 2.19 and older, but binaries built
2646 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2647
2648 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2649 Contributed by Roland McGrath (Google).
2650
2651 * The header <regexp.h> is deprecated, and will be removed in a future
2652 release. Use of this header will trigger a deprecation warning.
2653 Application developers should update their code to use <regex.h> instead.
2654
2655 This header was formerly part of SUS, but was deprecated in 1994 and
2656 removed from the standard in 2001. Also, the glibc implementation
2657 leaks memory. See BZ#18681 for more details.
2658 \f
2659 Version 2.21
2660
2661 * The following bugs are resolved with this release:
2662
2663 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2664 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2665 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2666 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2667 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2668 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2669 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2670 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2671 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2672 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2673 17892.
2674
2675 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2676 too little memory for the to-be-scanned arguments and overflow the
2677 allocated buffer. The implementation now correctly computes the required
2678 buffer size when using malloc, and switches to malloc from alloca as
2679 intended.
2680
2681 * A new semaphore algorithm has been implemented in generic C code for all
2682 machines. Previous custom assembly implementations of semaphore were
2683 difficult to reason about or ensure that they were safe. The new version
2684 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2685 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2686 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2687
2688 * Port to Altera Nios II has been contributed by Mentor Graphics.
2689
2690 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2691 implementations for powerpc64/powerpc64le.
2692 Implemented by Adhemerval Zanella (IBM).
2693
2694 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2695 and powerpc64le. This may improve lock scaling of existing programs on
2696 HTM capable systems. The lock elision code is only enabled with
2697 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2698 powerpc will issue a transaction abort on every syscall to avoid side
2699 effects being visible outside transactions.
2700
2701 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2702 AArch64. Contributed by ARM Ltd.
2703
2704 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2705
2706 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2707 under certain input conditions resulting in the execution of a shell for
2708 command substitution when the application did not request it. The
2709 implementation now checks WRDE_NOCMD immediately before executing the
2710 shell and returns the error WRDE_CMDSUB as expected.
2711
2712 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2713 processing format strings with a large number of format specifiers.
2714
2715 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2716 infinite loop if the DNS response contained a PTR record of an unexpected
2717 format.
2718
2719 * The minimum GCC version that can be used to build this version of the GNU
2720 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2721 still be used to compile programs using the GNU C Library.
2722
2723 * The GNU C Library is now built with -Werror by default. This can be
2724 disabled by configuring with --disable-werror.
2725
2726 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2727
2728 * The obsolete sigvec function has been removed. This was the original
2729 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2730 programs have been using instead for about 25 years. Of course, ABI
2731 compatibility for old binaries using sigvec remains intact.
2732
2733 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2734 with newer versions of bison.
2735
2736 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2737 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2738 registers overlay two consecutive single-precision registers. MIPS32R2
2739 introduced a new FPU mode (FR=1) where double-precision registers extend the
2740 corresponding single-precision registers which is incompatible with the
2741 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2742 require the use of FR=1 making a transition necessary. New o32 ABI
2743 extensions enable users to migrate over time from the original o32 ABI
2744 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2745 tracks the ABI of any loaded object and verifies that new objects are
2746 compatible. Mode transitions will also be requested as required and
2747 unsupportable objects will be rejected. The ABI checks include both soft and
2748 hard float ABIs for o32, n32 and n64.
2749
2750 GCC 5 with GNU binutils 2.25 onwards:
2751 It is strongly recommended that all o32 system libraries are built using the
2752 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2753 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2754 compiler using --with-fp-32=xx to set this by default.
2755 \f
2756 Version 2.20
2757
2758 * The following bugs are resolved with this release:
2759
2760 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2761 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2762 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2763 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2764 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2765 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2766 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2767 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2768 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2769 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2770 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2771 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2772 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2773 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2774 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2775
2776 * Reverted change of ABI data structures for s390 and s390x:
2777 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2778 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2779 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2780 are preserved pointing straight to the same implementation as the old ones.
2781 Given that, new callers will simply provide a too-big buffer to these
2782 functions. Any applications/libraries out there that embed jmp_buf or
2783 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2784 against 2.19 headers will have to rebuilt again. This is necessary in any
2785 case to revert the breakage in their ABI caused by the glibc change.
2786
2787 * Support for file description locks is added to systems running the
2788 Linux kernel. The standard file locking interfaces are extended to
2789 operate on file descriptions, not file descriptors, via the use of
2790 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2791 are associated with an open file instead of a process.
2792
2793 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2794
2795 * The minimum Linux kernel version that this version of the GNU C Library
2796 can be used with is 2.6.32.
2797
2798 * Running the testsuite no longer terminates as soon as a test fails.
2799 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2800 with PASS or FAIL lines for individual tests. A summary of the results is
2801 printed, including a list of failing lists, and "make check" exits with
2802 error status if there were any unexpected failures. "make check
2803 stop-on-test-failure=y" may be used to keep the old behavior.
2804
2805 * The am33 port, which had not worked for several years, has been removed
2806 from ports.
2807
2808 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2809 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2810 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2811 conflicted with POSIX (support for which was removed in 2.19), the
2812 interfaces those macros enabled remain available when compiling with
2813 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2814 test macros defined.
2815
2816 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2817
2818 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2819 This may improve lock scaling of existing programs on TX capable systems.
2820 The lock elision code is only built with --enable-lock-elision=yes and
2821 then requires a GCC version supporting the TX builtins. With lock elision
2822 default mutexes are elided via __builtin_tbegin, if the cpu supports
2823 transactions. By default lock elision is not enabled and the elision code
2824 is not built.
2825
2826 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2827 copy the path argument. This allowed programs to cause posix_spawn to
2828 deference a dangling pointer, or use an unexpected pathname argument if
2829 the string was modified after the posix_spawn_file_actions_addopen
2830 invocation.
2831
2832 * All supported architectures now use the main glibc sysdeps directory
2833 instead of some being in a separate "ports" directory (which was
2834 distributed separately before glibc 2.17).
2835
2836 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2837 On configurations that support it (all Linux configurations), it's now
2838 used regardless of the --enable-add-ons switch to configure. It is no
2839 longer possible to build such configurations without pthreads support.
2840
2841 * Locale names, including those obtained from environment variables (LANG
2842 and the LC_* variables), are more tightly checked for proper syntax.
2843 setlocale will now fail (with EINVAL) for locale names that are overly
2844 long, contain slashes without starting with a slash, or contain ".." path
2845 components. (CVE-2014-0475) Previously, some valid locale names were
2846 silently replaced with the "C" locale when running in AT_SECURE mode
2847 (e.g., in a SUID program). This is no longer necessary because of the
2848 additional checks.
2849
2850 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2851 with application code using Intel MPX instructions. (With all previous
2852 versions, the MPX register state could be clobbered when making calls
2853 into or out of a shared library.) Note that while the new dynamic
2854 linker is compatible with all known x86 hardware whether or not it
2855 supports Intel MPX, some x86 instruction-set emulators might fail to
2856 handle the new instruction encodings. This is known to affect Valgrind
2857 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2858 release), and might affect other tools that do instruction emulation.
2859
2860 * Support for loadable gconv transliteration modules has been removed.
2861 The support for transliteration modules has been non-functional for
2862 over a decade, and the removal is prompted by security defects. The
2863 normal gconv conversion modules are still supported. Transliteration
2864 with //TRANSLIT is still possible, and the //IGNORE specifier
2865 continues to be supported. (CVE-2014-5119)
2866
2867 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2868 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2869 resulting a denial-of-service security vulnerability in applications which
2870 use functions related to iconv. (CVE-2014-6040)
2871 \f
2872 Version 2.19
2873
2874 * The following bugs are resolved with this release:
2875
2876 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2877 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2878 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2879 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2880 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2881 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2882 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2883 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2884 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2885 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2886 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2887 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2888 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2889 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2890 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2891 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2892 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2893 16431, 16453, 16474, 16506, 16510, 16529
2894
2895 * Slovenian translations for glibc messages have been contributed by the
2896 Translation Project's Slovenian team of translators.
2897
2898 * The public headers no longer use __unused nor __block. This change is to
2899 support compiling programs that are derived from BSD sources and use
2900 __unused internally, and to support compiling with Clang's -fblock
2901 extension which uses __block.
2902
2903 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2904 large collation sequences to optimize multiple passes. This cache
2905 computation may overflow for large collation sequences and may cause a
2906 stack or buffer overflow. This is now fixed to use a slower algorithm
2907 which does not use a cache if there is an integer overflow.
2908
2909 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2910 rules for large collation sequences to optimize multiple passes and falls
2911 back to alloca if malloc fails, resulting in a possible stack overflow.
2912 The implementation now falls back to an uncached collation sequence lookup
2913 if malloc fails.
2914
2915 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2916 initialized for static applications resulting in the security feature
2917 being disabled. The pointer guard is now correctly initialized to a
2918 random value for static applications. Existing static applications need
2919 to be recompiled to take advantage of the fix (bug 15754).
2920
2921 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2922 to the d_name member of struct dirent, or omit the terminating NUL
2923 character. (Bugzilla #14699).
2924
2925 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2926 aligned_alloc functions could allocate too few bytes or corrupt the
2927 heap when passed very large allocation size values (Bugzilla #15855,
2928 #15856, #15857).
2929
2930 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2931 for AF_INET6 has been fixed (Bugzilla #16072).
2932
2933 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2934 pap_CW, quz_PE, the_NP.
2935
2936 * Substantially revised locales: gd_GB, ht_HT
2937
2938 * The LC_ADDRESS field was updated to support country_car for almost all
2939 supported locales.
2940
2941 * ISO 1427 definitions were updated.
2942
2943 * ISO 3166 definitions were updated.
2944
2945 * The localedef utility now supports --big-endian and --little-endian
2946 command-line options to generate locales for a different system from that
2947 for which the C library was built.
2948
2949 * Binary locale files now only depend on the endianness of the system for
2950 which they are generated and not on other properties of that system. As a
2951 consequence, binary files generated with new localedef may be incompatible
2952 with old versions of the GNU C Library, and binary files generated with
2953 old localedef may be incompatible with this version of the GNU C Library,
2954 in the following circumstances:
2955
2956 + Locale files may be incompatible on m68k systems.
2957
2958 + Locale archive files (but not separate files for individual locales) may
2959 be incompatible on systems where plain "char" is signed.
2960
2961 * The configure option --disable-versioning has been removed. Builds with
2962 --disable-versioning had not worked for several years.
2963
2964 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2965 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2966
2967 * SystemTap probes for malloc have been introduced.
2968
2969 * SystemTap probes for slow multiple precision fallback paths of
2970 transcendental functions have been introduced.
2971
2972 * Support for powerpc64le has been added.
2973
2974 * The soft-float powerpc port now supports e500 processors.
2975
2976 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2977
2978 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2979 set of header declarations that are enabled by default, even when other
2980 feature test macros or compiler options such as -std=c99 would otherwise
2981 disable some of those declarations.
2982
2983 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2984 conflict with POSIX. The libbsd-compat library (which was a dummy library
2985 that did nothing) has also been removed.
2986
2987 * Preliminary documentation about Multi-Thread, Async-Signal and
2988 Async-Cancel Safety has been added.
2989
2990 * Change of ABI data structures for s390 and s390x:
2991 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2992 allow for future hardware extensions. All library functions that accept or
2993 return these structures were versioned in order to provide backwards
2994 compatibility with existing code. However, not all cases can be handled
2995 with interface versioning. If an application embeds either structure into
2996 another structure and that structure is passed to another compilation unit
2997 which expects the newer larger-sized object then any access to the new
2998 fields will result in undefined behaviour. Similarly any access to fields
2999 that were moved by the enlarging of the structures will cause undefined
3000 behaviour. To fix the undefined behaviour all objects that are part of the
3001 final in-memory image must agree on the size of structures, and this may
3002 require recompilation.
3003 \f
3004 Version 2.18
3005
3006 * The following bugs are resolved with this release:
3007
3008 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
3009 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
3010 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
3011 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
3012 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
3013 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
3014 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
3015 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
3016 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
3017 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
3018 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
3019 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
3020 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
3021 15755, 15759.
3022
3023 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
3024 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
3025 Distributions can re-enable building and using pt_chown via the new configure
3026 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
3027 considerable security risks and should only be used if the distribution
3028 understands and accepts the risks.
3029
3030 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
3031 #15078).
3032
3033 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
3034 fixed (Bugzilla #15330).
3035
3036 * Add support for calling C++11 thread_local object destructors on thread
3037 and program exit. This needs compiler support for offloading C++11
3038 destructor calls to glibc.
3039
3040 * Improved worst case performance of libm functions with double inputs and
3041 output.
3042
3043 * Support for priority inherited mutexes in pthread condition variables on
3044 non-x86 architectures.
3045
3046 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
3047
3048 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
3049
3050 * Optimized string functions for ARM. Implemented by Will Newton and
3051 Richard Henderson.
3052
3053 * Optimized libm functions for SPARC. Implemented by David S. Miller.
3054
3055 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
3056 Richard Henderson.
3057
3058 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
3059 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
3060
3061 * Added a benchmark framework to track performance of functions in glibc.
3062
3063 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
3064 It is based on draft TS 18661 and currently enabled as a GNU extension.
3065
3066 * On Linux, the clock function now uses the clock_gettime system call
3067 for improved precision, rather than old times system call.
3068
3069 * Added support for version-3 tz format files. This is needed when using
3070 the tz database release 2013e or later, and affects a few unusual cases --
3071 currently only TZ='America/Godthab' for time stamps after 2037.
3072
3073 * Added new API functions pthread_getattr_default_np and
3074 pthread_setattr_default_np to get and set the default pthread
3075 attributes of a process.
3076
3077 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
3078 This may improve lock scaling of existing programs on TSX capable systems.
3079 When the --enable-lock-elision=yes parameter is specified at configure
3080 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
3081 mutexes.
3082
3083 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
3084 availability). Implemented by Ryan S. Arnold.
3085
3086 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
3087 \f
3088 Version 2.17
3089
3090 * The following bugs are resolved with this release:
3091
3092 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
3093 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
3094 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
3095 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
3096 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
3097 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
3098 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
3099 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
3100 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
3101 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
3102 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
3103 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
3104 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
3105 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
3106 14879, 14889, 14893, 14898, 14914.
3107
3108 * Optimization of memcpy for MIPS.
3109
3110 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
3111 EMFILE has been fixed (Bugzilla #14889).
3112
3113 * The add-on ports collection is now distributed in the "ports" subdirectory
3114 of the main GNU C Library distribution, rather than separately.
3115
3116 * Port to ARM AArch64 contributed by Linaro.
3117
3118 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
3119 Optimized versions of memcpy, memset, and memcmp added for System z10 and
3120 zEnterprise z196.
3121 Implemented by Andreas Krebbel.
3122
3123 * The new function secure_getenv allows secure access to the environment,
3124 returning NULL if running in a SUID/SGID process. This function replaces
3125 the internal function __secure_getenv.
3126
3127 * SystemTap static probes have been added into the dynamic linker.
3128 Implemented by Gary Benson.
3129
3130 * Optimizations of string functions strstr, strcasestr and memmem.
3131 Implemented by Maxim Kuvyrkov.
3132
3133 * The minimum Linux kernel version that this version of the GNU C Library
3134 can be used with is 2.6.16.
3135
3136 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
3137 powerpc POWER7. Implemented by Will Schmidt.
3138
3139 * New configure option --disable-nscd builds the C library such that it
3140 never attempts to contact the Name Service Caching Daemon (nscd).
3141 New configure option --disable-build-nscd avoids building nscd itself;
3142 this is the default if --disable-nscd is used.
3143
3144 * Improved support for cross-compilation, including cross-testing and
3145 bootstrap builds without a previously built glibc.
3146
3147 * Several testsuite tests are now able to test multiple IFUNC variants of an
3148 interface, rather than just testing the one that would be chooen by
3149 default.
3150
3151 * New configure options --with-bugurl and --with-pkgversion, for
3152 distributors to use to embed their bug-reporting and package version
3153 information in --help and --version output.
3154
3155 * The ttyname and ttyname_r functions on Linux now fall back to searching for
3156 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
3157 allows creation of chroots without the procfs mounted on /proc.
3158
3159 * The `crypt' function now fails if passed salt bytes that violate the
3160 specification for those values. On Linux, the `crypt' function will
3161 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
3162 enabled, and fail on encrypted strings using the MD5 or DES algorithm
3163 when the mode is enabled.
3164
3165 * The `clock_*' suite of functions (declared in <time.h>) is now available
3166 directly in the main C library. Previously it was necessary to link with
3167 -lrt to use these functions. This change has the effect that a
3168 single-threaded program that uses a function such as `clock_gettime' (and
3169 is not linked with -lrt) will no longer implicitly load the pthreads
3170 library at runtime and so will not suffer the overheads associated with
3171 multi-thread support in other code such as the C++ runtime library.
3172
3173 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
3174 sat_IN, and szl_PL.
3175 \f
3176 Version 2.16
3177
3178 * The following bugs are resolved with this release:
3179
3180 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
3181 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
3182 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
3183 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
3184 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
3185 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
3186 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
3187 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
3188 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
3189 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
3190 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
3191 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
3192 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
3193 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
3194 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
3195 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
3196 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
3197 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
3198 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
3199 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
3200 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
3201 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
3202 14277, 14278.
3203
3204 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
3205 configuring glibc with:
3206 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
3207 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
3208 Implemented by H.J. Lu.
3209
3210 * ISO C11 support:
3211
3212 + define static_assert
3213
3214 + do not declare gets
3215
3216 + declare at_quick_exit and quick_exit also for ISO C11
3217
3218 + aligned_alloc. NB: The code is deliberately allows the size parameter
3219 to not be a multiple of the alignment. This is a moronic requirement
3220 in the standard but it is only a requirement on the caller, not the
3221 implementation.
3222
3223 + timespec_get added
3224
3225 + uchar.h support added
3226
3227 + CMPLX, CMPLXF, CMPLXL added
3228
3229 Implemented by Ulrich Drepper.
3230
3231 * Support for the IA-64 has been moved to ports.
3232
3233 * Remove support for anything but ELF binary format
3234
3235 * Checking versions of poll, ppoll added.
3236 Implemented by Ulrich Drepper.
3237
3238 * More generic and 64-bit performance optimizations to math functions.
3239 Implemented by Ulrich Drepper.
3240
3241 * New configure option --enable-obsolete-rpc makes the deprecated RPC
3242 headers and functions available at compile time as they were before
3243 version 2.14. This option will be removed at some time in the future
3244 after the TI-RPC library becomes fully sufficient for the needs of
3245 existing applications.
3246
3247 * Compatibility code for Linux kernel versions before 2.4 has been removed.
3248 Note that glibc is not expected to work with any Linux kernel version
3249 before 2.6.
3250
3251 * New header <sys/auxv.h> and function getauxval allowing easy access to
3252 the AT_* key-value pairs passed from the Linux kernel. The header also
3253 defines the HWCAP_* bits associated with the AT_HWCAP key.
3254
3255 * New locales: mag_IN
3256
3257 * New configure option --enable-systemtap builds SystemTap static probes
3258 into libc for setjmp and longjmp and into libpthread for various operations.
3259 So far the setjmp/longjmp probes and some of the libpthread probes are
3260 provided only for i*86 and x86_64.
3261 Implemented by Roland McGrath and Rayson Ho.
3262
3263 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
3264
3265 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
3266 and Will Schmidt.
3267
3268 * More optimized functions for SPARC. Implemented by David S. Miller.
3269
3270 * Improved support for cross-compilation, especially bootstrap builds
3271 without a previously built glibc.
3272
3273 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
3274 Chris Metcalf from Tilera.
3275
3276 * Support for the old ARM ABI has been removed from ports. Only the EABI is
3277 now supported for ARM processors.
3278
3279 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
3280 as the name of the dynamic linker, to distinguish it from the
3281 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
3282
3283 * Support for CRIS has been removed from ports.
3284
3285 * A new class of installed header has been documented for low-level
3286 platform-specific functionality. PowerPC added the first instance with a
3287 function to provide time base register access. Contributed by Tulio
3288 Magno Quites Machado Filho.
3289
3290 * ix86 configurations now install header files that are consistent with
3291 what x86-64 configurations install. These same header files can be used
3292 for -m32, -m64, or -mx32 builds.
3293 Contributed by H.J. Lu.
3294
3295 * Math library bug fixes. A thorough audit of all open math library bugs was
3296 conducted by Joseph Myers. Significant progress was made on many math
3297 library bugs resulting in more accurate exceptions and function results.
3298 Many thanks to all those that contributed including Andreas Jaeger for his
3299 patch review and work on the x87 trigonometric instruction issues.
3300
3301 * Timezone data is no longer installed. Timezone-related binaries and scripts
3302 will continue to be installed. Users should obtain their timezone data from
3303 their distribution provider or from the tzdata package at
3304 <ftp://munnari.oz.au/pub/>.
3305 \f
3306 Version 2.15
3307
3308 * The following bugs are resolved with this release:
3309
3310 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
3311 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
3312 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
3313 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
3314 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
3315 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
3316 13484, 13506, 13515, 13523, 13524, 13538, 13540
3317
3318 * New program pldd to list loaded object of a process
3319 Implemented by Ulrich Drepper.
3320
3321 * Add nss_db support back to glibc. No more dependency on Berkeley db
3322 and support for initgroups lookups.
3323 Implemented by Ulrich Drepper.
3324
3325 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
3326 Contributed by HJ Lu.
3327
3328 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
3329 Contributed by HJ Lu.
3330
3331 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
3332 on x86-32 and x86-64.
3333 Contributed by Liubov Dmitrieva.
3334
3335 * Optimized strchr and strrchr for SSE on x86-32.
3336 Contributed by Liubov Dmitrieva.
3337
3338 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
3339 for x86-64 and x86-32.
3340 Contributed by Liubov Dmitrieva.
3341
3342 * New interfaces: scandirat, scandirat64
3343 Implemented by Ulrich Drepper.
3344
3345 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
3346 Implemented by Ulrich Drepper.
3347
3348 * nscd now also caches the netgroup database.
3349 Implemented by Ulrich Drepper.
3350
3351 * Integrate libm with gcc's -ffinite-math-only option.
3352 Implemented by Ulrich Drepper.
3353
3354 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
3355 to math functions. Implemented by Ulrich Drepper.
3356
3357 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
3358 Implemented by Ulrich Drepper.
3359
3360 * New Linux interfaces: process_vm_readv, process_vm_writev
3361
3362 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
3363 Implemented by Ulrich Drepper.
3364
3365 * Optimized nearbyint and strcasecmp for PPC.
3366 Implemented by Adhemerval Zanella.
3367
3368 * New locales: bho_IN, unm_US, es_CU, ta_LK
3369 \f
3370 Version 2.14
3371
3372 * The following bugs are resolved with this release:
3373
3374 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
3375 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
3376 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
3377 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
3378 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
3379 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
3380 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
3381 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
3382 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
3383 12795, 12811, 12813, 12814, 12841
3384
3385 * The RPC implementation in libc is obsoleted. Old programs keep working
3386 but new programs cannot be linked with the routines in libc anymore.
3387 Programs in need of RPC functionality must be linked against TI-RPC.
3388 The TI-RPC implementation is IPv6 enabled and there are other benefits.
3389
3390 Visible changes of this change include (obviously) the inability to link
3391 programs using RPC functions without referencing the TI-RPC library and the
3392 removal of the RPC headers from the glibc headers.
3393 Implemented by Ulrich Drepper.
3394
3395 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
3396 syncfs, setns, sendmmsg
3397
3398 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
3399 yue_HK, lij_IT, mhr_RU
3400
3401 * New encodings: CP770, CP771, CP772, CP773, CP774
3402
3403 * New program sotruss to trace calls through PLTs
3404 Implemented by Ulrich Drepper.
3405
3406 * The malloc hook implementation is marked deprecated and will be removed
3407 from the default implementation in the next version. The design never
3408 worked ever since the introduction of threads. Even programs which do
3409 not create threads themselves can use multiple threads created internally.
3410 \f
3411 Version 2.13
3412
3413 * The following bugs are resolved with this release:
3414
3415 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
3416 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
3417 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
3418 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
3419 12378, 12394, 12397
3420
3421 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
3422
3423 * POWER7 optimizations: memset, memcmp, strncmp
3424
3425 * New optimized string functions for x86-64: strnlen (SSE2),
3426 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
3427 Implemented by Ulrich Drepper.
3428 \f
3429 Version 2.12
3430
3431 * The following bugs are resolved with this release:
3432
3433 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
3434 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
3435 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
3436 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
3437 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
3438 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
3439 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
3440 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
3441
3442 * New interfaces: pthread_getname_np, pthread_setname_np
3443
3444 * New Linux interface: recvmmsg
3445
3446 * STT_GNU_IFUNC implemented for Sparc by David Miller.
3447
3448 * The dynamic linker now recognizes supported ABI versions from the
3449 EI_ABIVERSION field in the ELF header.
3450 Implemented by Ulrich Drepper.
3451
3452 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
3453 will not be used to fill in password fields in the passwd.byname replies.
3454 Instead it is used to synthesize the shadow.byname table, should it be
3455 missing. This is a useful mode in some installations involving Solaris.
3456 Implemented by Ulrich Drepper.
3457
3458 * New locales: kok_IN, sq_MK, cv_RU
3459 \f
3460 Version 2.11
3461
3462 * The following bugs are resolved with this release:
3463
3464 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
3465 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
3466 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
3467 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
3468 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
3469 10789, 10817, 10823, 10840, 10847
3470
3471 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
3472 mkostemps64
3473 Implemented by Ulrich Drepper.
3474
3475 * Checking version of longjmp added that fails if an uninitialized stack
3476 frame would be created. Implemented by Ulrich Drepper.
3477
3478 * STT_GNU_IFUNC is now supported in static executables.
3479 Implemented by H.J. Lu.
3480
3481 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
3482
3483 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
3484 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
3485 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
3486 Contributed by H.J. Lu.
3487
3488 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
3489 Implemented by Ulrich Drepper.
3490
3491 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
3492 strstr, strcasestr.
3493 Contributed by H.J. Lu.
3494
3495 * Support for fma instruction in AVX on x86-64.
3496 Implemented by H.J. Lu and Ulrich Drepper.
3497
3498 * AVX support in x86-64 auditing support in ld.so.
3499 Implemented by H.J. Lu.
3500
3501 * STB_GNU_UNIQUE support added.
3502 Implemented by Ulrich Drepper.
3503
3504 * Implement second fallback mode for optimized DNS lookup for even more
3505 broken environments. If two requests from the same port are not
3506 handled correctly close the socket and open a new one before sending
3507 the second request. The 'single-request-reopen' option in /etc/resolv.conf
3508 can be used to select this mode right away, instead of rediscovering the
3509 necessity is every process again.
3510 Implemented by Ulrich Drepper.
3511
3512 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3513 Implemented by Adam Tkac.
3514
3515 * Optimized iconv conversions for S390x.
3516 Implemented by Andreas Krebbel.
3517
3518 * Using condvars with PI mutexes is now more efficient due to kernel
3519 support for requeueing to PI futexes. NPTL support added for x86-64.
3520 Implemented by Ulrich Drepper.
3521
3522 * New locales: ps_AF, my_MM
3523 \f
3524 Version 2.10
3525
3526 * The following bugs are resolved with this release:
3527
3528 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3529 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3530 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3531 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3532 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3533
3534 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3535 Implemented by Ulrich Drepper.
3536
3537 * New Linux interfaces: accept4, fallocate, fallocate64.
3538 Implemented by Ulrich Drepper.
3539
3540 * Correct declarations of string function when used in C++ code. This
3541 could lead to compile errors for invalid C++ code.
3542
3543 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3544 now in POSIX.
3545
3546 * New POSIX 2008 interface: psiginfo
3547 Implemented by Ulrich Drepper.
3548
3549 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3550 Implemented by Ulrich Drepper.
3551
3552 * Support for selecting between multiple function definitions at runtime
3553 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3554
3555 * The libcrypt library can now use the hash function implementations in
3556 NSS. Implemented by Ulrich Drepper.
3557
3558 * The malloc implementation can be compiled to be less memory efficient
3559 but higher performing in multi-threaded programs.
3560 Implemented by Ulrich Drepper.
3561
3562 * New locales: nan_TW@latin, ks_IN
3563
3564 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3565 Implemented by Ulrich Drepper.
3566
3567 * Extended printf hook support. It is possible to use user-defined types
3568 and extend existing format specifiers.
3569 Implemented by Ulrich Drepper.
3570
3571 * Handling for group shadow files has been added.
3572 Implemented by Ulrich Drepper.
3573
3574 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3575 servers (the case, e.g., for some people using the built-in DNS
3576 server in ADSL modems/routers). There is a once-per-process timeout
3577 in case of a broken server. To avoid it, users can run nscd or put
3578 'options single-request' in /etc/resolv.conf.
3579 Implemented by Ulrich Drepper.
3580 \f
3581 Version 2.9
3582
3583 * The following bugs are resolved with this release:
3584
3585 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3586 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3587 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3588 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3589 6968, 6974, 6980, 6995, 7008, 7009, 7029
3590
3591 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3592 up at the same time. Implemented by Ulrich Drepper.
3593
3594 * TLS descriptors for LD and GD on x86 and x86-64.
3595 Implemented by Alexandre Oliva.
3596
3597 * getaddrinfo now handles DCCP and UDPlite.
3598 Implemented by Ulrich Drepper.
3599
3600 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3601 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3602 Implemented by Ulrich Drepper.
3603
3604 * New implementation of memmem, strstr, and strcasestr which is O(n).
3605 Implemented by Eric Blake.
3606
3607 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3608
3609 * Implement "e" option for popen to open file descriptor with the
3610 close-on-exec flag set. Implemented by Ulrich Drepper.
3611
3612 * Implement "b" mode for fmemopen. In this mode writes writes don't
3613 implicitly add a NUL byte and seeks from the end of the buffer really
3614 use the buffer end, not the string length as the basis.
3615 Implemented by Ulrich Drepper.
3616
3617 * Many functions, exported and internal, now atomically set the close-on-exec
3618 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3619
3620 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3621 Sinhala)
3622 Implemented by Pravin Satpute.
3623
3624 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3625 \f
3626 Version 2.8
3627
3628 * The following bugs are resolved with this release:
3629
3630 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3631 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3632 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3633 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3634 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3635 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3636 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3637
3638 * New locales: bo_CN, bo_IN, shs_CA.
3639
3640 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3641
3642 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3643 Implemented by Pravin Satpute.
3644
3645 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3646
3647 * nscd caches now all timeouts for DNS entries
3648 Implemented by Ulrich Drepper.
3649
3650 * nscd is more efficient and wakes up less often.
3651 Implemented by Ulrich Drepper.
3652
3653 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3654 vdprintf, and obstack_vprintf.
3655 Implemented by Jakub Jelinek.
3656
3657 * Faster memset for x86-64.
3658 Implemented by Harsha Jagasia and H.J. Lu.
3659
3660 * Faster memcpy on x86.
3661 Implemented by Ulrich Drepper.
3662
3663 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3664 Implemented by Ulrich Drepper.
3665
3666 * Faster sqrt and sqrtf implemention for some PPC variants.
3667 Implemented by Steven Munroe.
3668 \f
3669 Version 2.7
3670
3671 * The following bugs are resolved with this release:
3672
3673 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3674 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3675 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3676 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3677 5071, 5103, 5104, 5112, 5113, 5184, 5186
3678
3679 * More checking functions: fread, fread_unlocked, open*, mq_open.
3680 Implemented by Jakub Jelinek and Ulrich Drepper.
3681
3682 * Extend fortification to C++. Implemented by Jakub Jelinek.
3683
3684 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3685 by not recognizing 'a' as a modifier when those specs are requested.
3686 Implemented by Jakub Jelinek.
3687
3688 * PPC optimizations to math and string functions.
3689 Implemented by Steven Munroe.
3690
3691 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3692 options to be passed. Implemented by Ulrich Drepper.
3693
3694 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3695
3696 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3697 Implemented by Ulrich Drepper.
3698
3699 * Handle private futexes in the NPTL implementation.
3700 Implemented by Jakub Jelinek and Ulrich Drepper.
3701
3702 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3703 Implemented by Roland McGrath and Ulrich Drepper.
3704
3705 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3706
3707 * SHA-256 and SHA-512 based password encryption.
3708 Implemented by Ulrich Drepper.
3709
3710 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3711 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3712 yo_NG.
3713
3714 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3715 Implemented by Ulrich Drepper.
3716 \f
3717 Version 2.6
3718
3719 * The following bugs are resolved with this release:
3720
3721 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3722 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3723 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3724 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3725 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3726 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3727 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3728 4702, 4858
3729
3730 * New Linux interfaces: epoll_pwait, sched_getcpu.
3731
3732 * New generic interfaces: strerror_l.
3733
3734 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3735 \f
3736 Version 2.5
3737
3738 * The following bugs are resolved with this release:
3739
3740 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3741 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3742 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3743 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3744 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3745 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3746 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3747 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3748 3225, 3251, 3252, 3253, 3273
3749
3750 * For Linux, the sorting of addresses returned by getaddrinfo now also
3751 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3752 Implemented by Ulrich Drepper.
3753
3754 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3755 Implemented by Ulrich Drepper.
3756
3757 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3758
3759 * New iconv module for MIK. Contributed by Alexander Shopov.
3760
3761 * For sites with broken group and/or passwd database, the auto-propagate
3762 option of nscd can prevent creating ID lookup entries from the results
3763 of a name lookup and vice versa. This usually is no problem but some
3764 site might have problems with the default behavior.
3765 Implemented by Ulrich Drepper.
3766
3767 * Iterating over entire database in NIS can be slow. With the
3768 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3769 to trade time for memory. The entire database will be read at once.
3770 Implemented by Ulrich Drepper.
3771
3772 * The interfaces introduced in RFC 3542 have been implemented by
3773 Ulrich Drepper.
3774
3775 * Support for the new ELF hash table format was added by Ulrich Drepper.
3776
3777 * Support for priority inheritance mutexes added by Jakub Jelinek and
3778 Ulrich Drepper.
3779
3780 * Support for priority protected mutexes added by Jakub Jelinek.
3781
3782 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3783 or_IN, csb_PL, fy_NL, sr_ME.
3784 \f
3785 Version 2.4
3786
3787 * More overflow detection functions.
3788
3789 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3790 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3791
3792 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3793 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3794 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3795 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3796 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3797 by Masahide Washizawa.
3798
3799 * It is now possible to install audit libraries for the dynamic linker using
3800 LD_AUDIT. Implemented by Ulrich Drepper.
3801
3802 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3803 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3804 For a libc and libpthread that works well on Linux 2.4 kernels, we
3805 recommend using the stable 2.3 branch.
3806
3807 * The new function open_wmemstream acts like open_memstream,
3808 but uses a wchar_t wide character buffer.
3809
3810 * The new function ppoll is an improved version of poll, similar to pselect.
3811
3812 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3813 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3814 renameat, unlinkat, linkat, symlinkat, readlinkat.
3815
3816 * New Linux kernel system calls: unshare,
3817 inotify_init, inotify_add_watch, inotify_rm_watch.
3818
3819 * The euidaccess function is now also known by the alias eaccess,
3820 for compatibility with some other systems.
3821
3822 * Timezone data updated to 2006b version.
3823 \f
3824 Version 2.3.6
3825
3826 * The following bugs are resolved with this release:
3827
3828 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3829 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3830 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3831 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3832 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3833 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3834
3835 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3836
3837 * As of this release, GCC 4 can be used to compile the C Library.
3838
3839 * Timezone data updated to 2005m version.
3840 \f
3841 Version 2.3.5
3842
3843 * The following bugs are resolved with this release:
3844
3845 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3846 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3847 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3848 777, 787, 821, 822, 823, 825
3849
3850 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3851 \f
3852 Version 2.3.4
3853
3854 * Support for RFC 3678. Real implementations exist only for Linux so far.
3855 Implemented by Ulrich Drepper.
3856
3857 * nscd can now cache entries persistently. Expiring entries are reloaded.
3858 For speedups the cache can be shared in memory with client processes.
3859 Implemented by Ulrich Drepper.
3860
3861 * nscd can now perform SELinux checks.
3862 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3863
3864 * getaddrinfo queries are now cached. Canonical name lookup is performed
3865 efficiently.
3866 Implemented by Ulrich Drepper.
3867
3868 * The nothrow function attribute is used when headers are used by gcc when
3869 compiling C code. This can avoid significant amounts of exception
3870 handling data.
3871
3872 * The malloc functions perform more error checking and are stricter when
3873 it comes to reacting on errors. The default action is to terminate
3874 the process after showing an error message. Implemented by Ulrich Drepper.
3875
3876 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3877 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3878
3879 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3880 namespaces using the new function dlmopen(). This feature is of course,
3881 like most other dynamic loading functionality, not available in statically
3882 linked applications. Implemented by Ulrich Drepper.
3883
3884 * Low-overhead boundary checking variants of string and some stdio functions
3885 were added. These are to be used in conjunction with a gcc patch by
3886 Jakub Jelinek which adds calls to these functions if possible.
3887 Implemented by Jakub Jelinek and Ulrich Drepper.
3888
3889 * Old code for several operating systems and machine architectures that
3890 have not been in working condition in a long time have been removed from
3891 the main source tree maintained by the GNU C Library's maintainers.
3892 These files are now reside in the separate `ports' source module
3893 that is usable as an add-on when building the library.
3894 \f
3895 Version 2.3.3
3896
3897 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3898 interrogate the dynamic linker, compatible with the Solaris interface.
3899
3900 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3901 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3902
3903 * getifaddrs now uses the netlink interface on Linux to get its information.
3904 Implemented by Thorsten Kukuk.
3905
3906 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3907 Implemented by Ulrich Drepper.
3908
3909 * support for non-executable stacks on x86 has been added. Changes mostly
3910 by Roland McGrath.
3911
3912 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3913 and Ulrich Drepper.
3914
3915 * getaddrinfo now performs destination address selection according to
3916 RFC 3484.
3917 \f
3918 Version 2.3.2
3919
3920 * Thread-safe interfaces for many functions that access locale data
3921 were added in version 2.3, but these features were omitted from NEWS.
3922 Many functions have variants with an `_l' suffix that take a `locale_t'
3923 object as a parameter rather than consulting the current locale.
3924 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3925 create and maintain `locale_t' objects. Additionally, the new function
3926 `uselocale' sets "the current locale" (as used by functions not so
3927 parameterized) set for an individual thread. These features were added
3928 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3929
3930 * The functions getresuid, getresgid, setresuid, and setresgid, which
3931 have long been available on Linux, are now declared in <unistd.h>
3932 and are now also available on the Hurd.
3933
3934 * ELF thread-local storage support (TLS) now works on x86-64.
3935
3936 * The new dynamic string token $LIB is expanded in shared library names.
3937 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3938
3939 * Aldy Hernandez contributed complete software floating point support for
3940 PowerPC machines with no FPU.
3941
3942 * fexecve is implemented on Linux.
3943
3944 * The `btowc' function should work at least twice as fast due to
3945 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3946
3947 * With approriate thread add-ons cancelable functions are now implemented
3948 in libc.so as well. No need to call the function in libpthread. This
3949 change allowed to finally disable the incorrect and expensive handling
3950 of weak definition in ld.so.
3951
3952 * Yet more PLT entries in libc.so have been removed. We finally arrived
3953 at the bare minimum. Startup times improved appropriately.
3954
3955 * Support for the new Linux/x86 system call interface was added. The
3956 AT_SYSINFO auxiliary vector entry is recognized and handled.
3957 \f
3958 Version 2.3
3959
3960 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3961 charsets.
3962
3963 * iconv (the program and the interface) now accepts empty names (excluding
3964 options like //TRANSLIT) to mean "use charset of current locale".
3965
3966 * localedef can now transliterate characters in strings which are not in
3967 the provided charmap. The information from the input locale is used.
3968
3969 * Prelinking support was added for ELF targets. This requires additional
3970 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3971
3972 * Read-only stdio streams now use mmap to speed up operation by eliminating
3973 copying and buffer underflows. To use add 'm' to the mode string of
3974 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3975
3976 * The malloc functions were completely rewritten by Wolfram Gloger based
3977 on Doug Lea's malloc-2.7.0.c.
3978
3979 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3980 implementation of regex.
3981
3982 * Bruno Haible upgraded the iconv and locale implementation to support
3983 Unicode 3.2.
3984
3985 * Contents of the LC_* and LANG environment variables in the CEN style are
3986 not recognized anymore. It never was used. Change by Ulrich Drepper.
3987
3988 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3989 thread-local storage (TLS) ABI on some platforms.
3990 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3991
3992 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3993 EUC-JISX0213, and TSCII.
3994
3995 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3996 BSD-compatible interface for getting all network interface addresses.
3997 Implementation for IPv4 by Roland McGrath.
3998
3999 * Loading of locale data is faster due to the introduction of a locale
4000 archive. Implemented by Roland McGrath and Ulrich Drepper.
4001
4002 * Startup times are significantly reduced by not using exported functions
4003 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
4004 and Ulrich Drepper.
4005
4006 * Steven Munroe contributed a port to PowerPC64/Linux.
4007 \f
4008 Version 2.2.6
4009
4010 * The Hurd now uses the GNU libio implementation of stdio rather than the
4011 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
4012
4013 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
4014 with the same behavior as the Linux system call of the same name.
4015 \f
4016 Version 2.2.5
4017
4018 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
4019 128-bit long double format.
4020
4021 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
4022 IBM1161, and IBM1162 charsets.
4023
4024 * Andreas Jaeger contributed a port to x86-64/Linux.
4025
4026 * Peter Bruin contributed a port to PowerPC/Hurd.
4027
4028 * libc functions using I/O streams now can handle wide-oriented streams
4029 as well.
4030
4031 * optimizations in the dynamic linker. Binaries created by recent binutils
4032 versions start up quicker due to reduced time spent on relocations.
4033
4034 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
4035 \f
4036 Version 2.2.4
4037
4038 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
4039 asinh, atanh, j0 for the 128-bit long double format.
4040
4041 * Bruno Haible updated all the code handling Unicode in some form to
4042 support Unicode 3.1.
4043
4044 * Speed of regex for single-byte locales is back to previous levels.
4045 Patch by Isamu Hasegawa.
4046
4047 * Alpha, SPARC, and IA-64 now also using floating stacks.
4048
4049 * Startup time of internationalized applications greatly improved through
4050 iconv cache. Use iconvconfig to generate the cache file.
4051 Contributed by Ulrich Drepper.
4052
4053 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
4054 severe performance problems. Patch by David Mosberger.
4055
4056 * The Hurd port got a lot more functionality like AIO, various stdio
4057 extensions, etc. Mainly done by Roland McGrath.
4058
4059 * mtrace can now lookup symbols in shared libraries.
4060 \f
4061 Version 2.2.3
4062
4063 * Intel's IA-64 math library is largely integrated. It provides fast and
4064 accurate implementatations for most basic and standard math functions
4065 in float, double, and long double format.
4066
4067 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
4068 and asin for the 96-bit long double format and asin, log, tan for the
4069 128-bit long double format.
4070
4071 * The beginning of a last-bit accurate math library by IBM Haifa were added.
4072 The basic double functions exist today. Contributed by Abraham Ziv
4073 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
4074 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
4075
4076 * An asynchronous name lookup library was added. The interface is designed
4077 after POSIX AIO. The proposal was circulated beforehand to get comments.
4078 No negative ones came in. Implemented by Ulrich Drepper.
4079
4080 * Port to S390/64bit contributed by Martin Schwidefsky
4081 <schwidefsky@de.ibm.com>.
4082
4083 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
4084 of functions for Linux/IA-64.
4085
4086 * The RPC code is now thread safe. Threads can now use the same service
4087 of different services at the same time. Patch by Eric Norum
4088 <eric.norum@usask.ca> with some help by Ulrich Drepper.
4089
4090 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
4091 family of functions for Linux/S390.
4092
4093 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
4094 of functions for Linux/x86.
4095
4096 * Port to Linux/CRIS contributed by Axis Communications.
4097 \f
4098 Version 2.2.2
4099
4100 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
4101 we can now check for namespace violations and missing declarations. The
4102 result is that almost all headers are now Unix-compliant (as defined in
4103 the upcoming XPG6). The negative side is that some programs might need
4104 corrections, too, if they depend on the incorrect form of the headers in
4105 previous versions which defined too many symbols and included too many
4106 other headers.
4107
4108 * regex now handles multibyte character sets correctly.
4109 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
4110
4111 * iconv (the program) does now conform to the upcoming XPG6 and handles
4112 charmaps. Instead of the charset names the path of charmaps can be
4113 provided and the conversion happens based on this data.
4114 Contributed by Ulrich Drepper.
4115
4116 * The locale program now provides detailed information about the installed
4117 locales. While
4118
4119 locale -a
4120
4121 only lists the names of the supported locales
4122
4123 locale -a --verbose
4124
4125 provides details such as country, language, and codeset name.
4126 Contributed by Ulrich Drepper.
4127 \f
4128 Version 2.2.1
4129
4130 * The gencat program now parses the input file according to the charset
4131 selected by the LC_CTYPE category. This is important for stateful
4132 character sets. To make generating catalogs easier there is a way
4133 to overwrite the charset selected by the locale: before the first
4134 message or $ quote line the catalog can contain a line like
4135
4136 $ codeset=ISO-8859-2
4137
4138 to select the charset (ISO-8859-2 in this case).
4139
4140 Implemented by Shinya Hanataka and Ulrich Drepper.
4141
4142 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
4143 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
4144 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
4145
4146 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
4147 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
4148
4149 * The LANGUAGE environment variable is now ignored unless the locale is
4150 changed from the default "C" locale.
4151
4152 * The usual bug fixes.
4153 \f
4154 Version 2.2
4155
4156 * Greg McGary added runtime support for bounds checking using gcc's
4157 new -fbounded-pointers option. ix86 target is complete. PowerPC
4158 is in progress.
4159
4160 * Thorsten Kukuk added secure mode support to nscd.
4161
4162 * The Berkeley db libraries have been removed.
4163
4164 Related, the nss_db module is now in a separate package since it
4165 obviously requires a database library being available.
4166
4167 * Wide character I/O streams implemented by Ulrich Drepper.
4168
4169 * Functions from the extended socket API added by Ulrich Drepper.
4170
4171 * Functions feenableexcept and fedisableexcept to control the
4172 behaviour of individual exceptions have been added by Andreas Jaeger.
4173
4174 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
4175
4176 * The resolver code has been updated from bind 8.2.3-T5B which supports
4177 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
4178 and Mark Kettenis.
4179
4180 This change could in some situations effect backward compatibility. Since
4181 now `_res' is a thread-local instead of a global variable, modifying it
4182 in one thread does not have any effect in other threads.
4183
4184 The resolver library was also extended to allow IPv6 as the transport
4185 protocol for the requests. This work was done by Stig Venaas.
4186
4187 * Compatibility code for K&R C compilers has been removed from the
4188 header files. A ISO C compiler is needed to use the library
4189 (conforming to either C89 or C99 standard).
4190
4191 * Complete rewrite of the localedef program to support multibyte character
4192 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
4193 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
4194 Implemented by Ulrich Drepper.
4195
4196 Bruno Haible significantly improved the generation and use of the data
4197 structures for the wide character tables.
4198
4199 * Plural handling in gettext implemented by Ulrich Drepper.
4200
4201 * The utmp daemon has been removed.
4202
4203 * The port to MIPS-Linux has been finished by Andreas Jaeger.
4204
4205 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
4206 and Yutaka Niibe.
4207
4208 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
4209
4210 * POSIX barriers implemented by Kaz Kylheku.
4211
4212 * POSIX spawn function family implemented by Ulrich Drepper.
4213
4214 * POSIX spinlocks are now available.
4215
4216 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
4217
4218 * the configure option --enable-kernel=X.Y.Z allows to strip out
4219 compatibility for kernel versions before X.Y.Z. This is currently only
4220 implemented for Linux.
4221
4222 * the sockaddr_in6 structure changed. The IPv6 working group added a new
4223 field sin6_scope_id. This means that all programs using IPv6 should be
4224 recompiled. Don't expect binary compatibility with previous glibc
4225 versions.
4226
4227 * various conversion modules for IBM character sets contributed by
4228 Masahide Washizawa.
4229
4230 * IA-64 port by Jes Sorensen and HJ Lu.
4231 \f
4232 Version 2.1.3
4233
4234 * bug fixes
4235
4236 \f
4237 Version 2.1.2
4238
4239 * bug fixes
4240
4241 \f
4242 Version 2.1.1
4243
4244 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
4245
4246 * New xdr functions are added; some rpc functions are now 64bit clean.
4247
4248 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
4249
4250 * Fixed known incompatibilities with glibc 2.0.
4251
4252 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
4253
4254 * Optimized a number of functions (especially the ELF dynamic loader).
4255
4256 * Update timezone data files.
4257
4258 * lots of charmaps corrections
4259
4260 * some new locale definitions and charmaps
4261
4262 \f
4263 Version 2.1
4264
4265 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
4266 conform to POSIX member type requirements. Symbol versions have been
4267 adjusted as needed within the library, and for direct use by applications,
4268 but there is potential for problems if third-party libraries use
4269 struct timeval as part of their interface. This does not present
4270 a problem for X and other "essential" system libraries.
4271
4272 * An additional locale model to support C++ Standard Library locale
4273 model and probably more was implemented by Ulrich Drepper.
4274
4275 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
4276 symbol level.
4277
4278 * Miles Bader provided the `argp' function family to support hierachical
4279 command line argument parsing, layered on top of getopt.
4280
4281 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
4282
4283 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
4284 numbers.
4285
4286 * scanf recognizes the %a and %A format for scanning floating point numbers.
4287
4288 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
4289 information and interfaces for the available integer types.
4290
4291 * about 130 new math functions were added to implement the ISO C9x math
4292 library.
4293
4294 * the new header <complex.h> contains definitions of the complex math
4295 functions from ISO C 9X.
4296
4297 * the new header <tgmath.h> defines generic macros to use complex or
4298 real valued functions.
4299
4300 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
4301
4302 * Andreas Jaeger provided a test suite for the math library.
4303
4304 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
4305
4306 * Ulrich Drepper added character set conversion functions (iconv).
4307
4308 * Optimized string functions have been added.
4309
4310 * The localedata addon is now part of glibc.
4311
4312 * An implementation of profiling shared libraries was added by Ulrich Drepper.
4313
4314 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
4315 daemon for NSS (nscd).
4316
4317 Missing a better place here are some numbers on improvements. Under
4318 Linux 2.1.125 un-tar-ing the kernel sources takes
4319
4320 user system wall
4321
4322 using local files 12.19s 6.88s 22.91s
4323
4324 using NIS 13.92s 8.91s 26.34s
4325
4326 using NIS & nscd 10.37s 7.34s 25.30s
4327
4328 using NIS+ 27.57s 30.37s 640.46s
4329
4330 using NIS+ & nscd 10.25s 7.83s 26.51s
4331
4332 using NIS & old nscd [1] 13.83s 8.32s 29.60s
4333
4334 Keep in mind that non-namelookup related operations dominate above times.
4335 It was just a common complain that using NIS+ unpacking the kernel is
4336 horribly slow.
4337
4338 [1] The old nscd implementation is not available anymore since it was
4339 distributed with glibc up to version 2.0.98 and thus is now replaced.
4340
4341 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
4342
4343 * Mark Kettenis provided a Hesiod NSS module.
4344
4345 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
4346 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
4347
4348 * rcmd can now handle netgroups (Dick Streefland).
4349
4350 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
4351 Bambrough.
4352
4353 * Support for the IPv6 protocol has been added to the socket API, as per the
4354 latest draft standards.
4355
4356 * Support for Linux 2.2 has been added.
4357
4358 * Interface changes relative to the latest 2.0.x release:
4359 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4360 addseverity NEW: Unix98
4361 alphasort64 NEW: LFS
4362 argp_err_exit_status NEW: argp, GNU ext
4363 argp_error NEW: argp, GNU ext
4364 argp_failure NEW: argp, GNU ext
4365 argp_help NEW: argp, GNU ext
4366 argp_parse NEW: argp, GNU ext
4367 argp_program_bug_address NEW: argp, GNU ext
4368 argp_program_version NEW: argp, GNU ext
4369 argp_program_version_hook NEW: argp, GNU ext
4370 argp_state_help NEW: argp, GNU ext
4371 argp_usage NEW: argp, GNU ext
4372 authdes_create NEW: Secure RPC
4373 authdes_getucred NEW: Secure RPC
4374 authdes_pk_create NEW: Secure RPC
4375 backtrace NEW: GNU ext.
4376 backtrace_symbols NEW: GNU ext.
4377 backtrace_symbols_fd NEW: GNU ext.
4378 cacos NEW: ISO C 9x
4379 cacosf NEW: ISO C 9x
4380 cacosh NEW: ISO C 9x
4381 cacoshf NEW: ISO C 9x
4382 cacoshl NEW: ISO C 9x
4383 cacosl NEW: ISO C 9x
4384 capget NEW: kernel
4385 capset NEW: kernel
4386 carg NEW: ISO C 9x
4387 cargf NEW: ISO C 9x
4388 cargl NEW: ISO C 9x
4389 casin NEW: ISO C 9x
4390 casinf NEW: ISO C 9x
4391 casinh NEW: ISO C 9x
4392 casinhf NEW: ISO C 9x
4393 casinhl NEW: ISO C 9x
4394 casinl NEW: ISO C 9x
4395 catan NEW: ISO C 9x
4396 catanf NEW: ISO C 9x
4397 catanh NEW: ISO C 9x
4398 catanhf NEW: ISO C 9x
4399 catanhl NEW: ISO C 9x
4400 catanl NEW: ISO C 9x
4401 cbc_crypt NEW: Secure RPC
4402 ccos NEW: ISO C 9x
4403 ccosf NEW: ISO C 9x
4404 ccosh NEW: ISO C 9x
4405 ccoshf NEW: ISO C 9x
4406 ccoshl NEW: ISO C 9x
4407 ccosl NEW: ISO C 9x
4408 cexp NEW: ISO C 9x
4409 cexpf NEW: ISO C 9x
4410 cexpl NEW: ISO C 9x
4411 cimag NEW: ISO C 9x
4412 cimagf NEW: ISO C 9x
4413 cimagl NEW: ISO C 9x
4414 clearerr_locked REMOVED
4415 clntunix_create NEW: sunrpc ext
4416 clog NEW: ISO C 9x
4417 clog10 NEW: ISO C 9x
4418 clog10f NEW: ISO C 9x
4419 clog10l NEW: ISO C 9x
4420 clogf NEW: ISO C 9x
4421 clogl NEW: ISO C 9x
4422 conj NEW: ISO C 9x
4423 conjf NEW: ISO C 9x
4424 conjl NEW: ISO C 9x
4425 cpow NEW: ISO C 9x
4426 cpowf NEW: ISO C 9x
4427 cpowl NEW: ISO C 9x
4428 cproj NEW: ISO C 9x
4429 cprojf NEW: ISO C 9x
4430 cprojl NEW: ISO C 9x
4431 creal NEW: ISO C 9x
4432 crealf NEW: ISO C 9x
4433 creall NEW: ISO C 9x
4434 creat64 NEW: LFS
4435 csin NEW: ISO C 9x
4436 csinf NEW: ISO C 9x
4437 csinh NEW: ISO C 9x
4438 csinhf NEW: ISO C 9x
4439 csinhl NEW: ISO C 9x
4440 csinl NEW: ISO C 9x
4441 csqrt NEW: ISO C 9x
4442 csqrtf NEW: ISO C 9x
4443 csqrtl NEW: ISO C 9x
4444 ctan NEW: ISO C 9x
4445 ctanf NEW: ISO C 9x
4446 ctanh NEW: ISO C 9x
4447 ctanhf NEW: ISO C 9x
4448 ctanhl NEW: ISO C 9x
4449 ctanl NEW: ISO C 9x
4450 des_setparity NEW: Secure RPC
4451 ecb_crypt NEW: Secure RPC
4452 endutxent NEW: Unix98
4453 exp10 NEW: ISO C 9x
4454 exp10f NEW: ISO C 9x
4455 exp10l NEW: ISO C 9x
4456 exp2 NEW: ISO C 9x
4457 exp2f NEW: ISO C 9x
4458 exp2l NEW: ISO C 9x
4459 fattach NEW: STREAMS
4460 fdetach NEW: STREAMS
4461 fdim NEW: ISO C 9x
4462 fdimf NEW: ISO C 9x
4463 fdiml NEW: ISO C 9x
4464 feclearexcept NEW: ISO C 9x
4465 fegetenv NEW: ISO C 9x
4466 fegetexceptflag NEW: ISO C 9x
4467 fegetround NEW: ISO C 9x
4468 feholdexcept NEW: ISO C 9x
4469 feof_locked REMOVED
4470 feraiseexcept NEW: ISO C 9x
4471 ferror_locked REMOVED
4472 fesetenv NEW: ISO C 9x
4473 fesetexceptflag NEW: ISO C 9x
4474 fesetround NEW: ISO C 9x
4475 fetestexcept NEW: ISO C 9x
4476 feupdateenv NEW: ISO C 9x
4477 fflush_locked REMOVED
4478 ffsl NEW: GNU ext.
4479 ffsll NEW: GNU ext.
4480 fgetpos64 NEW: LFS
4481 fgets_unlocked NEW: GNU ext.
4482 fileno_locked REMOVED
4483 fma NEW: ISO C 9x
4484 fmaf NEW: ISO C 9x
4485 fmal NEW: ISO C 9x
4486 fmax NEW: ISO C 9x
4487 fmaxf NEW: ISO C 9x
4488 fmaxl NEW: ISO C 9x
4489 fmin NEW: ISO C 9x
4490 fminf NEW: ISO C 9x
4491 fminl NEW: ISO C 9x
4492 fmtmsg NEW: Unix98
4493 fopen64 NEW: LFS
4494 fputc_locked REMOVED
4495 fputs_unlocked NEW: GNU ext.
4496 fread_unlocked NEW: GNU ext.
4497 freopen64 NEW: LFS
4498 fseeko NEW: Unix98
4499 fsetpos64 NEW: LFS
4500 fstatfs64 NEW: LFS
4501 fstatvfs NEW: Unix98
4502 fstatvfs64 NEW: LFS
4503 ftello NEW: Unix98
4504 ftello64 NEW: LFS
4505 ftruncate64 NEW: LFS
4506 ftw64 NEW: LFS
4507 fwrite_unlocked NEW: GNU ext.
4508 gai_strerror NEW: IPv6
4509 gamma_r REMOVED
4510 gammaf_r REMOVED
4511 gammal_r REMOVED
4512 getchar_locked REMOVED
4513 getdate NEW: Unix98
4514 getdate_err NEW: Unix98
4515 getdate_r NEW: GNU ext.
4516 getmsg NEW: STREAMS
4517 getnameinfo NEW: IPv6
4518 getnetname NEW: Secure RPC
4519 getpmsg NEW: STREAMS
4520 getpt NEW: Unix98 PTY
4521 getrlimit64 NEW: LFS
4522 getutxent NEW: Unix98
4523 getutxid NEW: Unix98
4524 getutxline NEW: Unix98
4525 glob64 NEW: GNU ext.
4526 globfree64 NEW: GNU ext.
4527 gnu_get_libc_release NEW: GNU ext.
4528 gnu_get_libc_version NEW: GNU ext.
4529 grantpt NEW: Unix98 PTY
4530 host2netname NEW: Secure RPC
4531 iconv NEW: iconv
4532 iconv_close NEW: iconv
4533 iconv_open NEW: iconv
4534 if_freenameindex NEW: IPv6
4535 if_indextoname NEW: IPv6
4536 if_nameindex NEW: IPv6
4537 if_nametoindex NEW: IPv6
4538 in6addr_any NEW: IPv6
4539 in6addr_loopback NEW: IPv6
4540 inet6_isipv4mapped NEW: IPv6
4541 isastream NEW: STREAMS
4542 iswblank NEW: GNU ext.
4543 key_decryptsession NEW: Secure RPC
4544 key_decryptsession_pk NEW: Secure RPC
4545 key_encryptsession NEW: Secure RPC
4546 key_encryptsession_pk NEW: Secure RPC
4547 key_gendes NEW: Secure RPC
4548 key_get_conv NEW: Secure RPC
4549 key_secretkey_is_set NEW: Secure RPC
4550 key_setnet NEW: Secure RPC
4551 key_setsecret NEW: Secure RPC
4552 llrint NEW: ISO C 9x
4553 llrintf NEW: ISO C 9x
4554 llrintl NEW: ISO C 9x
4555 llround NEW: ISO C 9x
4556 llroundf NEW: ISO C 9x
4557 llroundl NEW: ISO C 9x
4558 log2 NEW: ISO C 9x
4559 log2f NEW: ISO C 9x
4560 log2l NEW: ISO C 9x
4561 lrint NEW: ISO C 9x
4562 lrintf NEW: ISO C 9x
4563 lrintl NEW: ISO C 9x
4564 lround NEW: ISO C 9x
4565 lroundf NEW: ISO C 9x
4566 lroundl NEW: ISO C 9x
4567 lseek64 NEW: LFS
4568 makecontext NEW: Unix98
4569 mempcpy NEW: GNU ext.
4570 mmap64 NEW: LFS
4571 moncontrol REMOVED
4572 modify_ldt NEW: kernel
4573 nan NEW: ISO C 9x
4574 nanf NEW: ISO C 9x
4575 nanl NEW: ISO C 9x
4576 nearbyint NEW: ISO C 9x
4577 nearbyintf NEW: ISO C 9x
4578 nearbyintl NEW: ISO C 9x
4579 netname2host NEW: Secure RPC
4580 netname2user NEW: Secure RPC
4581 nexttoward NEW: ISO C 9x
4582 nexttowardf NEW: ISO C 9x
4583 nexttowardl NEW: ISO C 9x
4584 nftw NEW: Unix98
4585 nftw64 NEW: LFS
4586 open64 NEW: LFS
4587 passwd2des NEW: Secure RPC
4588 pow10 NEW: GNU ext.
4589 pow10f NEW: GNU ext.
4590 pow10l NEW: GNU ext.
4591 pread NEW: Unix98
4592 pread64 NEW: LFS
4593 printf_size NEW: GNU ext.
4594 printf_size_info NEW: GNU ext.
4595 profil_counter REMOVED
4596 pthread_mutexattr_getkind_np REPLACED
4597 pthread_mutexattr_setkind_np REPLACED
4598 ptsname NEW: Unix98 PTY
4599 ptsname_r NEW: Unix98 PTY
4600 putc_locked REMOVED
4601 putchar_locked REMOVED
4602 putgrent NEW: GNU ext.
4603 putmsg NEW: STREAMS
4604 putpmsg NEW: STREAMS
4605 pututxline NEW: Unix98
4606 pwrite NEW: Unix98
4607 pwrite64 NEW: LFS
4608 readdir64 NEW: LFS
4609 readdir64_r NEW: LFS
4610 remquo NEW: ISO C 9x
4611 remquof NEW: ISO C 9x
4612 remquol NEW: ISO C 9x
4613 round NEW: ISO C 9x
4614 roundf NEW: ISO C 9x
4615 roundl NEW: ISO C 9x
4616 rtime NEW: GNU ext.
4617 scalbln NEW: ISO C 9x
4618 scalblnf NEW: ISO C 9x
4619 scalblnl NEW: ISO C 9x
4620 scandir64 NEW: LFS
4621 sendfile NEW: kernel
4622 setcontext NEW: Unix98
4623 setrlimit64 NEW: LFS
4624 setutxent NEW: Unix98
4625 sighold NEW: Unix98
4626 sigignore NEW: Unix98
4627 sigqueue NEW: POSIX.1b
4628 sigrelse NEW: Unix98
4629 sigset NEW: POSIX.1b
4630 sigtimedwait NEW: POSIX.1b
4631 sigwaitinfo NEW: POSIX.1b
4632 sincos NEW: GNU ext.
4633 sincosf NEW: GNU ext.
4634 sincosl NEW: GNU ext.
4635 statfs64 NEW: LFS
4636 statvfs NEW: Unix98
4637 statvfs64 NEW: LFS
4638 strcasestr NEW: GNU ext.
4639 strtoimax NEW: ISO C 9x
4640 strtoumax NEW: ISO C 9x
4641 strverscmp NEW: GNU ext.
4642 svcauthdes_stats NEW: Secure RPC
4643 svcunix_create NEW: sunrpc etx
4644 svcunixfd_create NEW: sunrpc ext
4645 swapcontext NEW: Unix98
4646 tcgetsid NEW: Unix98 PTY
4647 tdestroy NEW: GNU ext.
4648 tgamma NEW: ISO C 9x
4649 tgammaf NEW: ISO C 9x
4650 tgammal NEW: ISO C 9x
4651 tmpfile64 NEW: LFS
4652 trunc NEW: ISO C 9x
4653 truncate64 NEW: LFS
4654 truncf NEW: ISO C 9x
4655 truncl NEW: ISO C 9x
4656 umount2 NEW: kernel
4657 unlockpt NEW: Unix98 PTY
4658 updwtmpx NEW: Unix98
4659 user2netname NEW: Secure RPC
4660 utmpxname NEW: Unix98
4661 versionsort NEW: GNU ext.
4662 versionsort64 NEW: GNU ext.
4663 waitid NEW: Unix98
4664 wcscasecmp NEW: GNU ext.
4665 wcsncasecmp NEW: GNU ext.
4666 wcsnlen NEW: GNU ext.
4667 wcstoimax NEW: ISO C 9x
4668 wcstoll NEW: ISO C 9x
4669 wcstoull NEW: ISO C 9x
4670 wcstoumax NEW: ISO C 9x
4671 wcswcs NEW: Unix98
4672 wordexp NEW: POSIX.2
4673 wordfree NEW: POSIX.2
4674 write_profiling REMOVED
4675 xdecrypt NEW: Secure RPC
4676 xdr_authdes_cred NEW: Secure RPC
4677 xdr_authdes_verf NEW: Secure RPC
4678 xdr_cryptkeyarg NEW: Secure RPC
4679 xdr_cryptkeyarg2 NEW: Secure RPC
4680 xdr_cryptkeyres NEW: Secure RPC
4681 xdr_getcredres NEW: Secure RPC
4682 xdr_key_netstarg NEW: Secure RPC
4683 xdr_key_netstres NEW: Secure RPC
4684 xdr_keybuf NEW: Secure RPC
4685 xdr_keystatus NEW: Secure RPC
4686 xdr_netnamestr NEW: Secure RPC
4687 xdr_sizeof NEW: Secure RPC
4688 xdr_unixcred NEW: sunrpc ext
4689 xencrypt NEW: Secure RPC
4690 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4691 \f
4692 Version 2.0.6
4693
4694 * more bug fixes
4695
4696 \f
4697 Version 2.0.5
4698
4699 * more bug fixes
4700
4701 * inet_ntoa is thread-safe
4702
4703 * updwtmp is moved from libutil to libc
4704
4705 * rewrite of cbrt function
4706
4707 * update of timezone data
4708 \f
4709 Version 2.0.4
4710
4711 * more bug fixes
4712 \f
4713 Version 2.0.3
4714
4715 * more bug fixes
4716 \f
4717 Version 2.0.2
4718
4719 * more bug fixes
4720
4721 * add atoll function
4722
4723 * fix complex problems in Berkeley DB code
4724
4725 * fix math functions
4726 \f
4727 Version 2.0.1
4728
4729 * fixed lots of header problems (especially Linux/GNU specific)
4730
4731 * dynamic loader preserves all registers
4732
4733 * Roland McGrath provided support for handling of auxiliary objects in
4734 the ELF dynamic loader.
4735
4736 * support for parallel builds is improved
4737 \f
4738 Version 2.0
4739
4740 * GNU extensions are no longer declared by default. To enable them you
4741 must define the macro `_GNU_SOURCE' in your program or compile with
4742 `-D_GNU_SOURCE'.
4743
4744 * The library has changed from using GNU ld symbol aliases to using weak
4745 symbols where available. The ELF object file format supports weak
4746 symbols; GNU ld also supports weak symbols in the a.out format. (There
4747 is also now support for other GNU ld extensions in ELF. Use the
4748 `--with-elf' option to configure to indicate you have ELF, and
4749 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4750 of many files which contained only symbol aliases, reducing the size of
4751 the source and the compiled library; many other files were renamed to
4752 less cryptic names previously occupied by the symbol alias files.
4753 There is a new header file <elf.h> for programs which operate on
4754 files in the ELF format.
4755
4756 * Converted to Autoconf version 2, so `configure' has more options.
4757 Run `configure --help' to see the details.
4758
4759 * The library can now be configured to build profiling, highly-optimized
4760 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4761 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4762 `configure' enable building these extra libraries. The shared library is
4763 built by default when using both ELF and GNU ld. When shared libraries
4764 are enabled, the new library `-ldl' is available for arbitrary run-time
4765 loading of shared objects; its interface is defined in <dlfcn.h>. The
4766 new header file <link.h> gives access to the internals of the run-time
4767 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4768 application of same name on other systems and it provides information
4769 about dynamically linked binaries.
4770
4771 * The C library now provides the run-time support code for profiling
4772 executables compiled with `-pg'. Programs can control the profiling code
4773 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4774 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4775 the support for this file format was contributed by David Mosberger-Tang.
4776
4777 * The math code has been replaced with a math library based on fdlibm from
4778 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4779 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4780 functions. The math functions now reside in a separate library, so
4781 programs using them will need to use `-lm' their linking commands.
4782
4783 * John C. Bowman contributed optimized ix87 assembler inline functions.
4784
4785 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4786 mechanism similar to that found in Solaris 2. This is now used for the
4787 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4788 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4789 file controls what services are used for each individual database. This
4790 works by loading shared libraries with names specified in `nsswitch.conf',
4791 so service modules can be changed or added at any time without even
4792 relinking any program. Currently there are the file, db, and NIS based
4793 NSS services available.
4794
4795 * The new functions `strtoq' and `strtouq' parse integer values from
4796 strings, like `strtol' and `strtoul', but they return `long long int' and
4797 `unsigned long long int' values, respectively (64-bit quantities).
4798
4799 * The new functions `strtof' and `strtold' parse floating-point values from
4800 strings, like `strtod', but they return `float' and `long double' values,
4801 respectively (on some machines `double' and `long double' are the same).
4802
4803 * Ulrich Drepper has contributed new implementations of the floating-point
4804 printing and reading code used in the `printf' family of functions and
4805 `strtod', `strtof', and `strtold'. These new functions are perfectly
4806 accurate, and much faster than the old ones.
4807
4808 * The implementation of the POSIX locale model was completely rewritten by
4809 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4810 to compile the POSIX locale definition.
4811
4812 * The former dummy implementations of the strcoll and strxfrm function are
4813 now replaced by fully functional code contributed by Ulrich Drepper. The
4814 collation information comes from the POSIX locale definitions.
4815
4816 * The new header <langinfo.h> defines an interface for accessing
4817 various locale-dependent data (using the locale chosen with `setlocale').
4818
4819 * Ulrich Drepper has contributed a new suite of functions for operation on
4820 wide-character and multibyte-character strings, in <wchar.h>;
4821 and classification and case conversion of wide characters, in <wctype.h>.
4822 These new functions are conforming to the ISO C, Amendement 1 specification.
4823
4824 * There is now a second implementation of the standard I/O library available.
4825 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4826 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4827
4828 * You can now use positional parameter specifications in format strings
4829 for the `printf' and `scanf' families of functions. For example,
4830 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4831 ``Number 6, Mr Jones''. This is mainly useful when providing different
4832 format strings for different languages, whose grammars may dictate
4833 different orderings of the values being printed. To support this
4834 feature, the interface for `register_printf_handler' has changed; see
4835 the header file <printf.h> for details.
4836
4837 * The `printf' and `scanf' families of functions now understand a new
4838 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4839 to group numbers as indicated by the locale; for `scanf' and friends, this
4840 says to accept as valid only a number with all the proper grouping
4841 separators in the right places. In the default "C" locale, numbers are
4842 not grouped; but locales for specific countries will define the usual
4843 conventions (i.e. separate thousands with `,' in the US locale).
4844
4845 * The pgrp functions have been regularized, slightly incompatibly but much
4846 less confusingly. The core functions are now `getpgid' and `setpgid',
4847 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4848 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4849 provided for compatibility. There is no longer an incompatible `getpgrp'
4850 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4851
4852 * The new header file <fts.h> and suite of functions simplify programs that
4853 operate on directory trees. This code comes from 4.4 BSD.
4854
4855 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4856 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4857 NSS scheme used in glibc.
4858
4859 * There is a new malloc debugging hook `__memalign_hook'.
4860
4861 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4862 `unsigned int' in <sys/types.h>. These are for compatibility only and
4863 their use is discouraged.
4864
4865 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4866 done differently, so that it works even without GNU ld.
4867
4868 * New function `euidaccess' checks allowed access to a file like `access',
4869 but using the effective IDs instead of the real IDs.
4870
4871 * The time zone data files have been updated for the latest and greatest
4872 local time conventions of the countries of the world.
4873
4874 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4875 see <dirent.h>.
4876
4877 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4878 for formatting floating-point numbers. They are provided only for
4879 compatibility; new programs should use `sprintf' instead. There are
4880 also equivalent function for the `long double' floating-point type and
4881 all functions also exist in a reentrant form.
4882
4883 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4884 functions for maintaining the login-record files (primarily of use to
4885 system programs such as `login'), and convenient functions for
4886 allocating and initializing a pseudo-terminal (pty) device.
4887
4888 * Ulrich Drepper has contributed new support for System V style
4889 shared memory and IPC on systems that support it.
4890
4891 * Ulrich Drepper has contributed several miscellaneous new functions found
4892 in System V: The `hsearch' family of functions provide an effective
4893 implementation of hash tables; `a64l' and `l64a' provide a very simple
4894 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4895 number generator.
4896
4897 * Ulrich Drepper has contributed new reentrant counterparts for the
4898 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4899
4900 * Ulrich Drepper has contributed new, highly-optimized versions of several
4901 string functions for the i486/Pentium family of processors.
4902
4903 * Ulrich Drepper has updated the Linux-specific code, based largely
4904 on work done in Hongjiu Lu's version of GNU libc for Linux.
4905 The GNU library now supports Linux versions 2.0.10 and later,
4906 using the ELF object file format (i[3456]86-*-linux).
4907
4908 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4909
4910 * David Mosberger-Tang and Richard Henderson have ported the C library
4911 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4912 dynamic linking support for ELF/Alpha.
4913
4914 * Richard Henderson contributed several Alpha optimized assembler function
4915 for arithmetic and string handling.
4916
4917 * Ulrich Drepper has contributed a new set of message catalog functions to
4918 support multiple languages using the <libintl.h> interface, for use with
4919 his new package GNU gettext. Translation volunteers have contributed
4920 catalogs of the library's messages in Spanish, German, and Korean.
4921
4922 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4923 program and the `catgets' function for reading the catalog files it
4924 creates. (The <libintl.h> interface is preferred; we include the
4925 <nl_types.h> interface using `catgets' only for source compatibility with
4926 programs already written to use it.)
4927
4928 * New header file <values.h> gives SVID-compatible names for <limits.h>
4929 constants.
4930
4931 * Various new macros, declarations, and small header files for compatibility
4932 with 4.4 BSD.
4933
4934 * New function `group_member' is a convenient way to check if a process has
4935 a given effective group ID.
4936
4937 * When using GCC 2.7 and later, the socket functions are now declared in a
4938 special way so that passing an argument of type `struct sockaddr_in *',
4939 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4940 `struct sockaddr *' type, does not generate a type-clash warning.
4941
4942 * New function `error' declared in header file <error.h> is a convenient
4943 function for printing error messages and optionally exiting; this is the
4944 canonical function used in GNU programs. The new functions `err', `warn',
4945 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4946 doing the same thing.
4947
4948 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4949 POSIX.2 `glob' function to do ~ and {...} expansion.
4950
4951 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4952 `clearenv' which is used in POSIX.9 is also available.
4953
4954 * New function `getsid' returns session ID number on systems that support it.
4955
4956 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4957 files <db.h> and <mpool.h> provide a rich set of functions for several
4958 types of simple databases stored in memory and in files, and <ndbm.h> is
4959 an old `ndbm'-compatible interface using the `db' functions. Link with
4960 `-ldb' to get these functions.
4961
4962 * New macro `strdupa' copies a string like `strdup', but uses local stack
4963 space from `alloca' instead of dynamic heap space from `malloc'.
4964
4965 * New function `strnlen' is like `strlen' but searches only a given maximum
4966 number of characters for the null terminator. `stpncpy', `strndup' and
4967 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4968 function.
4969
4970 * New function `statfs' in header <sys/statfs.h>.
4971
4972 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4973 provide convenient functions for operating on blocks of null-terminated
4974 strings.
4975
4976 * A new suite of functions in <utmp.h> handle all the details of reading
4977 and writing the utmp file.
4978
4979 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4980 Thorsten Kukuk.
4981
4982 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4983 completely POSIX compliant and also implemented the extended functionality
4984 to handle alternate digit representation and alternate era date formats.
4985
4986 * Ulrich Drepper provided an implementation of the `strptime' function
4987 defined in XPG4.2 which transforms a string into a `struct tm' value.
4988
4989 * Paul Eggert provided the tzselect shell script as part of the timezone
4990 code. The shell script makes it easy to select the correct timezone
4991 specification.
4992
4993 * The implementation of the malloc family of functions is completely replaced
4994 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4995 The implementation uses the mmap function (if available) and it is
4996 optimized for the use in multi threaded programs.
4997
4998 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4999 functions. This new functionality is usable by specifying a special
5000 salt string and it is compatible with implementation on *BSD systems.
5001
5002 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
5003 `getsubopt' to handle second level command line options, `bsd_signal'
5004 to access BSD style `signal' functionality, the obsolete `regexp' style
5005 expression matcher.
5006
5007 * the `lchown' function is available on system which support this
5008 functionality.
5009
5010 * The implementation of the shadow password handling function was contributed
5011 by Ulrich Drepper.
5012
5013 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
5014
5015 * POSIX.1g support was added. The <sys/select.h> header is available,
5016 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
5017 implementation of `getaddrinfo'.
5018 \f
5019 Version 1.09
5020
5021 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
5022
5023 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
5024 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
5025
5026 * The new function `daemon' from 4.4 BSD is useful for server programs that
5027 want to put themselves in the background.
5028
5029 * Joel Sherrill has contributed support for several standalone boards that
5030 run without an operating system.
5031
5032 * `printf', `scanf' and friends now accept a `q' type modifier for long
5033 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
5034
5035 * All of the code taken from BSD (notably most of the math and networking
5036 routines) has been updated from the BSD 4.4-Lite release.
5037
5038 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
5039
5040 * The new functions `getdomainname' and `setdomainname' fetch or change the
5041 YP/NIS domain name. These are system calls which exist on systems which
5042 have YP (aka NIS).
5043
5044 * The time zone data files have been updated for the latest international
5045 conventions.
5046
5047 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
5048 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
5049 \f
5050 Version 1.08
5051
5052 * The C library now includes support for Sun RPC, from Sun's free
5053 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
5054 are included. (There is still no support for YP.)
5055
5056 * Tom Quinn has contributed a port of the C library to SGI machines running
5057 Irix 4 (mips-sgi-irix4).
5058
5059 * The new `lockf' function is a simplified interface to the locking
5060 facilities of `fcntl', included for compatibility.
5061
5062 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
5063
5064 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
5065
5066 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
5067 compatibility.
5068
5069 * The error message printed by `assert' for a failed assertion now includes
5070 the name of the program (if using GNU ld) and the name of the calling
5071 function (with versions of GCC that support this).
5072
5073 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
5074
5075 * The library now includes the <sys/mman.h> header file and memory
5076 management functions `mmap', `munmap', `mprotect', `msync', and
5077 `madvise', on systems that support those facilities.
5078
5079 * The interface for `mcheck' has changed slightly: the function called to
5080 abort the program when an allocation inconsistency is detected now takes
5081 an argument that indicates the type of failure. The new function
5082 `mprobe' lets you request a consistency check for a particular block at
5083 any time (checks are normally done only when you call `free' or `realloc'
5084 on a block).
5085
5086 * It is now possible to easily cross-compile the C library, building on one
5087 system a library to run on another machine and/or operating system. All
5088 you need to do is set the variable `HOST_CC' in `configparms' to the
5089 native compiler for programs to run on the machine you are building on (a
5090 few generator programs are used on Unix systems); set `CC' to the
5091 cross-compiler.
5092
5093 * The new function `fexecve' (only implemented on the GNU system) executes
5094 a program file given a file descriptor already open on the file.
5095 \f
5096 Version 1.07
5097
5098 * Brendan Kehoe has contributed most of a port to the DEC Alpha
5099 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
5100
5101 * You can set the variable `libprefix' in `configparms' to specify a prefix
5102 to be prepended to installed library files; this makes it easy to install
5103 the GNU C library to be linked as `-lgnuc' or whatever.
5104
5105 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
5106 copies a limited number of characters from a string, and returns the
5107 address of the last character written.
5108
5109 * You no longer need to check for whether the installed `stddef.h' is
5110 compatible with the GNU C library. configure now checks for you.
5111
5112 * You can now define a per-stream `fileno' function to convert the
5113 stream's cookie into an integral file descriptor.
5114
5115 * ``malloc (0)'' no longer returns a null pointer. Instead, it
5116 allocates zero bytes of storage, and returns a unique pointer which
5117 you can pass to `realloc' or `free'. The behavior is undefined if
5118 you dereference this pointer.
5119
5120 * The C library now runs on Sony NEWS m68k machines running either
5121 NewsOS 3 or NewsOS 4.
5122
5123 * The new `syscall' function is a system-dependent primitive function
5124 for invoking system calls. It has the canonical behavior on Unix
5125 systems, including unreliable return values for some calls (such as
5126 `pipe', `fork' and `getppid').
5127
5128 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
5129 to `EAGAIN', which is the preferred name. On systems whose kernels
5130 use two distinct codes, the C library now translates EWOULDBLOCK to
5131 EAGAIN in every system call function.
5132 \f
5133 Version 1.06
5134
5135 * The GNU C Library Reference Manual is now distributed with the library.
5136 `make dvi' will produce a DVI file of the printed manual.
5137 `make info' will produce Info files that you can read on line using C-h i
5138 in Emacs or the `info' program.
5139 Please send comments on the manual to bug-glibc-manual@gnu.org.
5140
5141 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
5142
5143 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
5144
5145 * Jason Merrill has contributed a port to the Sequent Symmetry running
5146 Dynix version 3 (i386-sequent-dynix).
5147
5148 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
5149 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
5150
5151 * New function `memory_warnings' lets you arrange to get warnings when
5152 malloc is running out of memory to allocate, like Emacs gives you.
5153
5154 * The C library now contains the relocating allocator used in Emacs 19 for
5155 its editing buffers. This allocator (ralloc) minimizes allocation
5156 overhead and fragmentation by moving allocated regions around whenever it
5157 needs to. You always refer to a ralloc'd region with a "handle" (a
5158 pointer to a pointer--an object of type `void **').
5159
5160 * There is a new `printf' format: `%m' gives you the string corresponding
5161 to the error code in `errno'.
5162
5163 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
5164 or `%[' conversion, but instead of filling in a fixed-sized buffer you
5165 pass, the `a' modifier says to fill in a `char **' you pass with a
5166 malloc'd string.
5167
5168 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
5169 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
5170 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
5171
5172 * `mkstemp' is a traditional Unix function to atomically create and open a
5173 uniquely-named temporary file.
5174 \f
5175 Version 1.05
5176
5177 * The standard location for the file that says what the local timezone is
5178 has changed again. It is now `/usr/local/etc/localtime' (or more
5179 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
5180
5181 * The distribution no longer contains any files with names longer than 14
5182 characters.
5183
5184 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
5185 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
5186
5187 * New functions `ttyslot' and `syslog' from 4.4 BSD.
5188 \f
5189 Version 1.04
5190
5191 * The configuration process has changed quite a bit. The `configure'
5192 script is now used just like the configuration scripts for other GNU
5193 packages. The `sysdeps' directory hierarchy is much rearranged.
5194 The file `INSTALL' explains the new scheme in detail.
5195
5196 * The header files no longer need to be processed into ANSI C and
5197 traditional C versions. There is just one set of files to install, and
5198 it will work with ANSI or old C compilers (including `gcc -traditional').
5199
5200 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
5201 MIPS DECStation running Ultrix 4.
5202
5203 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
5204 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
5205 made itself into a shared library.
5206
5207 * Yet further improved support for the i386, running 4.3 BSD-like systems
5208 (such as Mach 3 with the Unix single-server), or System V.
5209
5210 * New function `strncasecmp' to do case-insensitive string comparison
5211 with limited length.
5212
5213 * New function `strsep' is a reentrant alternative to `strtok'.
5214
5215 * New functions `scandir' and `alphasort' for searching directories.
5216
5217 * New function `setenv' is a better interface to `putenv'.
5218
5219 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
5220 function for traversing a directory tree.
5221
5222 * The GNU obstack package is now also part of the C library.
5223 The new function `open_obstack_stream' creates a stdio stream that
5224 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
5225 formatted output directly to an obstack.
5226
5227 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
5228 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
5229
5230 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
5231
5232 * The new functions `strfry' and `memfrob' do mysterious and wonderful
5233 things to your strings.
5234
5235 * There are some new test programs: test-fseek, testmb, and testrand.
5236
5237 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
5238 These ports are not finished, but are a good starting place for really
5239 supporting those systems.
5240
5241 * `/etc/localtime' is now the standard location for the file that says what
5242 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
5243 This follows the general principle that `/etc' is the place for all local
5244 configuration files.
5245
5246 * The C library header files now use `extern "C"' when used by the C++
5247 compiler, so the C library should now work with C++ code.
5248
5249 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
5250 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
5251 in <strings.h>.)
5252
5253 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
5254 function which uses the `merge sort' algorithm, and is said to be
5255 significantly faster than the old GNU `qsort' function. Merge sort is
5256 now the standard `qsort' function. The new algorithm can require a lot
5257 of temporary storage; so, the old sorting function is called when the
5258 required storage is not available.
5259
5260 * The C library now includes Michael Glad's Ultra Fast Crypt, which
5261 provides the Unix `crypt' function, plus some other entry points.
5262
5263 * The code and header files taken from 4.4 BSD have been updated with the
5264 latest files released from Berkeley.
5265 \f
5266 ----------------------------------------------------------------------
5267 Copyright information:
5268
5269 Copyright (C) 1992-2018 Free Software Foundation, Inc.
5270
5271 Permission is granted to anyone to make or distribute verbatim copies
5272 of this document as received, in any medium, provided that the
5273 copyright notice and this permission notice are preserved,
5274 thus giving the recipient permission to redistribute in turn.
5275
5276 Permission is granted to distribute modified versions
5277 of this document, or of portions of it,
5278 under the above conditions, provided also that they
5279 carry prominent notices stating who last changed them.
5280 \f
5281 Local variables:
5282 version-control: never
5283 fill-column: 76
5284 End: