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