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