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