1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 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 * The GNU C Library can now be compiled with support for building static
13 PIE executables (See --enable-static-pie in INSTALL). These static PIE
14 exectuables are like static executables but can be loaded at any address
15 and provide additional security hardening benefits at the cost of some
16 memory and performance. When the library is built with --enable-static-pie
17 the resulting libc.a is usable with GCC 8 and above to create static PIE
18 executables using the GCC option '-static-pie'. This feature is currently
19 supported on i386, x86_64, x32 and aarch64.
21 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
22 sinf and tan with FMA, contributed by Arjan van de Ven and H.J. Lu
25 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
27 * Optimized generic expf, exp2f, logf, log2f, powf and sinf.
29 * In order to support faster and safer process termination the malloc API
30 family of functions will no longer print a failure address and stack
31 backtrace after detecting heap corruption. The goal is to minimize the
32 amount of work done after corruption is detected and to avoid potential
33 security issues in continued process execution. Reducing shutdown time
34 leads to lower overall process restart latency, so there is benefit both
35 from a security and performance perspective.
37 * The abort function terminates the process immediately, without flushing
38 stdio streams. Previous glibc versions used to flush streams, resulting
39 in deadlocks and further data corruption. This change also affects
40 process aborts as the result of assertion failures.
42 * On platforms where long double has the IEEE binary128 format (aarch64,
43 alpha, mips64, s390 and sparc), the math library now implements _Float128
44 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
45 are the same interfaces added in version 2.26 for some platforms where
46 this format is supported but is not the format of long double.
48 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
49 mips64, powerpc64le, s390, sparc and x86_64), the math library now
50 implements interfaces for that type, as defined by ISO/IEC TS
51 18661-3:2015. These are corresponding interfaces to those supported for
54 * The math library now implements interfaces for the _Float32, _Float64 and
55 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
56 corresponding interfaces to those supported for _Float128.
58 * glibc now implements the memfd_create and mlock2 functions on Linux.
60 * Support for memory protection keys was added. The <sys/mman.h> header now
61 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
64 * The copy_file_range function was added.
66 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
68 * The ldconfig utility now processes `include' directives using the C/POSIX
69 collation ordering. Previous glibc versions used locale-specific
70 ordering, the change might break systems that relied on that.
72 Deprecated and removed features, and other changes affecting compatibility:
74 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
75 defined by <sys/ptrace.h>.
77 * libm no longer supports SVID error handling (calling a user-provided
78 matherr function on error) or the _LIB_VERSION variable to control error
79 handling. (SVID error handling and the _LIB_VERSION variable still work
80 for binaries linked against older versions of the GNU C Library.) The
81 libieee.a library is no longer provided. math.h no longer defines struct
82 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
83 TLOSS, PLOSS and HUGE.
85 * The libm functions pow10, pow10f and pow10l are no longer supported for
86 new programs. Programs should use the standard names exp10, exp10f and
87 exp10l for these functions instead.
89 * The mcontext_t type is no longer the same as struct sigcontext. On
90 platforms where it was previously the same, this changes the C++ name
91 mangling for interfaces involving this type.
93 * The add-ons mechanism for building additional packages at the same time as
94 glibc has been removed. The --enable-add-ons configure option is now
97 * The --without-fp configure option is now ignored. Whether hardware
98 floating-point instructions are used is now configured based on whether
99 the compiler used at configure time (without any options implied by a
100 --with-cpu= configure option) uses such instructions.
102 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
103 the specified string can be parsed as a domain name.
105 * In the malloc_info output, the <heap> element may contain another <aspace>
106 element, "subheaps", which contains the number of sub-heaps.
108 * The libresolv function p_secstodate is no longer supported for new
111 * The tilepro-*-linux-gnu configuration is no longer supported.
113 Changes to build and runtime requirements:
115 * bison version 2.7 or later is required to generate code in the 'intl'
118 Security related changes:
120 CVE-2009-5064: The ldd script would sometimes run the program under
121 examination directly, without preventing code execution through the
122 dynamic linker. (The glibc project disputes that this is a security
123 vulnerability; only trusted binaries must be examined using the ldd
126 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
127 suffered from a one-byte overflow during ~ operator processing (either
128 on the stack or the heap, depending on the length of the user name).
129 Reported by Tim Rühsen.
131 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
132 would sometimes fail to free memory allocated during ~ operator
133 processing, leading to a memory leak and, potentially, to a denial
136 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
137 without GLOB_NOESCAPE, could write past the end of a buffer while
138 unescaping user names. Reported by Tim Rühsen.
140 CVE-2017-17426: The malloc function, when called with an object size near
141 the value SIZE_MAX, would return a pointer to a buffer which is too small,
142 instead of NULL. This was a regression introduced with the new malloc
143 thread cache in glibc 2.26. Reported by Iain Buclaw.
145 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
146 to the allocation of too much memory. (This is not a security bug per se,
147 it is mentioned here only because of the CVE assignment.) Reported by
150 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
151 of the number of search path components. (This is not a security
152 vulnerability per se because no trust boundary is crossed if the fix for
153 CVE-2017-1000366 has been applied, but it is mentioned here only because
154 of the CVE assignment.) Reported by Qualys.
156 The following bugs are resolved with this release:
158 [The release manager will add the list generated by
159 scripts/list-fixed-bugs.py just before the release.]
166 * A per-thread cache has been added to malloc. Access to the cache requires
167 no locks and therefore significantly accelerates the fast path to allocate
168 and free small amounts of memory. Refilling an empty cache requires locking
169 the underlying arena. Performance measurements show significant gains in a
170 wide variety of user workloads. Workloads were captured using a special
171 instrumented malloc and analyzed with a malloc simulator. Contributed by
172 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
174 * Unicode 10.0.0 Support: Character encoding, character type info, and
175 transliteration tables are all updated to Unicode 10.0.0, using
176 generator scripts contributed by Mike FABIAN (Red Hat).
177 These updates cause user visible changes, especially the changes in
178 wcwidth for many emoji characters cause problems when emoji sequences
179 are rendered with pango, see for example:
180 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
182 * Collation of Hungarian has been overhauled and is now consistent with "The
183 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
186 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
188 - The GNU C Library will now detect when /etc/resolv.conf has been
189 modified and reload the changed configuration. The new resolver option
190 “no-reload” (RES_NORELOAD) disables this behavior.
192 - The GNU C Library now supports an arbitrary number of search domains
193 (configured using the “search” directive in /etc/resolv.conf);
194 previously, there was a hard limit of six domains. For backward
195 compatibility, applications that directly modify the ‘_res’ global
196 object are still limited to six search domains.
198 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
199 Library will now randomly pick a name server from the configuration as a
200 starting point. (Previously, the second name server was always used.)
202 * The tunables feature is now enabled by default. This allows users to tweak
203 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
205 * New function reallocarray, which resizes an allocated block (like realloc)
206 to the product of two sizes, with a guaranteed clean failure upon integer
207 overflow in the multiplication. Originally from OpenBSD, contributed by
208 Dennis Wölfing and Rüdiger Sonderfeld.
210 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
211 These are extended versions of preadv and pwritev, respectively, taking an
212 additional flags argument. The set of supported flags depends on the
213 running kernel; full support currently requires kernel 4.7 or later.
215 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
216 create a new session ID for the spawned process. This feature is
217 scheduled to be added to the next major revision of POSIX; for the time
218 being, it is available under _GNU_SOURCE.
220 * errno.h is now safe to use from C-preprocessed assembly language on all
221 supported operating systems. In this context, it will only define the
222 Exxxx constants, as preprocessor macros expanding to integer literals.
224 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
225 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
226 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
227 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
229 To compile programs that use this feature, the compiler must support
230 128-bit floating point with the type name _Float128 (as defined by TS
231 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
232 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
233 must be defined to make the new interfaces visible.
235 The new functions and macros correspond to those present for other
236 floating-point types (except for a few obsolescent interfaces not
237 supported for the new type), with F128 or f128 suffixes; for example,
238 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
239 printf or scanf formats for the new type; the strfromf128 and strtof128
240 interfaces should be used instead.
242 Deprecated and removed features, and other changes affecting compatibility:
244 * The synchronization that pthread_spin_unlock performs has been changed to
245 now be equivalent to a C11 atomic store with release memory order to the
246 spin lock's memory location. Previously, several (but not all)
247 architectures used stronger synchronization (e.g., containing what is
248 often called a full barrier). This change can improve performance, but
249 may affect odd fringe uses of spin locks that depend on the previous
250 behavior (e.g., using spin locks as atomic variables to try to implement
251 Dekker's mutual exclusion algorithm).
253 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
256 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
257 will only be built and installed when the GNU C Library is configured with
258 --enable-obsolete-rpc. This allows alternative RPC implementations, such
259 as TIRPC or rpcsvc-proto, to be used.
261 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
262 libnss_compat, are deprecated, and will not be built or installed by
265 The NIS(+) support library, libnsl, is also deprecated. By default, a
266 compatibility shared library will be built and installed, but not headers
267 or development libraries. Only a few NIS-related programs require this
268 library. (In particular, the GNU C Library has never required programs
269 that use 'gethostbyname' to be linked with libnsl.)
271 Replacement implementations based on TIRPC, which additionally support
272 IPv6, are available from <https://github.com/thkukuk/>. The configure
273 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
274 name service modules, to be built and installed.
276 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
277 support is enabled, the configured recursive resolver must support EDNS.
278 (Responding to EDNS-enabled queries with responses which are not
279 EDNS-enabled is fine, but FORMERR responses are not.)
281 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
282 servers have not supported this opcode for a long time.
284 * The _res_opcodes variable has been removed from libresolv. It had been
285 exported by accident.
287 * <string.h> no longer includes inline versions of any string functions,
288 as this kind of optimization is better done by the compiler. The macros
289 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
291 * The nonstandard header <xlocale.h> has been removed. Most programs should
292 use <locale.h> instead. If you have a specific need for the definition of
293 locale_t with no other declarations, please contact
294 libc-alpha@sourceware.org and explain.
296 * The obsolete header <sys/ultrasound.h> has been removed.
298 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
300 * The obsolete function cfree has been removed. Applications should use
303 * The stack_t type no longer has the name struct sigaltstack. This changes
304 the C++ name mangling for interfaces involving this type.
306 * The ucontext_t type no longer has the name struct ucontext. This changes
307 the C++ name mangling for interfaces involving this type.
309 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
310 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
311 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
312 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
313 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
314 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
315 fpregset_t type no longer has the name struct fpu. This changes the C++
316 name mangling for interfaces involving those types.
318 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
319 synced with the kernel:
321 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
322 are not supported on this architecture and have been removed.
324 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
325 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
326 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
328 Programs that assume the GET/SETREGS ptrace requests are universally
329 available will now fail to build, instead of malfunctioning at runtime.
331 Changes to build and runtime requirements:
333 * Linux kernel 3.2 or later is required at runtime, on all architectures
334 supported by that kernel. (This is a change from version 2.25 only for
337 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
339 * On most architectures, GCC 4.9 or later is required to build the GNU C
340 Library. On powerpc64le, GCC 6.2 or later is required.
342 Older GCC versions and non-GNU compilers are still supported when
343 compiling programs that use the GNU C Library. (We do not know exactly
344 how old, and some GNU extensions to C may be _de facto_ required. If you
345 are interested in helping us make this statement less vague, please
346 contact libc-alpha@sourceware.org.)
348 Security related changes:
350 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
351 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
353 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
354 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
356 * Avoid printing a backtrace from the __stack_chk_fail function since it is
357 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
360 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
361 fixed (CVE-2017-12133).
363 The following bugs are resolved with this release:
365 [984] network: Respond to changed resolv.conf in gethostbyname
366 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
367 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
369 [12189] libc: __stack_chk_fail should not attempt a backtrace
371 [14096] time: Race condition on timezone/tst-timezone.out
372 [14172] localedata: az_IR: new locale
373 [14995] build: glibc fails to build if gold is the default linker, even if
375 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
376 [16637] network: inet_pton function is accepting IPv6 with bad format
377 [16640] string: string/strtok.c: undefined behaviour inconsistent between
378 x86 and other generic code
379 [16875] localedata: ko_KR: fix lang_name
380 [17225] localedata: ar_SY: localized month names for May and June are
382 [17297] localedata: da_DK: wrong date_fmt string
383 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
384 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
385 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
386 [19066] localedata: ar_SA abbreviated day and month names are in English
387 [19569] network: resolv: Support an arbitrary number of search domains
388 [19570] network: Implement random DNS server selection in the stub
390 [19838] locale: localedef fails on PA-RISC
391 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
392 order of 0D36 and 0D37
393 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
395 [20098] libc: FAIL: debug/backtrace-tst on hppa
396 [20257] network: sunrpc: clntudp_call does not enforce timeout when
398 [20275] localedata: locale day/abday/mon/abmon should not have trailing
400 [20313] localedata: Update locale data to Unicode 9.0
401 [20424] manual: Document how to provide a malloc replacement
402 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
404 [20686] locale: Add el_GR@euro to SUPPORTED.
405 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
406 failures consistently
407 [21015] dynamic-link: Document and fix --enable-bind-now
408 [21016] nptl: pthread_cond support is broken on hppa
409 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
410 frame-pointer on i386
411 [21049] libc: segfault in longjmp_chk() due to clobbered processor
413 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
414 [21088] libc: Build fails with --enable-static-nss
415 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
416 generic c code is used
417 [21109] libc: Tunables broken on big-endian
418 [21112] math: powf has large ulp errors with base close to 1 and exponent
420 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
422 [21120] malloc: glibc malloc is incompatible with GCC 7
423 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
425 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
426 (-0.0) when linking with -lieee
427 [21171] math: log10, log2 and lgamma return incorrect results
428 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
429 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
430 [21207] localedata: ce_RU: update weekdays from CLDR
431 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
432 [21217] localedata: Update months from CLDR-31
433 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
434 [21243] libc: support_delete_temp_file should issue warning for failed
436 [21244] libc: support resolv_test_start() socket fd close should be
438 [21253] libc: localedef randomly segfaults when using -fstack-check due to
439 new posix_spawn implementation
440 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
441 leads to lower CPU frequency
442 [21259] libc: [alpha] termios.h missing IXANY for POSIX
443 [21261] libc: [sparc64] bits/setjmp.h namespace
444 [21267] network: [mips] bits/socket.h IOC* namespace
445 [21268] libc: [alpha] termios.h NL2, NL3 namespace
446 [21270] libc: mmap64 silently truncates large offset values
447 [21275] libc: posix_spawn always crashes on ia64 now
448 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
449 [21280] math: [powerpc] logbl for POWER7 return incorrect results
450 [21289] libc: Incorrect declaration for 32-bit platforms with
451 _FILE_OFFSET_BITS=64 causes build error
452 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
453 not support gethostbyname4_r
454 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
456 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
458 [21340] libc: Support POSIX_SPAWN_SETSID
459 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
460 [21359] network: ns_name_pack needs additional byte in destination buffer
461 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
462 against fragmentation attacks (CVE-2017-12132)
463 [21369] network: resolv: Remove EDNS fallback
464 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
466 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
467 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
468 [21393] stdio: Missing dup3 error check in freopen, freopen64
469 [21396] libc: Use AVX2 memcpy/memset on Skylake server
470 [21399] localedata: Bad description for U00EC in
471 localedata/charmaps/CP1254
472 [21411] malloc: realloc documentation error
473 [21426] network: sys/socket.h uio.h namespace
474 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
475 [21445] libc: signal.h bsd_signal namespace
476 [21455] network: Network headers stdint.h namespace
477 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
479 [21475] network: resolv: Overlong search path is truncated mid-label
480 [21511] libc: sigstack namespace
481 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
482 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
483 bits/syscall.h: No such file or directory
484 [21517] libc: struct sigaltstack namespace
485 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
486 [21533] localedata: Update locale data to Unicode 10.0
488 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
489 junk at end of line, first unrecognized character is `@'
490 [21538] libc: SIG_HOLD missing for XPG4
491 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
492 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
493 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
494 [21543] libc: sigevent namespace
495 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
496 size for general purpose registers in mcontext_t structure
497 [21550] libc: sigwait namespace
498 [21552] libc: XPG4 bsd_signal namespace
499 [21554] libc: sigpause namespace
500 [21560] libc: sys/wait.h signal.h namespace
501 [21561] libc: waitid namespace
502 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
503 '/usr/include/stdlib.h': Permission denied
504 [21575] libc: sys/wait.h missing struct rusage definition
505 [21584] libc: sigaltstack etc namespace
506 [21597] libc: siginterrupt namespace
507 [21607] math: hppa: FAIL: math/test-tgmath
508 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
510 [21622] libc: [tile] missing SA_* for POSIX.1:2008
511 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
512 stack and heap (CVE-2017-1000366)
513 [21625] libc: wait3 namespace
514 [21654] nss: Incorrect pointer alignment in NSS group merge result
516 [21657] network: Parse interface zone id for node-local multicast
517 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
519 [21666] libc: .symver is used on common symbol
520 [21668] network: resolv: res_init cross-thread broadcast introduces race
522 [21687] math: tgmath.h totalorder, totalordermag return type
523 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
525 [21696] libc: Incorrect assumption of of __cpu_mask in
526 posix/sched_cpucount.c
527 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
528 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
530 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
531 leading to relocation crash
532 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
533 `resolv_conf_matches (resp, conf)' failed.
534 [21710] localedata: Added Samoan language locale for Samoa
535 [21711] localedata: Pashto yesstr/nostr locale are missing
536 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
537 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
538 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
539 [21723] localedata: yesstr/nostr missing for Chinese language locale
540 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
542 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
544 [21728] localedata: New Locale for Tongan language
545 [21729] localedata: incorrect LC_NAME fields for hi_IN
546 [21733] localedata: yesstr and nostr are missing for zh_HK
547 [21734] localedata: Missing yesstr and nostr are for kw_GB
548 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
549 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
550 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
551 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
553 [21744] libc: Tests failing on --enable-tunables --enable-stack-
555 [21749] localedata: Wrong abbreviated day name (“abday”) for
557 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
558 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
559 [21759] localedata: missing yesstr and nostr for Tigrinya
560 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
561 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
562 [21767] localedata: Missing Bislama locales
563 [21768] localedata: Missing yesstr and nostr for aa_ET
564 [21770] localedata: Missing Field in li_NL
565 [21778] nptl: Robust mutex may deadlock
566 [21779] libc: MicroBlaze segfaults when loading libpthread
567 [21783] localedata: Fix int_select international_call_prefixes
568 [21784] localedata: Inconsistency in country_isbn
569 [21788] localedata: Missing Country Postal Abbreviations
570 [21794] localedata: Added-country_isbn-for-Italy
571 [21795] localedata: Add/Fix country_isbn for France
572 [21796] localedata: Added country_isbn for Republic of Korea
573 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
574 [21799] localedata: Added int_select international_call_prefixes
575 [21801] localedata: Added int_select international_call_prefixes
576 [21804] nptl: Double semicolon in thread-shared-types.h
577 [21807] localedata: LC_ADDRESS fix for pap_CW
578 [21808] localedata: Fix LC_ADDRESS for pap_AW
579 [21821] localedata: Added country_name in mai_IN
580 [21822] localedata: Fix LC_TIME for mai_IN
581 [21823] localedata: missing yesstr, nostr for sa_IN
582 [21825] localedata: Fix name_mrs for mag_IN
583 [21828] localedata: 2.26 changelog should mention user visible changes
585 [21835] localedata: Added Maithili language locale for Nepal
586 [21838] localedata: Removed redundant data for the_NP
587 [21839] localedata: Fix LC_MONETARY for ta_LK
588 [21844] localedata: Fix Latin characters and Months Sequence.
589 [21848] localedata: Fix mai_NP Title Name
594 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
595 24731-2:2010, is supported to enable declarations of functions from that
596 TR. Note that not all functions from that TR are supported by the GNU C
599 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
600 18661-1:2014, is supported to enable declarations of functions and macros
601 from that TS. Note that not all features from that TS are supported by
604 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
605 18661-4:2015, is supported to enable declarations of functions and macros
606 from that TS. Note that most features from that TS are not supported by
609 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
610 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
611 Since the GNU C Library defaults to a much newer revision of POSIX, this
612 will only affect programs that specifically request an old conformance
613 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
614 see a change in the visible declarations, but a program compiled with
615 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
618 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
619 defined by all multithreaded code, but glibc has not required this for
622 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
623 means that in a future release, the macros “major”, “minor”, and “makedev”
624 will only be available from <sys/sysmacros.h>.
626 These macros are not part of POSIX nor XSI, and their names frequently
627 collide with user code; see for instance glibc bug 19239 and Red Hat bug
628 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
629 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
632 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
633 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
634 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
636 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
637 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
638 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
639 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
640 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
641 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
642 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
643 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
644 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
645 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
646 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
649 * New <math.h> features are added from TS 18661-1:2014:
651 - Signaling NaN macros: SNANF, SNAN, SNANL.
653 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
654 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
655 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
657 - llogb functions: the llogb, llogbf and llogbl functions, and the
658 FP_LLOGB0 and FP_LLOGBNAN macros.
660 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
663 - Comparison macros: iseqsig.
665 - Classification macros: iscanonical, issubnormal, iszero.
667 - Total order functions: totalorder, totalorderf, totalorderl,
668 totalordermag, totalordermagf, totalordermagl.
670 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
672 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
673 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
675 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
676 are added to libc. They convert a floating-point number into string.
678 * Most of glibc can now be built with the stack smashing protector enabled.
679 It is recommended to build glibc with --enable-stack-protector=strong.
680 Implemented by Nick Alcock (Oracle).
682 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
683 intended to be used instead of memset() to erase sensitive data after use;
684 the compiler will not optimize out calls to explicit_bzero even if they
685 are "unnecessary" (in the sense that no _correct_ program can observe the
686 effects of the memory clear).
688 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
689 to float instead of double. This does not affect the ABI of any libraries
690 that are part of the GNU C Library, but may affect the ABI of other
691 libraries that use this type in their interfaces.
693 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
694 float_t and double_t types are now defined to long double instead of float
695 and double. These options are not the default, and this does not affect
696 the ABI of any libraries that are part of the GNU C Library, but it may
697 affect the ABI of other libraries that use this type in their interfaces,
698 if they are compiled or used with those options.
700 * The getentropy and getrandom functions, and the <sys/random.h> header file
703 * The buffer size for byte-oriented stdio streams is now limited to 8192
704 bytes by default. Previously, on Linux, the default buffer size on most
705 file systems was 4096 bytes (and thus remains unchanged), except on
706 network file systems, where the buffer size was unpredictable and could be
707 as large as several megabytes.
709 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
710 for the Linux quota interface which predates kernel version 2.4.22 has
713 * The malloc_get_state and malloc_set_state functions have been removed.
714 Already-existing binaries that dynamically link to these functions will
715 get a hidden implementation in which malloc_get_state is a stub. As far
716 as we know, these functions are used only by GNU Emacs and this change
717 will not adversely affect already-built Emacs executables. Any undumped
718 Emacs executables, which normally exist only during an Emacs build, should
719 be rebuilt by re-running “./configure; make” in the Emacs build tree.
721 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
722 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
723 “no-ip6-dotint” had already been the default, and support for the
724 “ip6-dotint” option was removed from the Internet in 2006.
726 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
727 flag from <resolv.h> have been removed. The option relied on a
728 backwards-incompatible DNS extension which was never deployed on the
731 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
732 RES_BLAST defined in the <resolv.h> header file have been deprecated.
733 They were already unimplemented.
735 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
736 _res.flags are deprecated. The flag was standardized in RFC 2133, but
737 removed again from the IETF name lookup interface specification in RFC
738 2553. Applications should use getaddrinfo instead.
740 * DNSSEC-related declarations and definitions have been removed from the
741 <arpa/nameser.h> header file, and libresolv will no longer attempt to
742 decode the data part of DNSSEC record types. Previous versions of glibc
743 only implemented minimal support for the previous version of DNSSEC, which
744 is incompatible with the currently deployed version.
746 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
747 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
748 <arpa/nameser.h> header file because the distinction between RR types and
749 meta-RR types is not officially standardized, subject to revision, and
750 thus not suitable for encoding in a macro.
752 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
753 and rhook members of the res_state type in <resolv.h> have been removed.
754 The glibc stub resolver did not support these hooks, but the header file
755 did not reflect that.
757 * For multi-arch support it is recommended to use a GCC which has
758 been built with support for GNU indirect functions. This ensures
759 that correct debugging information is generated for functions
760 selected by IFUNC resolvers. This support can either be enabled by
761 configuring GCC with '--enable-gnu-indirect-function', or by
762 enabling it by default by setting 'default_gnu_indirect_function'
763 variable for a particular architecture in the GCC source file
766 * GDB pretty printers have been added for mutex and condition variable
767 structures in POSIX Threads. When installed and loaded in gdb these pretty
768 printers show various pthread variables in human-readable form when read
769 using the 'print' or 'display' commands in gdb.
771 * Tunables feature added to allow tweaking of the runtime for an application
772 program. This feature can be enabled with the '--enable-tunables' configure
773 flag. The GNU C Library manual has details on usage and README.tunables has
774 instructions on adding new tunables to the library.
776 * A new version of condition variables functions have been implemented in
777 the NPTL implementation of POSIX Threads to provide stronger ordering
780 * A new version of pthread_rwlock functions have been implemented to use a more
781 scalable algorithm primarily through not using a critical section anymore to
784 Security related changes:
786 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
787 have been created with makecontext could fail to terminate due to a
788 missing .cantunwind annotation. This has been observed to lead to a hang
789 (denial of service) in some Go applications compiled with gccgo. Reported
790 by Andreas Schwab. (CVE-2016-6323)
792 * The DNS stub resolver functions would crash due to a NULL pointer
793 dereference when processing a query with a valid DNS question type which
794 was used internally in the implementation. The stub resolver now uses a
795 question type which is outside the range of valid question type values.
798 The following bugs are resolved with this release:
800 [4099] stdio: Overly agressive caching by stream i/o functions.
801 [7065] build: Support building glibc with -fstack-protector or -fstack-
803 [9842] localedata: en_CA: incorrect date format
804 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
805 before it started waiting
806 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
807 [15765] nptl: sem_open is wrongly a cancellation point
808 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
809 [16458] libc: endian.h and netinet/in.h byte order macros return values of
811 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
812 library linked with pthread
813 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
815 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
816 [17252] libc: getrandom and getentropy syscall
817 [17730] malloc: thread-local storage is sometimes improperly free()'d
818 after being __libc_memalign()'d
819 [18241] stdio: failed fseek on memstream does not set errno and can fail
821 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
823 [18463] nptl: pthread_cond_broadcast issue when surrounded by
824 PTHREAD_PRIO_INHERIT mutex on ARM
825 [18784] network: res_query and related function crash for special record
826 type queries (CVE-2015-5180)
827 [19380] math: strtod does not raise "inexact"
828 [19387] string: Integer overflow in memchr
829 [19390] string: Integer overflow in strncat
830 [19398] build: linknamespace tests fail with massively parallel build
831 [19402] nptl: Deadlock with robust shared mutex and asynchronous
833 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
834 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
835 [19514] libc: [PATCH] Fix spelling errors in spelling
836 "implement"/"implementation" in several places
837 [19582] network: Deprecate RES_USE_INET6
838 [19673] manual: clog10 docs appear to be erroneous
839 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
840 causes a segmentation fault
841 [19826] libc: invalid pointer returned from __tls_get_addr with static
843 [20016] network: resolv: Remove hooks support from the API
844 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
846 [20033] math: [x86_64] vectorized math function don't call the __finite
848 [20116] nptl: use after free in pthread_create
849 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
850 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
852 [20311] nptl: please install proc_service.h
853 [20366] libc: Compilation errors in installed headers in strict-compliance
855 [20370] malloc: malloc: Arena free list management is still racy
856 (incorrect fix in bug 19243)
857 [20386] libc: assert (X = 0) does not result in GCC warning
858 [20432] malloc: malloc: Minimize interface required for interposition
859 [20435] libc: Missing unwind info in __startcontext causes infinite loop
860 in _Unwind_Backtrace (CVE-2016-6323)
861 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
862 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
864 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
865 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
866 [20477] network: resolv: incorrect double-checked locking related to
868 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
870 [20495] math: x86_64 performance degradation due to AVX/SSE transition
872 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
873 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
874 cause transition penalty
875 [20517] math: sparcv9 missing fdiml compat symbol
876 [20524] manual: strverscmp is inconsistent
877 [20525] libc: <sys/quota.h> should be based on kernel headers
878 [20539] math: GCC 7 -static -lm fails to link at -O3
879 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
881 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
883 [20591] network: Remove obsolete DNSSEC support
884 [20592] network: DNS resource record type classification macros in
885 <arpa/nameser.h> are incorrect
886 [20593] network: Update DNS RR type definitions
887 [20611] network: getaddrinfo accepts invalid numeric scope IDs
888 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
890 [20629] network: libresolv: Remove support for bitstring labels
892 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
893 assertion: "maxidx >= 2"
894 [20660] math: [arm] Use VSQRT
895 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
896 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
897 [20689] libc: Test for FMA should also check for AVX.
898 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
899 [20715] math: iszero macro breaks existing code
900 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
901 [20728] libc: powerpc: Missing TOC stub in clone
902 [20729] build: glibc-2.24 fails to build for i486 with -Os
903 [20750] build: Build fails with default PIE enabled gcc-6.x
904 [20768] math: [alpha] sqrt fegetenv namespace
905 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
907 [20787] math: float_t is defined as float incorrectly on x86_64 even if
908 __FLT_EVAL_METHOD__ is 2
909 [20790] network: rpcgen buffer overrun in get_prog_declaration
910 [20822] nptl: race condition in __lll_unlock_elision on powerpc
911 [20829] libc: crypt snprintf namespace
912 [20847] libc: tst-vfork3 failure
913 [20855] math: Default bits/mathdef.h has inappropriate float_t
914 [20859] math: [sh4] FP_ILOGB0 invalid
915 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
916 U+20AC), not same as GBK
917 [20915] dynamic-link: global-dynamic TLS broken on aarch64
918 [20916] math: pow handling of sNaN arguments
919 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
920 [20919] math: Bad pow (qNaN, 0) result with -lieee
921 [20940] math: hypot sNaN handling
922 [20947] math: fmax, fmin sNaN handling
923 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
924 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
926 [20971] string: powerpc64/power7 memchr overflows internal pointer check
927 [20973] nptl: robust mutexes: Lost wake-ups
928 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
929 [20978] nis: strlen on null pointer in nss_nisplus
930 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
932 [21014] string: i686 memchr overflows internal pointer check
933 [21019] libc: [mips] n32 lseek incorrect on overflow
934 [21022] libc: [microblaze] __backtrace get_frame_size namespace
935 [21026] libc: [MIPS] readahead syscall is broken on n64
936 [21028] math: Fallback fesetexceptflag should always succeed
937 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
938 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
940 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
941 [21061] librt: [microblaze] librt lost clock_* exports
942 [21073] libc: tunables: insecure environment variables passed to
943 subprocesses with AT_SECURE
944 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
948 * The minimum Linux kernel version that this version of the GNU C Library
949 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
950 kernel version 2.6.32 or later suffices (on architectures that already
951 required kernel versions more recent than 3.2, those requirements remain
952 unchanged). Linux 3.2 or later kernel headers are required on all
955 * The pap_AN locale has been deleted. This has been deprecated for a long
956 time. It has been replaced by pap_AW & pap_CW, both of which have long
957 been included in previous releases.
959 * The readdir_r and readdir64_r functions have been deprecated. It is
960 recommended to use readdir and readdir64 instead.
962 * The type “union wait” has been removed. It was deprecated in the early
963 1990s and never part of POSIX. Application code should use the int type
964 instead of “union wait”.
966 * A new NSS action is added to facilitate large distributed system
967 administration. The action, MERGE, allows remote user stores like LDAP
968 to be merged into local user stores like /etc/groups in order to provide
969 easy to use, updated, and managed sets of merged credentials. The new
970 action can be used by configuring it in /etc/nsswitch.conf:
971 group: files [SUCCESS=merge] nis
972 Implemented by Stephen Gallagher (Red Hat).
974 * The deprecated __malloc_initialize_hook variable has been removed from the
977 * The long unused localedef --old-style option has been removed. It hasn't
978 done anything in over 16 years. Scripts using this option can safely
981 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
982 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
983 return the next representable value in the direction of positive infinity
984 and the nextdown functions return the next representable value in the
985 direction of negative infinity. These are currently enabled as GNU
988 Security related changes:
990 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
991 could result in a stack overflow when getnetbyname was called with an
992 overly long name. (CVE-2016-3075)
994 * Previously, getaddrinfo copied large amounts of address data to the stack,
995 even after the fix for CVE-2013-4458 has been applied, potentially
996 resulting in a stack overflow. getaddrinfo now uses a heap allocation
997 instead. Reported by Michael Petlan. (CVE-2016-3706)
999 * The glob function suffered from a stack-based buffer overflow when it was
1000 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1001 Reported by Alexander Cherepanov. (CVE-2016-1234)
1003 * The Sun RPC UDP client could exhaust all available stack space when
1004 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1005 alloca plugin for GCC. (CVE-2016-4429)
1007 * The IPv6 name server management code in libresolv could result in a memory
1008 leak for each thread which is created, performs a failing naming lookup,
1009 and exits. Over time, this could result in a denial of service due to
1010 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1012 The following bugs are resolved with this release:
1014 [1170] localedata: ne_NP: update Nepali locale definition file
1015 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1017 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1018 [6796] math: fdim() does not set errno on overflow
1019 [10354] libc: posix_spawn should use vfork() in more cases than presently
1020 [11213] localedata: localedata: add copyright disclaimer to locale files
1021 [12143] localedata: chr_US: new Cherokee locale
1022 [12450] localedata: sgs_LT: new locale
1023 [12676] localedata: ln_CD: new locale
1024 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1026 [13304] math: fma, fmaf, fmal produce wrong results
1027 [14259] build: --localedir arg to configure is ignored
1028 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1029 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1030 [14934] localedata: es_CL: wrong first weekday chilean locale
1031 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1033 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1035 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1036 [15368] nptl: raise() is not async-signal-safe
1037 [15479] math: ceil, floor, round and trunc raise inexact exception
1038 [15578] localedata: kk_KZ: various updates
1039 [16003] localedata: pap_AN: punt old locale
1040 [16137] localedata: iw_IL: punt old locale
1041 [16190] localedata: eo: new esperanto locale
1042 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1043 [16742] malloc: race condition: pthread_atfork() called before first
1044 malloc() results in unexpected locking behaviour/deadlocks
1045 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1047 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1048 [17565] localedata: pt_PT: wrong (work-)week start
1049 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1050 [17950] build: Build fails with -msse
1051 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1052 [18433] libc: posix_spawn does not return correctly upon failure to
1054 [18453] localedata: charmaps/IBM875: incorrect codes
1055 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1057 [18896] localedata: he_IL: improvements for currency
1058 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1059 Romanian locale data
1060 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1062 [19056] libc: Deprecate readdir_r
1063 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1065 [19198] localedata: nl_NL: small improvements for Dutch locales
1066 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1067 nameservers (CVE-2016-5417)
1068 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1070 [19400] locale: Language missing in "iso-639.def", trivial fix in
1072 [19431] malloc: Deadlock between fflush, getdelim, and fork
1073 [19505] libc: Incorrect file descriptor validity checks in
1074 posix_spawn_file_actions_add{open,close,dup2}
1075 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1076 when using RTLD_NEXT
1077 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1078 `intl/{gettextP,loadinfo}.h'
1079 [19534] libc: execle, execlp may use malloc
1080 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1081 [19573] network: res_nclose and __res_maybe_init disagree about name
1082 server initialization, breaking Hesiod
1083 [19575] localedata: Status of GB18030 tables
1084 [19581] localedata: sr_* date_fmt string contains additional newline
1085 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1087 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1088 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1089 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1090 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1092 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1094 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1096 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1097 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1098 [19633] locale: strfmon_l applies global locale to number formatting
1099 [19642] network: Memory leak in getnameinfo
1100 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1101 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1103 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1104 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1106 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1107 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1109 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1111 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1113 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1114 pointers and lengths in error-case.
1115 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1116 report errors on UTF-16 surrogates values.
1117 [19755] nscd: nscd assertion failure in gc
1118 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1119 [19759] libc: mempcpy shouldn't be inlined
1120 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1121 [19765] libc: s390 needs an optimized mempcpy
1122 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1123 NAME_MAX limit assumption (CVE-2016-1234)
1124 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1125 [19787] network: Missing and incorrect truncation checks in getnameinfo
1126 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1128 [19791] network: Assertion failure in res_query.c with un-connectable name
1130 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1131 [19822] math: libm.so install clobbers old version
1132 [19825] network: resolv: send_vc can return uninitialized data in second
1133 response to getaddrinfo
1134 [19830] network: nss_dns: should check RDATA length against buffer length
1135 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1136 confronted with A/AAAA records of invalid size
1137 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1138 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1139 [19853] stdio: Printing IBM long double in decimal with high precision is
1141 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1142 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1143 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1145 [19865] network: Assertion failure or memory leak in
1146 _nss_dns_getcanonname_r
1147 [19868] network: nss_dns: netent code does not skip over non-PTR records
1148 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1150 [19881] string: Improve x86-64 memset
1151 [19907] string: Incorrect memcpy tests
1152 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1154 [19925] libc: termios.h XCASE namespace
1155 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1156 [19929] libc: limits.h NL_NMAX namespace
1157 [19931] stdio: Memory leak in vfprintf
1158 [19957] libc: clone(CLONE_VM) access invalid parent memory
1159 [19963] localedata: en_IL: New locale
1160 [19989] stdio: stdio.h cuserid namespace
1161 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1163 [19996] locale: langinfo.h nl_langinfo_l namespace
1164 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1165 [20010] network: getaddrinfo: Stack overflow in hostent translation
1167 [20012] stdio: libio: fmemopen append mode failure
1168 [20014] stdio: stdio.h namespace for pre-threads POSIX
1169 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1170 [20023] libc: fcntl.h timespec namespace
1171 [20024] math: [x86_64] vectorized sincos trashes the stack
1172 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1173 [20041] time: sys/time.h timespec namespace
1174 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1175 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1176 [20051] libc: ttyslot in wrong header under wrong conditions
1177 [20054] libc: gethostname not declared for XPG4
1178 [20055] libc: termios.h missing tcgetsid for XPG4
1179 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1181 [20073] libc: sys/stat.h fchmod namespace
1182 [20074] libc: stdlib.h rand_r namespace
1183 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1184 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1186 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1187 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1189 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1190 [20119] libc: Wrong mask for processors level type from CPUID
1191 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1192 AS not supporting AVX512
1193 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1195 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1196 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1198 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1199 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1200 [20178] libc: posix_spawn{p} should not call exit
1201 [20191] stdio: libio: vtables hardening
1202 [20195] string: FMA4 detection requires CPUID execution with register
1204 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1205 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1207 [20212] math: acos (sNaN) returns sNaN
1208 [20213] math: asin (sNaN) returns sNaN
1209 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1210 [20218] math: [i386] asinhl (sNaN) returns sNaN
1211 [20219] math: [i386] atanhl (sNaN) returns sNaN
1212 [20222] stdio: fopencookie: Mangle function pointers
1213 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1214 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1215 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1217 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1218 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1219 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1220 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1221 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1222 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1223 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1224 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1225 [20240] math: modf (sNaN) returns sNaN
1226 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1227 [20250] math: frexp (sNaN) returns sNaN
1228 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1229 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1231 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1232 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1233 subscript is below array bounds [-Werror=array-bounds]
1234 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1235 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1237 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1238 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1240 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1241 "invalid" exceptions
1242 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1243 [20316] localedata: id_ID: Februari instead of Pebruari
1244 [20327] string: POWER8 strcasecmp returns incorrect result
1245 [20347] math: Failure: Test: j0_downward (0xap+0)
1246 [20348] libc: FAIL: misc/tst-preadvwritev64
1247 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1248 [20350] libc: There is no test for p{read,write}64
1249 [20357] math: Incorrect cos result for 1.5174239687223976
1250 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1254 * Unicode 8.0.0 Support: Character encoding, character type info, and
1255 transliteration tables are all updated to Unicode 8.0.0, using new
1256 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1257 These updates cause user visible changes, such as the fixes for bugs
1258 89, 16061, and 18568.
1260 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1261 kernel-internal CPU set size. This means that requests that change the
1262 CPU affinity which failed before (for example, an all-ones CPU mask) will
1263 now succeed. Applications that need to determine the effective CPU
1264 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1265 setting it because the kernel can adjust it (and the previous size check
1266 would not detect this in the majority of cases).
1268 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1269 the following new symbols are used: fts64_children, fts64_close,
1270 fts64_open, fts64_read and fts64_set.
1272 * getaddrinfo now detects certain invalid responses on an internal netlink
1273 socket. If such responses are received, an affected process will
1274 terminate with an error message of "Unexpected error <number> on netlink
1275 descriptor <number>" or "Unexpected netlink response of size <number> on
1276 descriptor <number>". The most likely cause for these errors is a
1277 multi-threaded application which erroneously closes and reuses the netlink
1278 file descriptor while it is used by getaddrinfo.
1280 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1281 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1282 unnecessary serialization of memory allocation requests across threads.
1283 The defect is now corrected. Users should see a substantial increase in
1284 the concurent throughput of allocation requests for applications which
1285 trigger this bug. Affected applications typically create and
1286 destroy threads frequently. (Bug 19048 was reported and analyzed by
1289 * There is now a --disable-timezone-tools configure option for disabling the
1290 building and installing of the timezone related utilities (zic, zdump, and
1291 tzselect). This is useful for people who build the timezone data and code
1292 independent of the GNU C Library.
1294 * The obsolete header <regexp.h> has been removed. Programs that require
1295 this header must be updated to use <regex.h> instead.
1297 * The obsolete functions bdflush, create_module, get_kernel_syms,
1298 query_module and uselib are no longer available to newly linked binaries;
1299 the header <sys/kdaemon.h> has been removed. These functions and header
1300 were specific to systems using the Linux kernel and could not usefully be
1301 used with the GNU C Library on systems with version 2.6 or later of the
1304 * Optimized string, wcsmbs and memory functions for IBM z13.
1305 Implemented by Stefan Liebler.
1307 * Newly linked programs that define a variable called signgam will no longer
1308 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1309 require signgam to be set by those functions must ensure that they use the
1310 variable provided by the GNU C Library and declared in <math.h>, without
1311 defining their own copy.
1313 * The minimum GCC version that can be used to build this version of the GNU
1314 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1315 still be used to compile programs using the GNU C Library.
1317 Security related changes:
1319 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1320 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1322 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1323 the pointer guard feature. It is always enabled. Previously,
1324 LD_POINTER_GUARD could be used to disable security hardening in binaries
1325 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1328 * An integer overflow in hcreate and hcreate_r could lead to an
1329 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1331 * The catopen function no longer has unbounded stack usage. Reported by
1332 Max. (CVE-2015-8779)
1334 * The nan, nanf and nanl functions no longer have unbounded stack usage
1335 depending on the length of the string passed as an argument to the
1336 functions. Reported by Joseph Myers. (CVE-2014-9761)
1338 * A stack-based buffer overflow was found in libresolv when invoked from
1339 libnss_dns, allowing specially crafted DNS responses to seize control
1340 of execution flow in the DNS client. The buffer overflow occurs in
1341 the functions send_dg (send datagram) and send_vc (send TCP) for the
1342 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1343 family. The use of AF_UNSPEC triggers the low-level resolver code to
1344 send out two parallel queries for A and AAAA. A mismanagement of the
1345 buffers used for those queries could result in the response of a query
1346 writing beyond the alloca allocated buffer created by
1347 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1348 the overflow. Thanks to the Google Security Team and Red Hat for
1349 reporting the security impact of this issue, and Robert Holiday of
1350 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1352 The following bugs are resolved with this release:
1354 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1355 [887] math: Math library function "logb" and "nextafter" inconsistent
1356 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1357 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1358 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1359 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1361 [4404] localedata: German translation of "Alarm clock" is misleading
1362 [6799] math: nextafter() and nexttoward() doen't set errno on
1363 overflow/underflow errors
1364 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1366 [10432] nis: _nss_nis_setnetgrent assertion failure
1367 [11460] libc: fts has no LFS support
1368 [12926] network: getaddrinfo()/make_request() may spin forever
1369 [13065] nptl: Race condition in pthread barriers
1370 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1371 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1373 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1374 [14912] libc: Rename non-installed bits/*.h headers
1375 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1376 [15367] math: Let gcc use __builtin_isinf
1377 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1378 [15421] math: lgamma wrongly sets signgam for ISO C
1379 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1380 FE_INVALID with argument out of range
1381 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1383 [15786] dynamic-link: ifunc resolver functions can smash function
1385 [15918] math: Unnecessary check for equality in hypotf()
1386 [16061] localedata: Review / update transliteration data
1387 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1389 [16141] time: strptime %z offset restriction
1390 [16171] math: drem should be alias of remainder
1391 [16296] math: fegetround is pure?
1392 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1393 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1395 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1396 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1398 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1399 "Exception "Inexact" set" on ppc32
1400 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1401 [16517] math: Missing underflow exception from tanf/tan/tanl
1402 [16519] math: Missing underflow exception from sinhf
1403 [16520] math: Missing underflow exception from tanhf
1404 [16521] math: Missing underflow exception from exp2
1405 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1407 [16734] stdio: fopen calls mmap to allocate its buffer
1408 [16961] math: nan function incorrect handling of bad sequences
1409 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1410 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1411 [16985] locale: localedef: confusing error message when opening output
1413 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1414 [17197] locale: Redundant shift character in iconv conversion output at
1416 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1417 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1418 [17250] dynamic-link: static linking breaks nss loading
1419 (getaddrinfo/getpwnam/etc...)
1420 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1422 [17441] math: isnan() should use __builtin_isnan() in GCC
1423 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1424 (related to lock elision)
1425 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1426 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1427 [17887] time: strptime should be able to parse "+01:00" style timezones
1428 [17905] libc: catopen() Multiple unbounded stack allocations
1430 [18084] libc: backtrace (..., 0) dumps core on x86
1431 [18086] libc: nice() sets errno to 0 on success
1432 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1433 is too large (CVE-2015-8778)
1434 [18251] dynamic-link: SONAME missing when audit modules provides path
1435 [18265] libc: add attributes for wchar string and memory functions
1436 [18370] math: csqrt missing underflows
1437 [18421] libc: [hppa] read-only segment has dynamic relocations
1438 [18472] libc: Obsolete syscall wrappers should be compat symbols
1439 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1440 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1441 [18525] localedata: Remove locale timezone information
1442 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1443 [18568] localedata: Update locale data to Unicode 8.0
1444 [18589] locale: sort-test.sh fails at random
1445 [18595] math: ctan, ctanh missing underflows
1446 [18604] libc: assert macro-expands its argument
1447 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1448 contains a vector instruction exception.
1449 [18611] math: j1, jn missing errno setting on underflow
1450 [18618] localedata: sync Chechen locale definitions with other *_RU
1452 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1453 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1454 [18665] network: In send_dg, the recvfrom function is NOT always using the
1455 buffer size of a newly created buffer (CVE-2015-7547)
1456 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1458 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1460 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1461 [18699] math: tilegx cproj() for various complex infinities does not yield
1463 [18724] libc: Harden put*ent functions against data injection
1464 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1465 [18755] build: build errors with -DNDEBUG
1466 [18757] stdio: fmemopen fails to set errno on failure
1467 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1469 [18781] libc: openat64 lacks O_LARGEFILE
1470 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1471 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1472 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1473 [18790] math: [ldbl-128ibm] tanhl inaccurate
1474 [18795] libc: stpncpy fortification misses buffer lengths that are
1475 statically too large
1476 [18796] build: build fails for --disable-mathvec
1477 [18803] math: hypot missing underflows
1478 [18820] stdio: fmemopen may leak memory on failure
1479 [18823] math: csqrt spurious underflows
1480 [18824] math: fma spurious underflows
1481 [18825] math: pow missing underflows
1482 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1483 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1484 [18870] build: sem_open.c fails to compile with missing symbol
1486 [18872] stdio: Fix memory leak in printf_positional
1487 [18873] libc: posix_fallocate overflow check ineffective
1488 [18875] math: Excess precision leads incorrect libm
1489 [18877] libc: arm: mmap offset regression
1490 [18887] libc: memory corruption when using getmntent on blank lines
1491 [18918] localedata: hu_HU: change time to HH:MM:SS format
1492 [18921] libc: Regression: extraneous stat() and fstat() performed by
1494 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1495 binaries (CVE-2015-8777)
1496 [18951] math: tgamma missing underflows
1497 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1499 [18953] localedata: lt_LT: change currency symbol to the euro
1500 [18956] math: powf inaccuracy
1501 [18961] math: [i386] exp missing underflows
1502 [18966] math: [i386] exp10 missing underflows
1503 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1504 [18969] build: multiple string test failures due to missing locale
1506 [18970] libc: Reference of pthread_setcancelstate in libc.a
1507 [18977] math: float / long double Bessel functions not in XSI POSIX
1508 [18980] math: i386 libm functions return with excess range and precision
1509 [18981] math: i386 scalb*, ldexp return with excess range and precision
1510 [18982] stdio: va_list and vprintf
1511 [18985] time: Passing out of range data to strftime() causes a segfault
1513 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1514 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1515 [19012] locale: iconv_open leaks memory on error path
1516 [19016] math: clog, clog10 inaccuracy
1517 [19018] nptl: Mangle function pointers in tls_dtor_list
1518 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1519 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1520 [19048] malloc: malloc: arena free list can become cyclic, increasing
1522 [19049] math: [powerpc] erfc incorrect zero sign
1523 [19050] math: [powerpc] log* incorrect zero sign
1524 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1525 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1526 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1527 [19074] network: Data race in _res_hconf_reorder_addrs
1528 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1529 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1530 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1532 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1534 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1535 [19086] manual: posix_fallocate64 documented argument order is wrong.
1536 [19088] math: lround, llround missing exceptions close to overflow
1538 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1539 [19095] math: dbl-64 lrint incorrect for 64-bit long
1540 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1541 [19124] dynamic-link: ld.so failed to build with older assmebler
1542 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1543 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1544 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1545 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1546 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1547 pthread_setaffinity_np
1548 [19156] math: [ldbl-128] j0l spurious underflows
1549 [19164] nptl: tst-getcpu fails with many possible CPUs
1550 [19168] math: math/test-ildoubl and math/test-ldouble failure
1551 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1552 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1554 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1555 (FE_NOMASK_ENV) do not clear SSE exceptions
1556 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1557 _int_new_arena/reused_arena
1558 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1559 [19201] math: dbl-64 remainder incorrect sign of zero result
1560 [19205] math: bits/math-finite.h conditions do not match math.h and
1562 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1563 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1564 for C99-based standards
1565 [19212] libc: features.h not -Wundef clean
1566 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1568 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1569 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1570 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1572 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1573 "inexact" exceptions
1574 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1576 [19242] libc: strtol incorrect in Turkish locales
1577 [19243] malloc: reused_arena can pick an arena on the free list, leading
1578 to an assertion failure and reference count corruption
1579 [19253] time: tzset() ineffective when temporary TZ did not include DST
1581 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1582 [19270] math: [hppa] Shared libm missing __isnanl
1583 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1584 MAP_HUGETLB and MAP_STACK defines
1585 [19313] nptl: Wrong __cpu_mask for x32
1586 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1587 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1588 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1589 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1590 [19363] time: x32: times() return value wrongly truncates/sign extends
1592 [19367] dynamic-link: Improve branch prediction on Silvermont
1593 [19369] network: Default domain name not reset by res_ninit when "search"
1594 / "domain" entry is removed from resolv.conf
1595 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1597 [19385] localedata: bg_BG: time separator should be colon, not comma
1598 [19408] libc: linux personality syscall wrapper may erroneously return an
1599 error on 32-bit architectures
1600 [19415] libc: dladdr returns wrong names on hppa
1601 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1602 IBM905, IBM907, and IBM909
1603 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1604 [19443] build: build failures with -DDEBUG
1605 [19451] build: Make check fails on test-double-vlen2
1606 [19462] libc: Glibc failed to build with -Os
1607 [19465] math: Wrong code with -Os
1608 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1610 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1612 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1613 [19486] math: S390: Math tests fail with "Exception Inexact set".
1614 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1615 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1617 [19590] math: Fail to build shared objects that use libmvec.so functions.
1621 * The following bugs are resolved with this release:
1623 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1624 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1625 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1626 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1627 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1628 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1629 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1630 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1631 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1632 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1633 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1634 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1635 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1636 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1637 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1638 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1639 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1640 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1641 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1642 18657, 18676, 18694, 18696, 18887.
1644 * Cache information can be queried via sysconf() function on s390 e.g. with
1645 _SC_LEVEL1_ICACHE_SIZE as argument.
1647 * A buffer overflow in gethostbyname_r and related functions performing DNS
1648 requests has been fixed. If the NSS functions were called with a
1649 misaligned buffer, the buffer length change due to pointer alignment was
1650 not taken into account. This could result in application crashes or,
1651 potentially arbitrary code execution, using crafted, but syntactically
1652 valid DNS responses. (CVE-2015-1781)
1654 * The time zone file parser has been made more robust against crafted time
1655 zone files, avoiding heap buffer overflows related to the processing of
1656 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1657 large time zone data files. Overly long time zone specifiers in the TZ
1658 variable no longer result in stack overflows and crashes.
1660 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1661 for LD and GD on x86 and x86-64, has been implemented. You will need
1662 binutils-2.24 or later to enable this optimization.
1664 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1665 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1666 Hat). These updates cause user visible changes, such as the fix for bug
1669 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1670 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1671 condition in some applications.
1673 * Added vector math library named libmvec with the following vectorized x86_64
1674 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1676 The library can be disabled with --disable-mathvec. Use of the functions is
1677 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1678 Shared library libmvec.so is linked in as needed when using -lm (no need to
1679 specify -lmvec explicitly for not static builds).
1680 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1682 * A new fmemopen implementation has been added with the goal of POSIX
1683 compliance. The new implementation fixes the following long-standing
1684 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1685 old implementation is still present for use be by existing binaries.
1687 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1688 releases. It has been fixed to match 2.19 and older, but binaries built
1689 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1691 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1692 Contributed by Roland McGrath (Google).
1694 * The header <regexp.h> is deprecated, and will be removed in a future
1695 release. Use of this header will trigger a deprecation warning.
1696 Application developers should update their code to use <regex.h> instead.
1698 This header was formerly part of SUS, but was deprecated in 1994 and
1699 removed from the standard in 2001. Also, the glibc implementation
1700 leaks memory. See BZ#18681 for more details.
1704 * The following bugs are resolved with this release:
1706 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1707 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1708 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1709 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1710 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1711 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1712 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1713 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1714 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1715 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1718 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1719 too little memory for the to-be-scanned arguments and overflow the
1720 allocated buffer. The implementation now correctly computes the required
1721 buffer size when using malloc, and switches to malloc from alloca as
1724 * A new semaphore algorithm has been implemented in generic C code for all
1725 machines. Previous custom assembly implementations of semaphore were
1726 difficult to reason about or ensure that they were safe. The new version
1727 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1728 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1729 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1731 * Port to Altera Nios II has been contributed by Mentor Graphics.
1733 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1734 implementations for powerpc64/powerpc64le.
1735 Implemented by Adhemerval Zanella (IBM).
1737 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1738 and powerpc64le. This may improve lock scaling of existing programs on
1739 HTM capable systems. The lock elision code is only enabled with
1740 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1741 powerpc will issue a transaction abort on every syscall to avoid side
1742 effects being visible outside transactions.
1744 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1745 AArch64. Contributed by ARM Ltd.
1747 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1749 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1750 under certain input conditions resulting in the execution of a shell for
1751 command substitution when the application did not request it. The
1752 implementation now checks WRDE_NOCMD immediately before executing the
1753 shell and returns the error WRDE_CMDSUB as expected.
1755 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1756 processing format strings with a large number of format specifiers.
1758 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1759 infinite loop if the DNS response contained a PTR record of an unexpected
1762 * The minimum GCC version that can be used to build this version of the GNU
1763 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1764 still be used to compile programs using the GNU C Library.
1766 * The GNU C Library is now built with -Werror by default. This can be
1767 disabled by configuring with --disable-werror.
1769 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1771 * The obsolete sigvec function has been removed. This was the original
1772 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1773 programs have been using instead for about 25 years. Of course, ABI
1774 compatibility for old binaries using sigvec remains intact.
1776 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1777 with newer versions of bison.
1779 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1780 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1781 registers overlay two consecutive single-precision registers. MIPS32R2
1782 introduced a new FPU mode (FR=1) where double-precision registers extend the
1783 corresponding single-precision registers which is incompatible with the
1784 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1785 require the use of FR=1 making a transition necessary. New o32 ABI
1786 extensions enable users to migrate over time from the original o32 ABI
1787 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1788 tracks the ABI of any loaded object and verifies that new objects are
1789 compatible. Mode transitions will also be requested as required and
1790 unsupportable objects will be rejected. The ABI checks include both soft and
1791 hard float ABIs for o32, n32 and n64.
1793 GCC 5 with GNU binutils 2.25 onwards:
1794 It is strongly recommended that all o32 system libraries are built using the
1795 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1796 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1797 compiler using --with-fp-32=xx to set this by default.
1801 * The following bugs are resolved with this release:
1803 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1804 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1805 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1806 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1807 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1808 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1809 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1810 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1811 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1812 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1813 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1814 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1815 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1816 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1817 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1819 * Reverted change of ABI data structures for s390 and s390x:
1820 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1821 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1822 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1823 are preserved pointing straight to the same implementation as the old ones.
1824 Given that, new callers will simply provide a too-big buffer to these
1825 functions. Any applications/libraries out there that embed jmp_buf or
1826 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1827 against 2.19 headers will have to rebuilt again. This is necessary in any
1828 case to revert the breakage in their ABI caused by the glibc change.
1830 * Support for file description locks is added to systems running the
1831 Linux kernel. The standard file locking interfaces are extended to
1832 operate on file descriptions, not file descriptors, via the use of
1833 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1834 are associated with an open file instead of a process.
1836 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1838 * The minimum Linux kernel version that this version of the GNU C Library
1839 can be used with is 2.6.32.
1841 * Running the testsuite no longer terminates as soon as a test fails.
1842 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1843 with PASS or FAIL lines for individual tests. A summary of the results is
1844 printed, including a list of failing lists, and "make check" exits with
1845 error status if there were any unexpected failures. "make check
1846 stop-on-test-failure=y" may be used to keep the old behavior.
1848 * The am33 port, which had not worked for several years, has been removed
1851 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1852 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1853 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1854 conflicted with POSIX (support for which was removed in 2.19), the
1855 interfaces those macros enabled remain available when compiling with
1856 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1857 test macros defined.
1859 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1861 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1862 This may improve lock scaling of existing programs on TX capable systems.
1863 The lock elision code is only built with --enable-lock-elision=yes and
1864 then requires a GCC version supporting the TX builtins. With lock elision
1865 default mutexes are elided via __builtin_tbegin, if the cpu supports
1866 transactions. By default lock elision is not enabled and the elision code
1869 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1870 copy the path argument. This allowed programs to cause posix_spawn to
1871 deference a dangling pointer, or use an unexpected pathname argument if
1872 the string was modified after the posix_spawn_file_actions_addopen
1875 * All supported architectures now use the main glibc sysdeps directory
1876 instead of some being in a separate "ports" directory (which was
1877 distributed separately before glibc 2.17).
1879 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1880 On configurations that support it (all Linux configurations), it's now
1881 used regardless of the --enable-add-ons switch to configure. It is no
1882 longer possible to build such configurations without pthreads support.
1884 * Locale names, including those obtained from environment variables (LANG
1885 and the LC_* variables), are more tightly checked for proper syntax.
1886 setlocale will now fail (with EINVAL) for locale names that are overly
1887 long, contain slashes without starting with a slash, or contain ".." path
1888 components. (CVE-2014-0475) Previously, some valid locale names were
1889 silently replaced with the "C" locale when running in AT_SECURE mode
1890 (e.g., in a SUID program). This is no longer necessary because of the
1893 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1894 with application code using Intel MPX instructions. (With all previous
1895 versions, the MPX register state could be clobbered when making calls
1896 into or out of a shared library.) Note that while the new dynamic
1897 linker is compatible with all known x86 hardware whether or not it
1898 supports Intel MPX, some x86 instruction-set emulators might fail to
1899 handle the new instruction encodings. This is known to affect Valgrind
1900 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1901 release), and might affect other tools that do instruction emulation.
1903 * Support for loadable gconv transliteration modules has been removed.
1904 The support for transliteration modules has been non-functional for
1905 over a decade, and the removal is prompted by security defects. The
1906 normal gconv conversion modules are still supported. Transliteration
1907 with //TRANSLIT is still possible, and the //IGNORE specifier
1908 continues to be supported. (CVE-2014-5119)
1910 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1911 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1912 resulting a denial-of-service security vulnerability in applications which
1913 use functions related to iconv. (CVE-2014-6040)
1917 * The following bugs are resolved with this release:
1919 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1920 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1921 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1922 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1923 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1924 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1925 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1926 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1927 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1928 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1929 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1930 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1931 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1932 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1933 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1934 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1935 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1936 16431, 16453, 16474, 16506, 16510, 16529
1938 * Slovenian translations for glibc messages have been contributed by the
1939 Translation Project's Slovenian team of translators.
1941 * The public headers no longer use __unused nor __block. This change is to
1942 support compiling programs that are derived from BSD sources and use
1943 __unused internally, and to support compiling with Clang's -fblock
1944 extension which uses __block.
1946 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1947 large collation sequences to optimize multiple passes. This cache
1948 computation may overflow for large collation sequences and may cause a
1949 stack or buffer overflow. This is now fixed to use a slower algorithm
1950 which does not use a cache if there is an integer overflow.
1952 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1953 rules for large collation sequences to optimize multiple passes and falls
1954 back to alloca if malloc fails, resulting in a possible stack overflow.
1955 The implementation now falls back to an uncached collation sequence lookup
1958 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1959 initialized for static applications resulting in the security feature
1960 being disabled. The pointer guard is now correctly initialized to a
1961 random value for static applications. Existing static applications need
1962 to be recompiled to take advantage of the fix (bug 15754).
1964 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1965 to the d_name member of struct dirent, or omit the terminating NUL
1966 character. (Bugzilla #14699).
1968 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1969 aligned_alloc functions could allocate too few bytes or corrupt the
1970 heap when passed very large allocation size values (Bugzilla #15855,
1973 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1974 for AF_INET6 has been fixed (Bugzilla #16072).
1976 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1977 pap_CW, quz_PE, the_NP.
1979 * Substantially revised locales: gd_GB, ht_HT
1981 * The LC_ADDRESS field was updated to support country_car for almost all
1984 * ISO 1427 definitions were updated.
1986 * ISO 3166 definitions were updated.
1988 * The localedef utility now supports --big-endian and --little-endian
1989 command-line options to generate locales for a different system from that
1990 for which the C library was built.
1992 * Binary locale files now only depend on the endianness of the system for
1993 which they are generated and not on other properties of that system. As a
1994 consequence, binary files generated with new localedef may be incompatible
1995 with old versions of the GNU C Library, and binary files generated with
1996 old localedef may be incompatible with this version of the GNU C Library,
1997 in the following circumstances:
1999 + Locale files may be incompatible on m68k systems.
2001 + Locale archive files (but not separate files for individual locales) may
2002 be incompatible on systems where plain "char" is signed.
2004 * The configure option --disable-versioning has been removed. Builds with
2005 --disable-versioning had not worked for several years.
2007 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2008 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2010 * SystemTap probes for malloc have been introduced.
2012 * SystemTap probes for slow multiple precision fallback paths of
2013 transcendental functions have been introduced.
2015 * Support for powerpc64le has been added.
2017 * The soft-float powerpc port now supports e500 processors.
2019 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2021 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2022 set of header declarations that are enabled by default, even when other
2023 feature test macros or compiler options such as -std=c99 would otherwise
2024 disable some of those declarations.
2026 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2027 conflict with POSIX. The libbsd-compat library (which was a dummy library
2028 that did nothing) has also been removed.
2030 * Preliminary documentation about Multi-Thread, Async-Signal and
2031 Async-Cancel Safety has been added.
2033 * Change of ABI data structures for s390 and s390x:
2034 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2035 allow for future hardware extensions. All library functions that accept or
2036 return these structures were versioned in order to provide backwards
2037 compatibility with existing code. However, not all cases can be handled
2038 with interface versioning. If an application embeds either structure into
2039 another structure and that structure is passed to another compilation unit
2040 which expects the newer larger-sized object then any access to the new
2041 fields will result in undefined behaviour. Similarly any access to fields
2042 that were moved by the enlarging of the structures will cause undefined
2043 behaviour. To fix the undefined behaviour all objects that are part of the
2044 final in-memory image must agree on the size of structures, and this may
2045 require recompilation.
2049 * The following bugs are resolved with this release:
2051 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2052 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2053 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2054 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2055 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2056 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2057 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2058 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2059 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2060 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2061 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2062 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2063 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2066 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2067 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2068 Distributions can re-enable building and using pt_chown via the new configure
2069 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2070 considerable security risks and should only be used if the distribution
2071 understands and accepts the risks.
2073 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2076 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2077 fixed (Bugzilla #15330).
2079 * Add support for calling C++11 thread_local object destructors on thread
2080 and program exit. This needs compiler support for offloading C++11
2081 destructor calls to glibc.
2083 * Improved worst case performance of libm functions with double inputs and
2086 * Support for priority inherited mutexes in pthread condition variables on
2087 non-x86 architectures.
2089 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2091 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2093 * Optimized string functions for ARM. Implemented by Will Newton and
2096 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2098 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2101 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2102 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2104 * Added a benchmark framework to track performance of functions in glibc.
2106 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2107 It is based on draft TS 18661 and currently enabled as a GNU extension.
2109 * On Linux, the clock function now uses the clock_gettime system call
2110 for improved precision, rather than old times system call.
2112 * Added support for version-3 tz format files. This is needed when using
2113 the tz database release 2013e or later, and affects a few unusual cases --
2114 currently only TZ='America/Godthab' for time stamps after 2037.
2116 * Added new API functions pthread_getattr_default_np and
2117 pthread_setattr_default_np to get and set the default pthread
2118 attributes of a process.
2120 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2121 This may improve lock scaling of existing programs on TSX capable systems.
2122 When the --enable-lock-elision=yes parameter is specified at configure
2123 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2126 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2127 availability). Implemented by Ryan S. Arnold.
2129 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2133 * The following bugs are resolved with this release:
2135 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2136 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2137 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2138 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2139 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2140 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2141 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2142 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2143 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2144 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2145 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2146 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2147 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2148 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2149 14879, 14889, 14893, 14898, 14914.
2151 * Optimization of memcpy for MIPS.
2153 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2154 EMFILE has been fixed (Bugzilla #14889).
2156 * The add-on ports collection is now distributed in the "ports" subdirectory
2157 of the main GNU C Library distribution, rather than separately.
2159 * Port to ARM AArch64 contributed by Linaro.
2161 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2162 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2164 Implemented by Andreas Krebbel.
2166 * The new function secure_getenv allows secure access to the environment,
2167 returning NULL if running in a SUID/SGID process. This function replaces
2168 the internal function __secure_getenv.
2170 * SystemTap static probes have been added into the dynamic linker.
2171 Implemented by Gary Benson.
2173 * Optimizations of string functions strstr, strcasestr and memmem.
2174 Implemented by Maxim Kuvyrkov.
2176 * The minimum Linux kernel version that this version of the GNU C Library
2177 can be used with is 2.6.16.
2179 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2180 powerpc POWER7. Implemented by Will Schmidt.
2182 * New configure option --disable-nscd builds the C library such that it
2183 never attempts to contact the Name Service Caching Daemon (nscd).
2184 New configure option --disable-build-nscd avoids building nscd itself;
2185 this is the default if --disable-nscd is used.
2187 * Improved support for cross-compilation, including cross-testing and
2188 bootstrap builds without a previously built glibc.
2190 * Several testsuite tests are now able to test multiple IFUNC variants of an
2191 interface, rather than just testing the one that would be chooen by
2194 * New configure options --with-bugurl and --with-pkgversion, for
2195 distributors to use to embed their bug-reporting and package version
2196 information in --help and --version output.
2198 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2199 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2200 allows creation of chroots without the procfs mounted on /proc.
2202 * The `crypt' function now fails if passed salt bytes that violate the
2203 specification for those values. On Linux, the `crypt' function will
2204 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2205 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2206 when the mode is enabled.
2208 * The `clock_*' suite of functions (declared in <time.h>) is now available
2209 directly in the main C library. Previously it was necessary to link with
2210 -lrt to use these functions. This change has the effect that a
2211 single-threaded program that uses a function such as `clock_gettime' (and
2212 is not linked with -lrt) will no longer implicitly load the pthreads
2213 library at runtime and so will not suffer the overheads associated with
2214 multi-thread support in other code such as the C++ runtime library.
2216 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2221 * The following bugs are resolved with this release:
2223 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2224 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2225 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2226 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2227 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2228 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2229 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2230 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2231 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2232 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2233 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2234 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2235 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2236 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2237 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2238 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2239 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2240 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2241 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2242 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2243 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2244 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2247 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2248 configuring glibc with:
2249 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2250 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2251 Implemented by H.J. Lu.
2255 + define static_assert
2257 + do not declare gets
2259 + declare at_quick_exit and quick_exit also for ISO C11
2261 + aligned_alloc. NB: The code is deliberately allows the size parameter
2262 to not be a multiple of the alignment. This is a moronic requirement
2263 in the standard but it is only a requirement on the caller, not the
2266 + timespec_get added
2268 + uchar.h support added
2270 + CMPLX, CMPLXF, CMPLXL added
2272 Implemented by Ulrich Drepper.
2274 * Support for the IA-64 has been moved to ports.
2276 * Remove support for anything but ELF binary format
2278 * Checking versions of poll, ppoll added.
2279 Implemented by Ulrich Drepper.
2281 * More generic and 64-bit performance optimizations to math functions.
2282 Implemented by Ulrich Drepper.
2284 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2285 headers and functions available at compile time as they were before
2286 version 2.14. This option will be removed at some time in the future
2287 after the TI-RPC library becomes fully sufficient for the needs of
2288 existing applications.
2290 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2291 Note that glibc is not expected to work with any Linux kernel version
2294 * New header <sys/auxv.h> and function getauxval allowing easy access to
2295 the AT_* key-value pairs passed from the Linux kernel. The header also
2296 defines the HWCAP_* bits associated with the AT_HWCAP key.
2298 * New locales: mag_IN
2300 * New configure option --enable-systemtap builds SystemTap static probes
2301 into libc for setjmp and longjmp and into libpthread for various operations.
2302 So far the setjmp/longjmp probes and some of the libpthread probes are
2303 provided only for i*86 and x86_64.
2304 Implemented by Roland McGrath and Rayson Ho.
2306 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2308 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2311 * More optimized functions for SPARC. Implemented by David S. Miller.
2313 * Improved support for cross-compilation, especially bootstrap builds
2314 without a previously built glibc.
2316 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2317 Chris Metcalf from Tilera.
2319 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2320 now supported for ARM processors.
2322 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2323 as the name of the dynamic linker, to distinguish it from the
2324 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2326 * Support for CRIS has been removed from ports.
2328 * A new class of installed header has been documented for low-level
2329 platform-specific functionality. PowerPC added the first instance with a
2330 function to provide time base register access. Contributed by Tulio
2331 Magno Quites Machado Filho.
2333 * ix86 configurations now install header files that are consistent with
2334 what x86-64 configurations install. These same header files can be used
2335 for -m32, -m64, or -mx32 builds.
2336 Contributed by H.J. Lu.
2338 * Math library bug fixes. A thorough audit of all open math library bugs was
2339 conducted by Joseph Myers. Significant progress was made on many math
2340 library bugs resulting in more accurate exceptions and function results.
2341 Many thanks to all those that contributed including Andreas Jaeger for his
2342 patch review and work on the x87 trigonometric instruction issues.
2344 * Timezone data is no longer installed. Timezone-related binaries and scripts
2345 will continue to be installed. Users should obtain their timezone data from
2346 their distribution provider or from the tzdata package at
2347 <ftp://munnari.oz.au/pub/>.
2351 * The following bugs are resolved with this release:
2353 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2354 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2355 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2356 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2357 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2358 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2359 13484, 13506, 13515, 13523, 13524, 13538, 13540
2361 * New program pldd to list loaded object of a process
2362 Implemented by Ulrich Drepper.
2364 * Add nss_db support back to glibc. No more dependency on Berkeley db
2365 and support for initgroups lookups.
2366 Implemented by Ulrich Drepper.
2368 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2369 Contributed by HJ Lu.
2371 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2372 Contributed by HJ Lu.
2374 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2375 on x86-32 and x86-64.
2376 Contributed by Liubov Dmitrieva.
2378 * Optimized strchr and strrchr for SSE on x86-32.
2379 Contributed by Liubov Dmitrieva.
2381 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2382 for x86-64 and x86-32.
2383 Contributed by Liubov Dmitrieva.
2385 * New interfaces: scandirat, scandirat64
2386 Implemented by Ulrich Drepper.
2388 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2389 Implemented by Ulrich Drepper.
2391 * nscd now also caches the netgroup database.
2392 Implemented by Ulrich Drepper.
2394 * Integrate libm with gcc's -ffinite-math-only option.
2395 Implemented by Ulrich Drepper.
2397 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2398 to math functions. Implemented by Ulrich Drepper.
2400 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2401 Implemented by Ulrich Drepper.
2403 * New Linux interfaces: process_vm_readv, process_vm_writev
2405 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2406 Implemented by Ulrich Drepper.
2408 * Optimized nearbyint and strcasecmp for PPC.
2409 Implemented by Adhemerval Zanella.
2411 * New locales: bho_IN, unm_US, es_CU, ta_LK
2415 * The following bugs are resolved with this release:
2417 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2418 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2419 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2420 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2421 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2422 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2423 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2424 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2425 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2426 12795, 12811, 12813, 12814, 12841
2428 * The RPC implementation in libc is obsoleted. Old programs keep working
2429 but new programs cannot be linked with the routines in libc anymore.
2430 Programs in need of RPC functionality must be linked against TI-RPC.
2431 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2433 Visible changes of this change include (obviously) the inability to link
2434 programs using RPC functions without referencing the TI-RPC library and the
2435 removal of the RPC headers from the glibc headers.
2436 Implemented by Ulrich Drepper.
2438 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2439 syncfs, setns, sendmmsg
2441 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2442 yue_HK, lij_IT, mhr_RU
2444 * New encodings: CP770, CP771, CP772, CP773, CP774
2446 * New program sotruss to trace calls through PLTs
2447 Implemented by Ulrich Drepper.
2449 * The malloc hook implementation is marked deprecated and will be removed
2450 from the default implementation in the next version. The design never
2451 worked ever since the introduction of threads. Even programs which do
2452 not create threads themselves can use multiple threads created internally.
2456 * The following bugs are resolved with this release:
2458 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2459 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2460 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2461 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2464 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2466 * POWER7 optimizations: memset, memcmp, strncmp
2468 * New optimized string functions for x86-64: strnlen (SSE2),
2469 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2470 Implemented by Ulrich Drepper.
2474 * The following bugs are resolved with this release:
2476 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2477 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2478 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2479 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2480 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2481 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2482 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2483 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2485 * New interfaces: pthread_getname_np, pthread_setname_np
2487 * New Linux interface: recvmmsg
2489 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2491 * The dynamic linker now recognizes supported ABI versions from the
2492 EI_ABIVERSION field in the ELF header.
2493 Implemented by Ulrich Drepper.
2495 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2496 will not be used to fill in password fields in the passwd.byname replies.
2497 Instead it is used to synthesize the shadow.byname table, should it be
2498 missing. This is a useful mode in some installations involving Solaris.
2499 Implemented by Ulrich Drepper.
2501 * New locales: kok_IN, sq_MK, cv_RU
2505 * The following bugs are resolved with this release:
2507 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2508 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2509 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2510 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2511 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2512 10789, 10817, 10823, 10840, 10847
2514 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2516 Implemented by Ulrich Drepper.
2518 * Checking version of longjmp added that fails if an uninitialized stack
2519 frame would be created. Implemented by Ulrich Drepper.
2521 * STT_GNU_IFUNC is now supported in static executables.
2522 Implemented by H.J. Lu.
2524 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2526 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2527 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2528 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2529 Contributed by H.J. Lu.
2531 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2532 Implemented by Ulrich Drepper.
2534 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2536 Contributed by H.J. Lu.
2538 * Support for fma instruction in AVX on x86-64.
2539 Implemented by H.J. Lu and Ulrich Drepper.
2541 * AVX support in x86-64 auditing support in ld.so.
2542 Implemented by H.J. Lu.
2544 * STB_GNU_UNIQUE support added.
2545 Implemented by Ulrich Drepper.
2547 * Implement second fallback mode for optimized DNS lookup for even more
2548 broken environments. If two requests from the same port are not
2549 handled correctly close the socket and open a new one before sending
2550 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2551 can be used to select this mode right away, instead of rediscovering the
2552 necessity is every process again.
2553 Implemented by Ulrich Drepper.
2555 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2556 Implemented by Adam Tkac.
2558 * Optimized iconv conversions for S390x.
2559 Implemented by Andreas Krebbel.
2561 * Using condvars with PI mutexes is now more efficient due to kernel
2562 support for requeueing to PI futexes. NPTL support added for x86-64.
2563 Implemented by Ulrich Drepper.
2565 * New locales: ps_AF, my_MM
2569 * The following bugs are resolved with this release:
2571 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2572 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2573 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2574 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2575 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2577 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2578 Implemented by Ulrich Drepper.
2580 * New Linux interfaces: accept4, fallocate, fallocate64.
2581 Implemented by Ulrich Drepper.
2583 * Correct declarations of string function when used in C++ code. This
2584 could lead to compile errors for invalid C++ code.
2586 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2589 * New POSIX 2008 interface: psiginfo
2590 Implemented by Ulrich Drepper.
2592 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2593 Implemented by Ulrich Drepper.
2595 * Support for selecting between multiple function definitions at runtime
2596 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2598 * The libcrypt library can now use the hash function implementations in
2599 NSS. Implemented by Ulrich Drepper.
2601 * The malloc implementation can be compiled to be less memory efficient
2602 but higher performing in multi-threaded programs.
2603 Implemented by Ulrich Drepper.
2605 * New locales: nan_TW@latin, ks_IN
2607 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2608 Implemented by Ulrich Drepper.
2610 * Extended printf hook support. It is possible to use user-defined types
2611 and extend existing format specifiers.
2612 Implemented by Ulrich Drepper.
2614 * Handling for group shadow files has been added.
2615 Implemented by Ulrich Drepper.
2617 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2618 servers (the case, e.g., for some people using the built-in DNS
2619 server in ADSL modems/routers). There is a once-per-process timeout
2620 in case of a broken server. To avoid it, users can run nscd or put
2621 'options single-request' in /etc/resolv.conf.
2622 Implemented by Ulrich Drepper.
2626 * The following bugs are resolved with this release:
2628 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2629 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2630 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2631 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2632 6968, 6974, 6980, 6995, 7008, 7009, 7029
2634 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2635 up at the same time. Implemented by Ulrich Drepper.
2637 * TLS descriptors for LD and GD on x86 and x86-64.
2638 Implemented by Alexandre Oliva.
2640 * getaddrinfo now handles DCCP and UDPlite.
2641 Implemented by Ulrich Drepper.
2643 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2644 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2645 Implemented by Ulrich Drepper.
2647 * New implementation of memmem, strstr, and strcasestr which is O(n).
2648 Implemented by Eric Blake.
2650 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2652 * Implement "e" option for popen to open file descriptor with the
2653 close-on-exec flag set. Implemented by Ulrich Drepper.
2655 * Implement "b" mode for fmemopen. In this mode writes writes don't
2656 implicitly add a NUL byte and seeks from the end of the buffer really
2657 use the buffer end, not the string length as the basis.
2658 Implemented by Ulrich Drepper.
2660 * Many functions, exported and internal, now atomically set the close-on-exec
2661 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2663 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2665 Implemented by Pravin Satpute.
2667 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2671 * The following bugs are resolved with this release:
2673 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2674 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2675 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2676 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2677 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2678 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2679 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2681 * New locales: bo_CN, bo_IN, shs_CA.
2683 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2685 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2686 Implemented by Pravin Satpute.
2688 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2690 * nscd caches now all timeouts for DNS entries
2691 Implemented by Ulrich Drepper.
2693 * nscd is more efficient and wakes up less often.
2694 Implemented by Ulrich Drepper.
2696 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2697 vdprintf, and obstack_vprintf.
2698 Implemented by Jakub Jelinek.
2700 * Faster memset for x86-64.
2701 Implemented by Harsha Jagasia and H.J. Lu.
2703 * Faster memcpy on x86.
2704 Implemented by Ulrich Drepper.
2706 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2707 Implemented by Ulrich Drepper.
2709 * Faster sqrt and sqrtf implemention for some PPC variants.
2710 Implemented by Steven Munroe.
2714 * The following bugs are resolved with this release:
2716 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2717 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2718 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2719 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2720 5071, 5103, 5104, 5112, 5113, 5184, 5186
2722 * More checking functions: fread, fread_unlocked, open*, mq_open.
2723 Implemented by Jakub Jelinek and Ulrich Drepper.
2725 * Extend fortification to C++. Implemented by Jakub Jelinek.
2727 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2728 by not recognizing 'a' as a modifier when those specs are requested.
2729 Implemented by Jakub Jelinek.
2731 * PPC optimizations to math and string functions.
2732 Implemented by Steven Munroe.
2734 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2735 options to be passed. Implemented by Ulrich Drepper.
2737 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2739 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2740 Implemented by Ulrich Drepper.
2742 * Handle private futexes in the NPTL implementation.
2743 Implemented by Jakub Jelinek and Ulrich Drepper.
2745 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2746 Implemented by Roland McGrath and Ulrich Drepper.
2748 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2750 * SHA-256 and SHA-512 based password encryption.
2751 Implemented by Ulrich Drepper.
2753 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2754 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2757 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2758 Implemented by Ulrich Drepper.
2762 * The following bugs are resolved with this release:
2764 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2765 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2766 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2767 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2768 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2769 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2770 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2773 * New Linux interfaces: epoll_pwait, sched_getcpu.
2775 * New generic interfaces: strerror_l.
2777 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2781 * The following bugs are resolved with this release:
2783 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2784 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2785 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2786 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2787 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2788 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2789 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2790 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2791 3225, 3251, 3252, 3253, 3273
2793 * For Linux, the sorting of addresses returned by getaddrinfo now also
2794 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2795 Implemented by Ulrich Drepper.
2797 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2798 Implemented by Ulrich Drepper.
2800 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2802 * New iconv module for MIK. Contributed by Alexander Shopov.
2804 * For sites with broken group and/or passwd database, the auto-propagate
2805 option of nscd can prevent creating ID lookup entries from the results
2806 of a name lookup and vice versa. This usually is no problem but some
2807 site might have problems with the default behavior.
2808 Implemented by Ulrich Drepper.
2810 * Iterating over entire database in NIS can be slow. With the
2811 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2812 to trade time for memory. The entire database will be read at once.
2813 Implemented by Ulrich Drepper.
2815 * The interfaces introduced in RFC 3542 have been implemented by
2818 * Support for the new ELF hash table format was added by Ulrich Drepper.
2820 * Support for priority inheritance mutexes added by Jakub Jelinek and
2823 * Support for priority protected mutexes added by Jakub Jelinek.
2825 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2826 or_IN, csb_PL, fy_NL, sr_ME.
2830 * More overflow detection functions.
2832 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2833 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2835 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2836 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2837 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2838 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2839 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2840 by Masahide Washizawa.
2842 * It is now possible to install audit libraries for the dynamic linker using
2843 LD_AUDIT. Implemented by Ulrich Drepper.
2845 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2846 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2847 For a libc and libpthread that works well on Linux 2.4 kernels, we
2848 recommend using the stable 2.3 branch.
2850 * The new function open_wmemstream acts like open_memstream,
2851 but uses a wchar_t wide character buffer.
2853 * The new function ppoll is an improved version of poll, similar to pselect.
2855 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2856 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2857 renameat, unlinkat, linkat, symlinkat, readlinkat.
2859 * New Linux kernel system calls: unshare,
2860 inotify_init, inotify_add_watch, inotify_rm_watch.
2862 * The euidaccess function is now also known by the alias eaccess,
2863 for compatibility with some other systems.
2865 * Timezone data updated to 2006b version.
2869 * The following bugs are resolved with this release:
2871 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2872 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2873 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2874 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2875 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2876 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2878 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2880 * As of this release, GCC 4 can be used to compile the C Library.
2882 * Timezone data updated to 2005m version.
2886 * The following bugs are resolved with this release:
2888 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2889 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2890 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2891 777, 787, 821, 822, 823, 825
2893 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2897 * Support for RFC 3678. Real implementations exist only for Linux so far.
2898 Implemented by Ulrich Drepper.
2900 * nscd can now cache entries persistently. Expiring entries are reloaded.
2901 For speedups the cache can be shared in memory with client processes.
2902 Implemented by Ulrich Drepper.
2904 * nscd can now perform SELinux checks.
2905 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2907 * getaddrinfo queries are now cached. Canonical name lookup is performed
2909 Implemented by Ulrich Drepper.
2911 * The nothrow function attribute is used when headers are used by gcc when
2912 compiling C code. This can avoid significant amounts of exception
2915 * The malloc functions perform more error checking and are stricter when
2916 it comes to reacting on errors. The default action is to terminate
2917 the process after showing an error message. Implemented by Ulrich Drepper.
2919 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2920 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2922 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2923 namespaces using the new function dlmopen(). This feature is of course,
2924 like most other dynamic loading functionality, not available in statically
2925 linked applications. Implemented by Ulrich Drepper.
2927 * Low-overhead boundary checking variants of string and some stdio functions
2928 were added. These are to be used in conjunction with a gcc patch by
2929 Jakub Jelinek which adds calls to these functions if possible.
2930 Implemented by Jakub Jelinek and Ulrich Drepper.
2932 * Old code for several operating systems and machine architectures that
2933 have not been in working condition in a long time have been removed from
2934 the main source tree maintained by the GNU C Library's maintainers.
2935 These files are now reside in the separate `ports' source module
2936 that is usable as an add-on when building the library.
2940 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2941 interrogate the dynamic linker, compatible with the Solaris interface.
2943 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2944 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2946 * getifaddrs now uses the netlink interface on Linux to get its information.
2947 Implemented by Thorsten Kukuk.
2949 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2950 Implemented by Ulrich Drepper.
2952 * support for non-executable stacks on x86 has been added. Changes mostly
2955 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2958 * getaddrinfo now performs destination address selection according to
2963 * Thread-safe interfaces for many functions that access locale data
2964 were added in version 2.3, but these features were omitted from NEWS.
2965 Many functions have variants with an `_l' suffix that take a `locale_t'
2966 object as a parameter rather than consulting the current locale.
2967 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2968 create and maintain `locale_t' objects. Additionally, the new function
2969 `uselocale' sets "the current locale" (as used by functions not so
2970 parameterized) set for an individual thread. These features were added
2971 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2973 * The functions getresuid, getresgid, setresuid, and setresgid, which
2974 have long been available on Linux, are now declared in <unistd.h>
2975 and are now also available on the Hurd.
2977 * ELF thread-local storage support (TLS) now works on x86-64.
2979 * The new dynamic string token $LIB is expanded in shared library names.
2980 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2982 * Aldy Hernandez contributed complete software floating point support for
2983 PowerPC machines with no FPU.
2985 * fexecve is implemented on Linux.
2987 * The `btowc' function should work at least twice as fast due to
2988 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2990 * With approriate thread add-ons cancelable functions are now implemented
2991 in libc.so as well. No need to call the function in libpthread. This
2992 change allowed to finally disable the incorrect and expensive handling
2993 of weak definition in ld.so.
2995 * Yet more PLT entries in libc.so have been removed. We finally arrived
2996 at the bare minimum. Startup times improved appropriately.
2998 * Support for the new Linux/x86 system call interface was added. The
2999 AT_SYSINFO auxiliary vector entry is recognized and handled.
3003 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3006 * iconv (the program and the interface) now accepts empty names (excluding
3007 options like //TRANSLIT) to mean "use charset of current locale".
3009 * localedef can now transliterate characters in strings which are not in
3010 the provided charmap. The information from the input locale is used.
3012 * Prelinking support was added for ELF targets. This requires additional
3013 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3015 * Read-only stdio streams now use mmap to speed up operation by eliminating
3016 copying and buffer underflows. To use add 'm' to the mode string of
3017 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3019 * The malloc functions were completely rewritten by Wolfram Gloger based
3020 on Doug Lea's malloc-2.7.0.c.
3022 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3023 implementation of regex.
3025 * Bruno Haible upgraded the iconv and locale implementation to support
3028 * Contents of the LC_* and LANG environment variables in the CEN style are
3029 not recognized anymore. It never was used. Change by Ulrich Drepper.
3031 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3032 thread-local storage (TLS) ABI on some platforms.
3033 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3035 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3036 EUC-JISX0213, and TSCII.
3038 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3039 BSD-compatible interface for getting all network interface addresses.
3040 Implementation for IPv4 by Roland McGrath.
3042 * Loading of locale data is faster due to the introduction of a locale
3043 archive. Implemented by Roland McGrath and Ulrich Drepper.
3045 * Startup times are significantly reduced by not using exported functions
3046 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3049 * Steven Munroe contributed a port to PowerPC64/Linux.
3053 * The Hurd now uses the GNU libio implementation of stdio rather than the
3054 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3056 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3057 with the same behavior as the Linux system call of the same name.
3061 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3062 128-bit long double format.
3064 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3065 IBM1161, and IBM1162 charsets.
3067 * Andreas Jaeger contributed a port to x86-64/Linux.
3069 * Peter Bruin contributed a port to PowerPC/Hurd.
3071 * libc functions using I/O streams now can handle wide-oriented streams
3074 * optimizations in the dynamic linker. Binaries created by recent binutils
3075 versions start up quicker due to reduced time spent on relocations.
3077 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3081 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3082 asinh, atanh, j0 for the 128-bit long double format.
3084 * Bruno Haible updated all the code handling Unicode in some form to
3085 support Unicode 3.1.
3087 * Speed of regex for single-byte locales is back to previous levels.
3088 Patch by Isamu Hasegawa.
3090 * Alpha, SPARC, and IA-64 now also using floating stacks.
3092 * Startup time of internationalized applications greatly improved through
3093 iconv cache. Use iconvconfig to generate the cache file.
3094 Contributed by Ulrich Drepper.
3096 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3097 severe performance problems. Patch by David Mosberger.
3099 * The Hurd port got a lot more functionality like AIO, various stdio
3100 extensions, etc. Mainly done by Roland McGrath.
3102 * mtrace can now lookup symbols in shared libraries.
3106 * Intel's IA-64 math library is largely integrated. It provides fast and
3107 accurate implementatations for most basic and standard math functions
3108 in float, double, and long double format.
3110 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3111 and asin for the 96-bit long double format and asin, log, tan for the
3112 128-bit long double format.
3114 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3115 The basic double functions exist today. Contributed by Abraham Ziv
3116 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3117 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3119 * An asynchronous name lookup library was added. The interface is designed
3120 after POSIX AIO. The proposal was circulated beforehand to get comments.
3121 No negative ones came in. Implemented by Ulrich Drepper.
3123 * Port to S390/64bit contributed by Martin Schwidefsky
3124 <schwidefsky@de.ibm.com>.
3126 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3127 of functions for Linux/IA-64.
3129 * The RPC code is now thread safe. Threads can now use the same service
3130 of different services at the same time. Patch by Eric Norum
3131 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3133 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3134 family of functions for Linux/S390.
3136 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3137 of functions for Linux/x86.
3139 * Port to Linux/CRIS contributed by Axis Communications.
3143 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3144 we can now check for namespace violations and missing declarations. The
3145 result is that almost all headers are now Unix-compliant (as defined in
3146 the upcoming XPG6). The negative side is that some programs might need
3147 corrections, too, if they depend on the incorrect form of the headers in
3148 previous versions which defined too many symbols and included too many
3151 * regex now handles multibyte character sets correctly.
3152 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3154 * iconv (the program) does now conform to the upcoming XPG6 and handles
3155 charmaps. Instead of the charset names the path of charmaps can be
3156 provided and the conversion happens based on this data.
3157 Contributed by Ulrich Drepper.
3159 * The locale program now provides detailed information about the installed
3164 only lists the names of the supported locales
3168 provides details such as country, language, and codeset name.
3169 Contributed by Ulrich Drepper.
3173 * The gencat program now parses the input file according to the charset
3174 selected by the LC_CTYPE category. This is important for stateful
3175 character sets. To make generating catalogs easier there is a way
3176 to overwrite the charset selected by the locale: before the first
3177 message or $ quote line the catalog can contain a line like
3179 $ codeset=ISO-8859-2
3181 to select the charset (ISO-8859-2 in this case).
3183 Implemented by Shinya Hanataka and Ulrich Drepper.
3185 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3186 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3187 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3189 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3190 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3192 * The LANGUAGE environment variable is now ignored unless the locale is
3193 changed from the default "C" locale.
3195 * The usual bug fixes.
3199 * Greg McGary added runtime support for bounds checking using gcc's
3200 new -fbounded-pointers option. ix86 target is complete. PowerPC
3203 * Thorsten Kukuk added secure mode support to nscd.
3205 * The Berkeley db libraries have been removed.
3207 Related, the nss_db module is now in a separate package since it
3208 obviously requires a database library being available.
3210 * Wide character I/O streams implemented by Ulrich Drepper.
3212 * Functions from the extended socket API added by Ulrich Drepper.
3214 * Functions feenableexcept and fedisableexcept to control the
3215 behaviour of individual exceptions have been added by Andreas Jaeger.
3217 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3219 * The resolver code has been updated from bind 8.2.3-T5B which supports
3220 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3223 This change could in some situations effect backward compatibility. Since
3224 now `_res' is a thread-local instead of a global variable, modifying it
3225 in one thread does not have any effect in other threads.
3227 The resolver library was also extended to allow IPv6 as the transport
3228 protocol for the requests. This work was done by Stig Venaas.
3230 * Compatibility code for K&R C compilers has been removed from the
3231 header files. A ISO C compiler is needed to use the library
3232 (conforming to either C89 or C99 standard).
3234 * Complete rewrite of the localedef program to support multibyte character
3235 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3236 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3237 Implemented by Ulrich Drepper.
3239 Bruno Haible significantly improved the generation and use of the data
3240 structures for the wide character tables.
3242 * Plural handling in gettext implemented by Ulrich Drepper.
3244 * The utmp daemon has been removed.
3246 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3248 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3251 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3253 * POSIX barriers implemented by Kaz Kylheku.
3255 * POSIX spawn function family implemented by Ulrich Drepper.
3257 * POSIX spinlocks are now available.
3259 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3261 * the configure option --enable-kernel=X.Y.Z allows to strip out
3262 compatibility for kernel versions before X.Y.Z. This is currently only
3263 implemented for Linux.
3265 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3266 field sin6_scope_id. This means that all programs using IPv6 should be
3267 recompiled. Don't expect binary compatibility with previous glibc
3270 * various conversion modules for IBM character sets contributed by
3273 * IA-64 port by Jes Sorensen and HJ Lu.
3287 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3289 * New xdr functions are added; some rpc functions are now 64bit clean.
3291 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3293 * Fixed known incompatibilities with glibc 2.0.
3295 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3297 * Optimized a number of functions (especially the ELF dynamic loader).
3299 * Update timezone data files.
3301 * lots of charmaps corrections
3303 * some new locale definitions and charmaps
3308 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3309 conform to POSIX member type requirements. Symbol versions have been
3310 adjusted as needed within the library, and for direct use by applications,
3311 but there is potential for problems if third-party libraries use
3312 struct timeval as part of their interface. This does not present
3313 a problem for X and other "essential" system libraries.
3315 * An additional locale model to support C++ Standard Library locale
3316 model and probably more was implemented by Ulrich Drepper.
3318 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3321 * Miles Bader provided the `argp' function family to support hierachical
3322 command line argument parsing, layered on top of getopt.
3324 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3326 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3329 * scanf recognizes the %a and %A format for scanning floating point numbers.
3331 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3332 information and interfaces for the available integer types.
3334 * about 130 new math functions were added to implement the ISO C9x math
3337 * the new header <complex.h> contains definitions of the complex math
3338 functions from ISO C 9X.
3340 * the new header <tgmath.h> defines generic macros to use complex or
3341 real valued functions.
3343 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3345 * Andreas Jaeger provided a test suite for the math library.
3347 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3349 * Ulrich Drepper added character set conversion functions (iconv).
3351 * Optimized string functions have been added.
3353 * The localedata addon is now part of glibc.
3355 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3357 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3358 daemon for NSS (nscd).
3360 Missing a better place here are some numbers on improvements. Under
3361 Linux 2.1.125 un-tar-ing the kernel sources takes
3365 using local files 12.19s 6.88s 22.91s
3367 using NIS 13.92s 8.91s 26.34s
3369 using NIS & nscd 10.37s 7.34s 25.30s
3371 using NIS+ 27.57s 30.37s 640.46s
3373 using NIS+ & nscd 10.25s 7.83s 26.51s
3375 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3377 Keep in mind that non-namelookup related operations dominate above times.
3378 It was just a common complain that using NIS+ unpacking the kernel is
3381 [1] The old nscd implementation is not available anymore since it was
3382 distributed with glibc up to version 2.0.98 and thus is now replaced.
3384 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3386 * Mark Kettenis provided a Hesiod NSS module.
3388 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3389 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3391 * rcmd can now handle netgroups (Dick Streefland).
3393 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3396 * Support for the IPv6 protocol has been added to the socket API, as per the
3397 latest draft standards.
3399 * Support for Linux 2.2 has been added.
3401 * Interface changes relative to the latest 2.0.x release:
3402 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3403 addseverity NEW: Unix98
3404 alphasort64 NEW: LFS
3405 argp_err_exit_status NEW: argp, GNU ext
3406 argp_error NEW: argp, GNU ext
3407 argp_failure NEW: argp, GNU ext
3408 argp_help NEW: argp, GNU ext
3409 argp_parse NEW: argp, GNU ext
3410 argp_program_bug_address NEW: argp, GNU ext
3411 argp_program_version NEW: argp, GNU ext
3412 argp_program_version_hook NEW: argp, GNU ext
3413 argp_state_help NEW: argp, GNU ext
3414 argp_usage NEW: argp, GNU ext
3415 authdes_create NEW: Secure RPC
3416 authdes_getucred NEW: Secure RPC
3417 authdes_pk_create NEW: Secure RPC
3418 backtrace NEW: GNU ext.
3419 backtrace_symbols NEW: GNU ext.
3420 backtrace_symbols_fd NEW: GNU ext.
3422 cacosf NEW: ISO C 9x
3423 cacosh NEW: ISO C 9x
3424 cacoshf NEW: ISO C 9x
3425 cacoshl NEW: ISO C 9x
3426 cacosl NEW: ISO C 9x
3433 casinf NEW: ISO C 9x
3434 casinh NEW: ISO C 9x
3435 casinhf NEW: ISO C 9x
3436 casinhl NEW: ISO C 9x
3437 casinl NEW: ISO C 9x
3439 catanf NEW: ISO C 9x
3440 catanh NEW: ISO C 9x
3441 catanhf NEW: ISO C 9x
3442 catanhl NEW: ISO C 9x
3443 catanl NEW: ISO C 9x
3444 cbc_crypt NEW: Secure RPC
3448 ccoshf NEW: ISO C 9x
3449 ccoshl NEW: ISO C 9x
3455 cimagf NEW: ISO C 9x
3456 cimagl NEW: ISO C 9x
3457 clearerr_locked REMOVED
3458 clntunix_create NEW: sunrpc ext
3460 clog10 NEW: ISO C 9x
3461 clog10f NEW: ISO C 9x
3462 clog10l NEW: ISO C 9x
3472 cprojf NEW: ISO C 9x
3473 cprojl NEW: ISO C 9x
3475 crealf NEW: ISO C 9x
3476 creall NEW: ISO C 9x
3481 csinhf NEW: ISO C 9x
3482 csinhl NEW: ISO C 9x
3485 csqrtf NEW: ISO C 9x
3486 csqrtl NEW: ISO C 9x
3490 ctanhf NEW: ISO C 9x
3491 ctanhl NEW: ISO C 9x
3493 des_setparity NEW: Secure RPC
3494 ecb_crypt NEW: Secure RPC
3495 endutxent NEW: Unix98
3497 exp10f NEW: ISO C 9x
3498 exp10l NEW: ISO C 9x
3502 fattach NEW: STREAMS
3503 fdetach NEW: STREAMS
3507 feclearexcept NEW: ISO C 9x
3508 fegetenv NEW: ISO C 9x
3509 fegetexceptflag NEW: ISO C 9x
3510 fegetround NEW: ISO C 9x
3511 feholdexcept NEW: ISO C 9x
3513 feraiseexcept NEW: ISO C 9x
3514 ferror_locked REMOVED
3515 fesetenv NEW: ISO C 9x
3516 fesetexceptflag NEW: ISO C 9x
3517 fesetround NEW: ISO C 9x
3518 fetestexcept NEW: ISO C 9x
3519 feupdateenv NEW: ISO C 9x
3520 fflush_locked REMOVED
3524 fgets_unlocked NEW: GNU ext.
3525 fileno_locked REMOVED
3537 fputc_locked REMOVED
3538 fputs_unlocked NEW: GNU ext.
3539 fread_unlocked NEW: GNU ext.
3544 fstatvfs NEW: Unix98
3548 ftruncate64 NEW: LFS
3550 fwrite_unlocked NEW: GNU ext.
3551 gai_strerror NEW: IPv6
3555 getchar_locked REMOVED
3557 getdate_err NEW: Unix98
3558 getdate_r NEW: GNU ext.
3560 getnameinfo NEW: IPv6
3561 getnetname NEW: Secure RPC
3562 getpmsg NEW: STREAMS
3563 getpt NEW: Unix98 PTY
3564 getrlimit64 NEW: LFS
3565 getutxent NEW: Unix98
3566 getutxid NEW: Unix98
3567 getutxline NEW: Unix98
3568 glob64 NEW: GNU ext.
3569 globfree64 NEW: GNU ext.
3570 gnu_get_libc_release NEW: GNU ext.
3571 gnu_get_libc_version NEW: GNU ext.
3572 grantpt NEW: Unix98 PTY
3573 host2netname NEW: Secure RPC
3575 iconv_close NEW: iconv
3576 iconv_open NEW: iconv
3577 if_freenameindex NEW: IPv6
3578 if_indextoname NEW: IPv6
3579 if_nameindex NEW: IPv6
3580 if_nametoindex NEW: IPv6
3581 in6addr_any NEW: IPv6
3582 in6addr_loopback NEW: IPv6
3583 inet6_isipv4mapped NEW: IPv6
3584 isastream NEW: STREAMS
3585 iswblank NEW: GNU ext.
3586 key_decryptsession NEW: Secure RPC
3587 key_decryptsession_pk NEW: Secure RPC
3588 key_encryptsession NEW: Secure RPC
3589 key_encryptsession_pk NEW: Secure RPC
3590 key_gendes NEW: Secure RPC
3591 key_get_conv NEW: Secure RPC
3592 key_secretkey_is_set NEW: Secure RPC
3593 key_setnet NEW: Secure RPC
3594 key_setsecret NEW: Secure RPC
3595 llrint NEW: ISO C 9x
3596 llrintf NEW: ISO C 9x
3597 llrintl NEW: ISO C 9x
3598 llround NEW: ISO C 9x
3599 llroundf NEW: ISO C 9x
3600 llroundl NEW: ISO C 9x
3605 lrintf NEW: ISO C 9x
3606 lrintl NEW: ISO C 9x
3607 lround NEW: ISO C 9x
3608 lroundf NEW: ISO C 9x
3609 lroundl NEW: ISO C 9x
3611 makecontext NEW: Unix98
3612 mempcpy NEW: GNU ext.
3615 modify_ldt NEW: kernel
3619 nearbyint NEW: ISO C 9x
3620 nearbyintf NEW: ISO C 9x
3621 nearbyintl NEW: ISO C 9x
3622 netname2host NEW: Secure RPC
3623 netname2user NEW: Secure RPC
3624 nexttoward NEW: ISO C 9x
3625 nexttowardf NEW: ISO C 9x
3626 nexttowardl NEW: ISO C 9x
3630 passwd2des NEW: Secure RPC
3632 pow10f NEW: GNU ext.
3633 pow10l NEW: GNU ext.
3636 printf_size NEW: GNU ext.
3637 printf_size_info NEW: GNU ext.
3638 profil_counter REMOVED
3639 pthread_mutexattr_getkind_np REPLACED
3640 pthread_mutexattr_setkind_np REPLACED
3641 ptsname NEW: Unix98 PTY
3642 ptsname_r NEW: Unix98 PTY
3644 putchar_locked REMOVED
3645 putgrent NEW: GNU ext.
3647 putpmsg NEW: STREAMS
3648 pututxline NEW: Unix98
3652 readdir64_r NEW: LFS
3653 remquo NEW: ISO C 9x
3654 remquof NEW: ISO C 9x
3655 remquol NEW: ISO C 9x
3657 roundf NEW: ISO C 9x
3658 roundl NEW: ISO C 9x
3660 scalbln NEW: ISO C 9x
3661 scalblnf NEW: ISO C 9x
3662 scalblnl NEW: ISO C 9x
3664 sendfile NEW: kernel
3665 setcontext NEW: Unix98
3666 setrlimit64 NEW: LFS
3667 setutxent NEW: Unix98
3669 sigignore NEW: Unix98
3670 sigqueue NEW: POSIX.1b
3671 sigrelse NEW: Unix98
3672 sigset NEW: POSIX.1b
3673 sigtimedwait NEW: POSIX.1b
3674 sigwaitinfo NEW: POSIX.1b
3675 sincos NEW: GNU ext.
3676 sincosf NEW: GNU ext.
3677 sincosl NEW: GNU ext.
3681 strcasestr NEW: GNU ext.
3682 strtoimax NEW: ISO C 9x
3683 strtoumax NEW: ISO C 9x
3684 strverscmp NEW: GNU ext.
3685 svcauthdes_stats NEW: Secure RPC
3686 svcunix_create NEW: sunrpc etx
3687 svcunixfd_create NEW: sunrpc ext
3688 swapcontext NEW: Unix98
3689 tcgetsid NEW: Unix98 PTY
3690 tdestroy NEW: GNU ext.
3691 tgamma NEW: ISO C 9x
3692 tgammaf NEW: ISO C 9x
3693 tgammal NEW: ISO C 9x
3697 truncf NEW: ISO C 9x
3698 truncl NEW: ISO C 9x
3700 unlockpt NEW: Unix98 PTY
3701 updwtmpx NEW: Unix98
3702 user2netname NEW: Secure RPC
3703 utmpxname NEW: Unix98
3704 versionsort NEW: GNU ext.
3705 versionsort64 NEW: GNU ext.
3707 wcscasecmp NEW: GNU ext.
3708 wcsncasecmp NEW: GNU ext.
3709 wcsnlen NEW: GNU ext.
3710 wcstoimax NEW: ISO C 9x
3711 wcstoll NEW: ISO C 9x
3712 wcstoull NEW: ISO C 9x
3713 wcstoumax NEW: ISO C 9x
3715 wordexp NEW: POSIX.2
3716 wordfree NEW: POSIX.2
3717 write_profiling REMOVED
3718 xdecrypt NEW: Secure RPC
3719 xdr_authdes_cred NEW: Secure RPC
3720 xdr_authdes_verf NEW: Secure RPC
3721 xdr_cryptkeyarg NEW: Secure RPC
3722 xdr_cryptkeyarg2 NEW: Secure RPC
3723 xdr_cryptkeyres NEW: Secure RPC
3724 xdr_getcredres NEW: Secure RPC
3725 xdr_key_netstarg NEW: Secure RPC
3726 xdr_key_netstres NEW: Secure RPC
3727 xdr_keybuf NEW: Secure RPC
3728 xdr_keystatus NEW: Secure RPC
3729 xdr_netnamestr NEW: Secure RPC
3730 xdr_sizeof NEW: Secure RPC
3731 xdr_unixcred NEW: sunrpc ext
3732 xencrypt NEW: Secure RPC
3733 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3744 * inet_ntoa is thread-safe
3746 * updwtmp is moved from libutil to libc
3748 * rewrite of cbrt function
3750 * update of timezone data
3764 * add atoll function
3766 * fix complex problems in Berkeley DB code
3768 * fix math functions
3772 * fixed lots of header problems (especially Linux/GNU specific)
3774 * dynamic loader preserves all registers
3776 * Roland McGrath provided support for handling of auxiliary objects in
3777 the ELF dynamic loader.
3779 * support for parallel builds is improved
3783 * GNU extensions are no longer declared by default. To enable them you
3784 must define the macro `_GNU_SOURCE' in your program or compile with
3787 * The library has changed from using GNU ld symbol aliases to using weak
3788 symbols where available. The ELF object file format supports weak
3789 symbols; GNU ld also supports weak symbols in the a.out format. (There
3790 is also now support for other GNU ld extensions in ELF. Use the
3791 `--with-elf' option to configure to indicate you have ELF, and
3792 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3793 of many files which contained only symbol aliases, reducing the size of
3794 the source and the compiled library; many other files were renamed to
3795 less cryptic names previously occupied by the symbol alias files.
3796 There is a new header file <elf.h> for programs which operate on
3797 files in the ELF format.
3799 * Converted to Autoconf version 2, so `configure' has more options.
3800 Run `configure --help' to see the details.
3802 * The library can now be configured to build profiling, highly-optimized
3803 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3804 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3805 `configure' enable building these extra libraries. The shared library is
3806 built by default when using both ELF and GNU ld. When shared libraries
3807 are enabled, the new library `-ldl' is available for arbitrary run-time
3808 loading of shared objects; its interface is defined in <dlfcn.h>. The
3809 new header file <link.h> gives access to the internals of the run-time
3810 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3811 application of same name on other systems and it provides information
3812 about dynamically linked binaries.
3814 * The C library now provides the run-time support code for profiling
3815 executables compiled with `-pg'. Programs can control the profiling code
3816 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3817 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3818 the support for this file format was contributed by David Mosberger-Tang.
3820 * The math code has been replaced with a math library based on fdlibm from
3821 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3822 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3823 functions. The math functions now reside in a separate library, so
3824 programs using them will need to use `-lm' their linking commands.
3826 * John C. Bowman contributed optimized ix87 assembler inline functions.
3828 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3829 mechanism similar to that found in Solaris 2. This is now used for the
3830 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3831 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3832 file controls what services are used for each individual database. This
3833 works by loading shared libraries with names specified in `nsswitch.conf',
3834 so service modules can be changed or added at any time without even
3835 relinking any program. Currently there are the file, db, and NIS based
3836 NSS services available.
3838 * The new functions `strtoq' and `strtouq' parse integer values from
3839 strings, like `strtol' and `strtoul', but they return `long long int' and
3840 `unsigned long long int' values, respectively (64-bit quantities).
3842 * The new functions `strtof' and `strtold' parse floating-point values from
3843 strings, like `strtod', but they return `float' and `long double' values,
3844 respectively (on some machines `double' and `long double' are the same).
3846 * Ulrich Drepper has contributed new implementations of the floating-point
3847 printing and reading code used in the `printf' family of functions and
3848 `strtod', `strtof', and `strtold'. These new functions are perfectly
3849 accurate, and much faster than the old ones.
3851 * The implementation of the POSIX locale model was completely rewritten by
3852 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3853 to compile the POSIX locale definition.
3855 * The former dummy implementations of the strcoll and strxfrm function are
3856 now replaced by fully functional code contributed by Ulrich Drepper. The
3857 collation information comes from the POSIX locale definitions.
3859 * The new header <langinfo.h> defines an interface for accessing
3860 various locale-dependent data (using the locale chosen with `setlocale').
3862 * Ulrich Drepper has contributed a new suite of functions for operation on
3863 wide-character and multibyte-character strings, in <wchar.h>;
3864 and classification and case conversion of wide characters, in <wctype.h>.
3865 These new functions are conforming to the ISO C, Amendement 1 specification.
3867 * There is now a second implementation of the standard I/O library available.
3868 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3869 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3871 * You can now use positional parameter specifications in format strings
3872 for the `printf' and `scanf' families of functions. For example,
3873 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3874 ``Number 6, Mr Jones''. This is mainly useful when providing different
3875 format strings for different languages, whose grammars may dictate
3876 different orderings of the values being printed. To support this
3877 feature, the interface for `register_printf_handler' has changed; see
3878 the header file <printf.h> for details.
3880 * The `printf' and `scanf' families of functions now understand a new
3881 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3882 to group numbers as indicated by the locale; for `scanf' and friends, this
3883 says to accept as valid only a number with all the proper grouping
3884 separators in the right places. In the default "C" locale, numbers are
3885 not grouped; but locales for specific countries will define the usual
3886 conventions (i.e. separate thousands with `,' in the US locale).
3888 * The pgrp functions have been regularized, slightly incompatibly but much
3889 less confusingly. The core functions are now `getpgid' and `setpgid',
3890 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3891 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3892 provided for compatibility. There is no longer an incompatible `getpgrp'
3893 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3895 * The new header file <fts.h> and suite of functions simplify programs that
3896 operate on directory trees. This code comes from 4.4 BSD.
3898 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3899 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3900 NSS scheme used in glibc.
3902 * There is a new malloc debugging hook `__memalign_hook'.
3904 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3905 `unsigned int' in <sys/types.h>. These are for compatibility only and
3906 their use is discouraged.
3908 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3909 done differently, so that it works even without GNU ld.
3911 * New function `euidaccess' checks allowed access to a file like `access',
3912 but using the effective IDs instead of the real IDs.
3914 * The time zone data files have been updated for the latest and greatest
3915 local time conventions of the countries of the world.
3917 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3920 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3921 for formatting floating-point numbers. They are provided only for
3922 compatibility; new programs should use `sprintf' instead. There are
3923 also equivalent function for the `long double' floating-point type and
3924 all functions also exist in a reentrant form.
3926 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3927 functions for maintaining the login-record files (primarily of use to
3928 system programs such as `login'), and convenient functions for
3929 allocating and initializing a pseudo-terminal (pty) device.
3931 * Ulrich Drepper has contributed new support for System V style
3932 shared memory and IPC on systems that support it.
3934 * Ulrich Drepper has contributed several miscellaneous new functions found
3935 in System V: The `hsearch' family of functions provide an effective
3936 implementation of hash tables; `a64l' and `l64a' provide a very simple
3937 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3940 * Ulrich Drepper has contributed new reentrant counterparts for the
3941 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3943 * Ulrich Drepper has contributed new, highly-optimized versions of several
3944 string functions for the i486/Pentium family of processors.
3946 * Ulrich Drepper has updated the Linux-specific code, based largely
3947 on work done in Hongjiu Lu's version of GNU libc for Linux.
3948 The GNU library now supports Linux versions 2.0.10 and later,
3949 using the ELF object file format (i[3456]86-*-linux).
3951 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3953 * David Mosberger-Tang and Richard Henderson have ported the C library
3954 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3955 dynamic linking support for ELF/Alpha.
3957 * Richard Henderson contributed several Alpha optimized assembler function
3958 for arithmetic and string handling.
3960 * Ulrich Drepper has contributed a new set of message catalog functions to
3961 support multiple languages using the <libintl.h> interface, for use with
3962 his new package GNU gettext. Translation volunteers have contributed
3963 catalogs of the library's messages in Spanish, German, and Korean.
3965 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3966 program and the `catgets' function for reading the catalog files it
3967 creates. (The <libintl.h> interface is preferred; we include the
3968 <nl_types.h> interface using `catgets' only for source compatibility with
3969 programs already written to use it.)
3971 * New header file <values.h> gives SVID-compatible names for <limits.h>
3974 * Various new macros, declarations, and small header files for compatibility
3977 * New function `group_member' is a convenient way to check if a process has
3978 a given effective group ID.
3980 * When using GCC 2.7 and later, the socket functions are now declared in a
3981 special way so that passing an argument of type `struct sockaddr_in *',
3982 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3983 `struct sockaddr *' type, does not generate a type-clash warning.
3985 * New function `error' declared in header file <error.h> is a convenient
3986 function for printing error messages and optionally exiting; this is the
3987 canonical function used in GNU programs. The new functions `err', `warn',
3988 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3989 doing the same thing.
3991 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3992 POSIX.2 `glob' function to do ~ and {...} expansion.
3994 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3995 `clearenv' which is used in POSIX.9 is also available.
3997 * New function `getsid' returns session ID number on systems that support it.
3999 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4000 files <db.h> and <mpool.h> provide a rich set of functions for several
4001 types of simple databases stored in memory and in files, and <ndbm.h> is
4002 an old `ndbm'-compatible interface using the `db' functions. Link with
4003 `-ldb' to get these functions.
4005 * New macro `strdupa' copies a string like `strdup', but uses local stack
4006 space from `alloca' instead of dynamic heap space from `malloc'.
4008 * New function `strnlen' is like `strlen' but searches only a given maximum
4009 number of characters for the null terminator. `stpncpy', `strndup' and
4010 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4013 * New function `statfs' in header <sys/statfs.h>.
4015 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4016 provide convenient functions for operating on blocks of null-terminated
4019 * A new suite of functions in <utmp.h> handle all the details of reading
4020 and writing the utmp file.
4022 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4025 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4026 completely POSIX compliant and also implemented the extended functionality
4027 to handle alternate digit representation and alternate era date formats.
4029 * Ulrich Drepper provided an implementation of the `strptime' function
4030 defined in XPG4.2 which transforms a string into a `struct tm' value.
4032 * Paul Eggert provided the tzselect shell script as part of the timezone
4033 code. The shell script makes it easy to select the correct timezone
4036 * The implementation of the malloc family of functions is completely replaced
4037 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4038 The implementation uses the mmap function (if available) and it is
4039 optimized for the use in multi threaded programs.
4041 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4042 functions. This new functionality is usable by specifying a special
4043 salt string and it is compatible with implementation on *BSD systems.
4045 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4046 `getsubopt' to handle second level command line options, `bsd_signal'
4047 to access BSD style `signal' functionality, the obsolete `regexp' style
4050 * the `lchown' function is available on system which support this
4053 * The implementation of the shadow password handling function was contributed
4056 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4058 * POSIX.1g support was added. The <sys/select.h> header is available,
4059 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4060 implementation of `getaddrinfo'.
4064 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4066 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4067 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4069 * The new function `daemon' from 4.4 BSD is useful for server programs that
4070 want to put themselves in the background.
4072 * Joel Sherrill has contributed support for several standalone boards that
4073 run without an operating system.
4075 * `printf', `scanf' and friends now accept a `q' type modifier for long
4076 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4078 * All of the code taken from BSD (notably most of the math and networking
4079 routines) has been updated from the BSD 4.4-Lite release.
4081 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4083 * The new functions `getdomainname' and `setdomainname' fetch or change the
4084 YP/NIS domain name. These are system calls which exist on systems which
4087 * The time zone data files have been updated for the latest international
4090 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4091 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4095 * The C library now includes support for Sun RPC, from Sun's free
4096 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4097 are included. (There is still no support for YP.)
4099 * Tom Quinn has contributed a port of the C library to SGI machines running
4100 Irix 4 (mips-sgi-irix4).
4102 * The new `lockf' function is a simplified interface to the locking
4103 facilities of `fcntl', included for compatibility.
4105 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4107 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4109 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4112 * The error message printed by `assert' for a failed assertion now includes
4113 the name of the program (if using GNU ld) and the name of the calling
4114 function (with versions of GCC that support this).
4116 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4118 * The library now includes the <sys/mman.h> header file and memory
4119 management functions `mmap', `munmap', `mprotect', `msync', and
4120 `madvise', on systems that support those facilities.
4122 * The interface for `mcheck' has changed slightly: the function called to
4123 abort the program when an allocation inconsistency is detected now takes
4124 an argument that indicates the type of failure. The new function
4125 `mprobe' lets you request a consistency check for a particular block at
4126 any time (checks are normally done only when you call `free' or `realloc'
4129 * It is now possible to easily cross-compile the C library, building on one
4130 system a library to run on another machine and/or operating system. All
4131 you need to do is set the variable `HOST_CC' in `configparms' to the
4132 native compiler for programs to run on the machine you are building on (a
4133 few generator programs are used on Unix systems); set `CC' to the
4136 * The new function `fexecve' (only implemented on the GNU system) executes
4137 a program file given a file descriptor already open on the file.
4141 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4142 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4144 * You can set the variable `libprefix' in `configparms' to specify a prefix
4145 to be prepended to installed library files; this makes it easy to install
4146 the GNU C library to be linked as `-lgnuc' or whatever.
4148 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4149 copies a limited number of characters from a string, and returns the
4150 address of the last character written.
4152 * You no longer need to check for whether the installed `stddef.h' is
4153 compatible with the GNU C library. configure now checks for you.
4155 * You can now define a per-stream `fileno' function to convert the
4156 stream's cookie into an integral file descriptor.
4158 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4159 allocates zero bytes of storage, and returns a unique pointer which
4160 you can pass to `realloc' or `free'. The behavior is undefined if
4161 you dereference this pointer.
4163 * The C library now runs on Sony NEWS m68k machines running either
4164 NewsOS 3 or NewsOS 4.
4166 * The new `syscall' function is a system-dependent primitive function
4167 for invoking system calls. It has the canonical behavior on Unix
4168 systems, including unreliable return values for some calls (such as
4169 `pipe', `fork' and `getppid').
4171 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4172 to `EAGAIN', which is the preferred name. On systems whose kernels
4173 use two distinct codes, the C library now translates EWOULDBLOCK to
4174 EAGAIN in every system call function.
4178 * The GNU C Library Reference Manual is now distributed with the library.
4179 `make dvi' will produce a DVI file of the printed manual.
4180 `make info' will produce Info files that you can read on line using C-h i
4181 in Emacs or the `info' program.
4182 Please send comments on the manual to bug-glibc-manual@gnu.org.
4184 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4186 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4188 * Jason Merrill has contributed a port to the Sequent Symmetry running
4189 Dynix version 3 (i386-sequent-dynix).
4191 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4192 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4194 * New function `memory_warnings' lets you arrange to get warnings when
4195 malloc is running out of memory to allocate, like Emacs gives you.
4197 * The C library now contains the relocating allocator used in Emacs 19 for
4198 its editing buffers. This allocator (ralloc) minimizes allocation
4199 overhead and fragmentation by moving allocated regions around whenever it
4200 needs to. You always refer to a ralloc'd region with a "handle" (a
4201 pointer to a pointer--an object of type `void **').
4203 * There is a new `printf' format: `%m' gives you the string corresponding
4204 to the error code in `errno'.
4206 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4207 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4208 pass, the `a' modifier says to fill in a `char **' you pass with a
4211 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4212 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4213 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4215 * `mkstemp' is a traditional Unix function to atomically create and open a
4216 uniquely-named temporary file.
4220 * The standard location for the file that says what the local timezone is
4221 has changed again. It is now `/usr/local/etc/localtime' (or more
4222 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4224 * The distribution no longer contains any files with names longer than 14
4227 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4228 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4230 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4234 * The configuration process has changed quite a bit. The `configure'
4235 script is now used just like the configuration scripts for other GNU
4236 packages. The `sysdeps' directory hierarchy is much rearranged.
4237 The file `INSTALL' explains the new scheme in detail.
4239 * The header files no longer need to be processed into ANSI C and
4240 traditional C versions. There is just one set of files to install, and
4241 it will work with ANSI or old C compilers (including `gcc -traditional').
4243 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4244 MIPS DECStation running Ultrix 4.
4246 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4247 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4248 made itself into a shared library.
4250 * Yet further improved support for the i386, running 4.3 BSD-like systems
4251 (such as Mach 3 with the Unix single-server), or System V.
4253 * New function `strncasecmp' to do case-insensitive string comparison
4254 with limited length.
4256 * New function `strsep' is a reentrant alternative to `strtok'.
4258 * New functions `scandir' and `alphasort' for searching directories.
4260 * New function `setenv' is a better interface to `putenv'.
4262 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4263 function for traversing a directory tree.
4265 * The GNU obstack package is now also part of the C library.
4266 The new function `open_obstack_stream' creates a stdio stream that
4267 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4268 formatted output directly to an obstack.
4270 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4271 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4273 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4275 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4276 things to your strings.
4278 * There are some new test programs: test-fseek, testmb, and testrand.
4280 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4281 These ports are not finished, but are a good starting place for really
4282 supporting those systems.
4284 * `/etc/localtime' is now the standard location for the file that says what
4285 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4286 This follows the general principle that `/etc' is the place for all local
4287 configuration files.
4289 * The C library header files now use `extern "C"' when used by the C++
4290 compiler, so the C library should now work with C++ code.
4292 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4293 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4296 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4297 function which uses the `merge sort' algorithm, and is said to be
4298 significantly faster than the old GNU `qsort' function. Merge sort is
4299 now the standard `qsort' function. The new algorithm can require a lot
4300 of temporary storage; so, the old sorting function is called when the
4301 required storage is not available.
4303 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4304 provides the Unix `crypt' function, plus some other entry points.
4306 * The code and header files taken from 4.4 BSD have been updated with the
4307 latest files released from Berkeley.
4309 ----------------------------------------------------------------------
4310 Copyright information:
4312 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4314 Permission is granted to anyone to make or distribute verbatim copies
4315 of this document as received, in any medium, provided that the
4316 copyright notice and this permission notice are preserved,
4317 thus giving the recipient permission to redistribute in turn.
4319 Permission is granted to distribute modified versions
4320 of this document, or of portions of it,
4321 under the above conditions, provided also that they
4322 carry prominent notices stating who last changed them.
4325 version-control: never