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