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