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