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 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
114 and will be removed in a future release. Software that is still using
115 either header should be updated to use standard <stdio.h> interfaces
118 libio.h was originally the header for a set of supported GNU extensions,
119 but they have not been maintained as such in many years, they are now
120 standing in the way of improvements to stdio, and we don't think there are
121 any remaining external users. _G_config.h was never intended for public
122 use, but predates the bits convention.
124 Changes to build and runtime requirements:
126 * bison version 2.7 or later is required to generate code in the 'intl'
129 Security related changes:
131 CVE-2009-5064: The ldd script would sometimes run the program under
132 examination directly, without preventing code execution through the
133 dynamic linker. (The glibc project disputes that this is a security
134 vulnerability; only trusted binaries must be examined using the ldd
137 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
138 suffered from a one-byte overflow during ~ operator processing (either
139 on the stack or the heap, depending on the length of the user name).
140 Reported by Tim Rühsen.
142 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
143 would sometimes fail to free memory allocated during ~ operator
144 processing, leading to a memory leak and, potentially, to a denial
147 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
148 without GLOB_NOESCAPE, could write past the end of a buffer while
149 unescaping user names. Reported by Tim Rühsen.
151 CVE-2017-17426: The malloc function, when called with an object size near
152 the value SIZE_MAX, would return a pointer to a buffer which is too small,
153 instead of NULL. This was a regression introduced with the new malloc
154 thread cache in glibc 2.26. Reported by Iain Buclaw.
156 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
157 to the allocation of too much memory. (This is not a security bug per se,
158 it is mentioned here only because of the CVE assignment.) Reported by
161 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
162 of the number of search path components. (This is not a security
163 vulnerability per se because no trust boundary is crossed if the fix for
164 CVE-2017-1000366 has been applied, but it is mentioned here only because
165 of the CVE assignment.) Reported by Qualys.
167 The following bugs are resolved with this release:
169 [The release manager will add the list generated by
170 scripts/list-fixed-bugs.py just before the release.]
177 * A per-thread cache has been added to malloc. Access to the cache requires
178 no locks and therefore significantly accelerates the fast path to allocate
179 and free small amounts of memory. Refilling an empty cache requires locking
180 the underlying arena. Performance measurements show significant gains in a
181 wide variety of user workloads. Workloads were captured using a special
182 instrumented malloc and analyzed with a malloc simulator. Contributed by
183 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
185 * Unicode 10.0.0 Support: Character encoding, character type info, and
186 transliteration tables are all updated to Unicode 10.0.0, using
187 generator scripts contributed by Mike FABIAN (Red Hat).
188 These updates cause user visible changes, especially the changes in
189 wcwidth for many emoji characters cause problems when emoji sequences
190 are rendered with pango, see for example:
191 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
193 * Collation of Hungarian has been overhauled and is now consistent with "The
194 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
197 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
199 - The GNU C Library will now detect when /etc/resolv.conf has been
200 modified and reload the changed configuration. The new resolver option
201 “no-reload” (RES_NORELOAD) disables this behavior.
203 - The GNU C Library now supports an arbitrary number of search domains
204 (configured using the “search” directive in /etc/resolv.conf);
205 previously, there was a hard limit of six domains. For backward
206 compatibility, applications that directly modify the ‘_res’ global
207 object are still limited to six search domains.
209 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
210 Library will now randomly pick a name server from the configuration as a
211 starting point. (Previously, the second name server was always used.)
213 * The tunables feature is now enabled by default. This allows users to tweak
214 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
216 * New function reallocarray, which resizes an allocated block (like realloc)
217 to the product of two sizes, with a guaranteed clean failure upon integer
218 overflow in the multiplication. Originally from OpenBSD, contributed by
219 Dennis Wölfing and Rüdiger Sonderfeld.
221 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
222 These are extended versions of preadv and pwritev, respectively, taking an
223 additional flags argument. The set of supported flags depends on the
224 running kernel; full support currently requires kernel 4.7 or later.
226 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
227 create a new session ID for the spawned process. This feature is
228 scheduled to be added to the next major revision of POSIX; for the time
229 being, it is available under _GNU_SOURCE.
231 * errno.h is now safe to use from C-preprocessed assembly language on all
232 supported operating systems. In this context, it will only define the
233 Exxxx constants, as preprocessor macros expanding to integer literals.
235 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
236 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
237 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
238 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
240 To compile programs that use this feature, the compiler must support
241 128-bit floating point with the type name _Float128 (as defined by TS
242 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
243 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
244 must be defined to make the new interfaces visible.
246 The new functions and macros correspond to those present for other
247 floating-point types (except for a few obsolescent interfaces not
248 supported for the new type), with F128 or f128 suffixes; for example,
249 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
250 printf or scanf formats for the new type; the strfromf128 and strtof128
251 interfaces should be used instead.
253 Deprecated and removed features, and other changes affecting compatibility:
255 * The synchronization that pthread_spin_unlock performs has been changed to
256 now be equivalent to a C11 atomic store with release memory order to the
257 spin lock's memory location. Previously, several (but not all)
258 architectures used stronger synchronization (e.g., containing what is
259 often called a full barrier). This change can improve performance, but
260 may affect odd fringe uses of spin locks that depend on the previous
261 behavior (e.g., using spin locks as atomic variables to try to implement
262 Dekker's mutual exclusion algorithm).
264 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
267 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
268 will only be built and installed when the GNU C Library is configured with
269 --enable-obsolete-rpc. This allows alternative RPC implementations, such
270 as TIRPC or rpcsvc-proto, to be used.
272 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
273 libnss_compat, are deprecated, and will not be built or installed by
276 The NIS(+) support library, libnsl, is also deprecated. By default, a
277 compatibility shared library will be built and installed, but not headers
278 or development libraries. Only a few NIS-related programs require this
279 library. (In particular, the GNU C Library has never required programs
280 that use 'gethostbyname' to be linked with libnsl.)
282 Replacement implementations based on TIRPC, which additionally support
283 IPv6, are available from <https://github.com/thkukuk/>. The configure
284 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
285 name service modules, to be built and installed.
287 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
288 support is enabled, the configured recursive resolver must support EDNS.
289 (Responding to EDNS-enabled queries with responses which are not
290 EDNS-enabled is fine, but FORMERR responses are not.)
292 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
293 servers have not supported this opcode for a long time.
295 * The _res_opcodes variable has been removed from libresolv. It had been
296 exported by accident.
298 * <string.h> no longer includes inline versions of any string functions,
299 as this kind of optimization is better done by the compiler. The macros
300 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
302 * The nonstandard header <xlocale.h> has been removed. Most programs should
303 use <locale.h> instead. If you have a specific need for the definition of
304 locale_t with no other declarations, please contact
305 libc-alpha@sourceware.org and explain.
307 * The obsolete header <sys/ultrasound.h> has been removed.
309 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
311 * The obsolete function cfree has been removed. Applications should use
314 * The stack_t type no longer has the name struct sigaltstack. This changes
315 the C++ name mangling for interfaces involving this type.
317 * The ucontext_t type no longer has the name struct ucontext. This changes
318 the C++ name mangling for interfaces involving this type.
320 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
321 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
322 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
323 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
324 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
325 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
326 fpregset_t type no longer has the name struct fpu. This changes the C++
327 name mangling for interfaces involving those types.
329 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
330 synced with the kernel:
332 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
333 are not supported on this architecture and have been removed.
335 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
336 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
337 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
339 Programs that assume the GET/SETREGS ptrace requests are universally
340 available will now fail to build, instead of malfunctioning at runtime.
342 Changes to build and runtime requirements:
344 * Linux kernel 3.2 or later is required at runtime, on all architectures
345 supported by that kernel. (This is a change from version 2.25 only for
348 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
350 * On most architectures, GCC 4.9 or later is required to build the GNU C
351 Library. On powerpc64le, GCC 6.2 or later is required.
353 Older GCC versions and non-GNU compilers are still supported when
354 compiling programs that use the GNU C Library. (We do not know exactly
355 how old, and some GNU extensions to C may be _de facto_ required. If you
356 are interested in helping us make this statement less vague, please
357 contact libc-alpha@sourceware.org.)
359 Security related changes:
361 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
362 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
364 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
365 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
367 * Avoid printing a backtrace from the __stack_chk_fail function since it is
368 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
371 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
372 fixed (CVE-2017-12133).
374 The following bugs are resolved with this release:
376 [984] network: Respond to changed resolv.conf in gethostbyname
377 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
378 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
380 [12189] libc: __stack_chk_fail should not attempt a backtrace
382 [14096] time: Race condition on timezone/tst-timezone.out
383 [14172] localedata: az_IR: new locale
384 [14995] build: glibc fails to build if gold is the default linker, even if
386 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
387 [16637] network: inet_pton function is accepting IPv6 with bad format
388 [16640] string: string/strtok.c: undefined behaviour inconsistent between
389 x86 and other generic code
390 [16875] localedata: ko_KR: fix lang_name
391 [17225] localedata: ar_SY: localized month names for May and June are
393 [17297] localedata: da_DK: wrong date_fmt string
394 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
395 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
396 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
397 [19066] localedata: ar_SA abbreviated day and month names are in English
398 [19569] network: resolv: Support an arbitrary number of search domains
399 [19570] network: Implement random DNS server selection in the stub
401 [19838] locale: localedef fails on PA-RISC
402 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
403 order of 0D36 and 0D37
404 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
406 [20098] libc: FAIL: debug/backtrace-tst on hppa
407 [20257] network: sunrpc: clntudp_call does not enforce timeout when
409 [20275] localedata: locale day/abday/mon/abmon should not have trailing
411 [20313] localedata: Update locale data to Unicode 9.0
412 [20424] manual: Document how to provide a malloc replacement
413 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
415 [20686] locale: Add el_GR@euro to SUPPORTED.
416 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
417 failures consistently
418 [21015] dynamic-link: Document and fix --enable-bind-now
419 [21016] nptl: pthread_cond support is broken on hppa
420 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
421 frame-pointer on i386
422 [21049] libc: segfault in longjmp_chk() due to clobbered processor
424 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
425 [21088] libc: Build fails with --enable-static-nss
426 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
427 generic c code is used
428 [21109] libc: Tunables broken on big-endian
429 [21112] math: powf has large ulp errors with base close to 1 and exponent
431 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
433 [21120] malloc: glibc malloc is incompatible with GCC 7
434 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
436 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
437 (-0.0) when linking with -lieee
438 [21171] math: log10, log2 and lgamma return incorrect results
439 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
440 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
441 [21207] localedata: ce_RU: update weekdays from CLDR
442 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
443 [21217] localedata: Update months from CLDR-31
444 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
445 [21243] libc: support_delete_temp_file should issue warning for failed
447 [21244] libc: support resolv_test_start() socket fd close should be
449 [21253] libc: localedef randomly segfaults when using -fstack-check due to
450 new posix_spawn implementation
451 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
452 leads to lower CPU frequency
453 [21259] libc: [alpha] termios.h missing IXANY for POSIX
454 [21261] libc: [sparc64] bits/setjmp.h namespace
455 [21267] network: [mips] bits/socket.h IOC* namespace
456 [21268] libc: [alpha] termios.h NL2, NL3 namespace
457 [21270] libc: mmap64 silently truncates large offset values
458 [21275] libc: posix_spawn always crashes on ia64 now
459 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
460 [21280] math: [powerpc] logbl for POWER7 return incorrect results
461 [21289] libc: Incorrect declaration for 32-bit platforms with
462 _FILE_OFFSET_BITS=64 causes build error
463 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
464 not support gethostbyname4_r
465 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
467 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
469 [21340] libc: Support POSIX_SPAWN_SETSID
470 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
471 [21359] network: ns_name_pack needs additional byte in destination buffer
472 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
473 against fragmentation attacks (CVE-2017-12132)
474 [21369] network: resolv: Remove EDNS fallback
475 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
477 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
478 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
479 [21393] stdio: Missing dup3 error check in freopen, freopen64
480 [21396] libc: Use AVX2 memcpy/memset on Skylake server
481 [21399] localedata: Bad description for U00EC in
482 localedata/charmaps/CP1254
483 [21411] malloc: realloc documentation error
484 [21426] network: sys/socket.h uio.h namespace
485 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
486 [21445] libc: signal.h bsd_signal namespace
487 [21455] network: Network headers stdint.h namespace
488 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
490 [21475] network: resolv: Overlong search path is truncated mid-label
491 [21511] libc: sigstack namespace
492 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
493 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
494 bits/syscall.h: No such file or directory
495 [21517] libc: struct sigaltstack namespace
496 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
497 [21533] localedata: Update locale data to Unicode 10.0
499 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
500 junk at end of line, first unrecognized character is `@'
501 [21538] libc: SIG_HOLD missing for XPG4
502 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
503 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
504 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
505 [21543] libc: sigevent namespace
506 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
507 size for general purpose registers in mcontext_t structure
508 [21550] libc: sigwait namespace
509 [21552] libc: XPG4 bsd_signal namespace
510 [21554] libc: sigpause namespace
511 [21560] libc: sys/wait.h signal.h namespace
512 [21561] libc: waitid namespace
513 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
514 '/usr/include/stdlib.h': Permission denied
515 [21575] libc: sys/wait.h missing struct rusage definition
516 [21584] libc: sigaltstack etc namespace
517 [21597] libc: siginterrupt namespace
518 [21607] math: hppa: FAIL: math/test-tgmath
519 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
521 [21622] libc: [tile] missing SA_* for POSIX.1:2008
522 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
523 stack and heap (CVE-2017-1000366)
524 [21625] libc: wait3 namespace
525 [21654] nss: Incorrect pointer alignment in NSS group merge result
527 [21657] network: Parse interface zone id for node-local multicast
528 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
530 [21666] libc: .symver is used on common symbol
531 [21668] network: resolv: res_init cross-thread broadcast introduces race
533 [21687] math: tgmath.h totalorder, totalordermag return type
534 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
536 [21696] libc: Incorrect assumption of of __cpu_mask in
537 posix/sched_cpucount.c
538 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
539 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
541 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
542 leading to relocation crash
543 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
544 `resolv_conf_matches (resp, conf)' failed.
545 [21710] localedata: Added Samoan language locale for Samoa
546 [21711] localedata: Pashto yesstr/nostr locale are missing
547 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
548 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
549 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
550 [21723] localedata: yesstr/nostr missing for Chinese language locale
551 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
553 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
555 [21728] localedata: New Locale for Tongan language
556 [21729] localedata: incorrect LC_NAME fields for hi_IN
557 [21733] localedata: yesstr and nostr are missing for zh_HK
558 [21734] localedata: Missing yesstr and nostr are for kw_GB
559 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
560 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
561 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
562 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
564 [21744] libc: Tests failing on --enable-tunables --enable-stack-
566 [21749] localedata: Wrong abbreviated day name (“abday”) for
568 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
569 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
570 [21759] localedata: missing yesstr and nostr for Tigrinya
571 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
572 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
573 [21767] localedata: Missing Bislama locales
574 [21768] localedata: Missing yesstr and nostr for aa_ET
575 [21770] localedata: Missing Field in li_NL
576 [21778] nptl: Robust mutex may deadlock
577 [21779] libc: MicroBlaze segfaults when loading libpthread
578 [21783] localedata: Fix int_select international_call_prefixes
579 [21784] localedata: Inconsistency in country_isbn
580 [21788] localedata: Missing Country Postal Abbreviations
581 [21794] localedata: Added-country_isbn-for-Italy
582 [21795] localedata: Add/Fix country_isbn for France
583 [21796] localedata: Added country_isbn for Republic of Korea
584 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
585 [21799] localedata: Added int_select international_call_prefixes
586 [21801] localedata: Added int_select international_call_prefixes
587 [21804] nptl: Double semicolon in thread-shared-types.h
588 [21807] localedata: LC_ADDRESS fix for pap_CW
589 [21808] localedata: Fix LC_ADDRESS for pap_AW
590 [21821] localedata: Added country_name in mai_IN
591 [21822] localedata: Fix LC_TIME for mai_IN
592 [21823] localedata: missing yesstr, nostr for sa_IN
593 [21825] localedata: Fix name_mrs for mag_IN
594 [21828] localedata: 2.26 changelog should mention user visible changes
596 [21835] localedata: Added Maithili language locale for Nepal
597 [21838] localedata: Removed redundant data for the_NP
598 [21839] localedata: Fix LC_MONETARY for ta_LK
599 [21844] localedata: Fix Latin characters and Months Sequence.
600 [21848] localedata: Fix mai_NP Title Name
605 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
606 24731-2:2010, is supported to enable declarations of functions from that
607 TR. Note that not all functions from that TR are supported by the GNU C
610 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
611 18661-1:2014, is supported to enable declarations of functions and macros
612 from that TS. Note that not all features from that TS are supported by
615 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
616 18661-4:2015, is supported to enable declarations of functions and macros
617 from that TS. Note that most features from that TS are not supported by
620 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
621 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
622 Since the GNU C Library defaults to a much newer revision of POSIX, this
623 will only affect programs that specifically request an old conformance
624 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
625 see a change in the visible declarations, but a program compiled with
626 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
629 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
630 defined by all multithreaded code, but glibc has not required this for
633 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
634 means that in a future release, the macros “major”, “minor”, and “makedev”
635 will only be available from <sys/sysmacros.h>.
637 These macros are not part of POSIX nor XSI, and their names frequently
638 collide with user code; see for instance glibc bug 19239 and Red Hat bug
639 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
640 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
643 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
644 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
645 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
647 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
648 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
649 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
650 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
651 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
652 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
653 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
654 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
655 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
656 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
657 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
660 * New <math.h> features are added from TS 18661-1:2014:
662 - Signaling NaN macros: SNANF, SNAN, SNANL.
664 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
665 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
666 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
668 - llogb functions: the llogb, llogbf and llogbl functions, and the
669 FP_LLOGB0 and FP_LLOGBNAN macros.
671 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
674 - Comparison macros: iseqsig.
676 - Classification macros: iscanonical, issubnormal, iszero.
678 - Total order functions: totalorder, totalorderf, totalorderl,
679 totalordermag, totalordermagf, totalordermagl.
681 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
683 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
684 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
686 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
687 are added to libc. They convert a floating-point number into string.
689 * Most of glibc can now be built with the stack smashing protector enabled.
690 It is recommended to build glibc with --enable-stack-protector=strong.
691 Implemented by Nick Alcock (Oracle).
693 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
694 intended to be used instead of memset() to erase sensitive data after use;
695 the compiler will not optimize out calls to explicit_bzero even if they
696 are "unnecessary" (in the sense that no _correct_ program can observe the
697 effects of the memory clear).
699 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
700 to float instead of double. This does not affect the ABI of any libraries
701 that are part of the GNU C Library, but may affect the ABI of other
702 libraries that use this type in their interfaces.
704 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
705 float_t and double_t types are now defined to long double instead of float
706 and double. These options are not the default, and this does not affect
707 the ABI of any libraries that are part of the GNU C Library, but it may
708 affect the ABI of other libraries that use this type in their interfaces,
709 if they are compiled or used with those options.
711 * The getentropy and getrandom functions, and the <sys/random.h> header file
714 * The buffer size for byte-oriented stdio streams is now limited to 8192
715 bytes by default. Previously, on Linux, the default buffer size on most
716 file systems was 4096 bytes (and thus remains unchanged), except on
717 network file systems, where the buffer size was unpredictable and could be
718 as large as several megabytes.
720 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
721 for the Linux quota interface which predates kernel version 2.4.22 has
724 * The malloc_get_state and malloc_set_state functions have been removed.
725 Already-existing binaries that dynamically link to these functions will
726 get a hidden implementation in which malloc_get_state is a stub. As far
727 as we know, these functions are used only by GNU Emacs and this change
728 will not adversely affect already-built Emacs executables. Any undumped
729 Emacs executables, which normally exist only during an Emacs build, should
730 be rebuilt by re-running “./configure; make” in the Emacs build tree.
732 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
733 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
734 “no-ip6-dotint” had already been the default, and support for the
735 “ip6-dotint” option was removed from the Internet in 2006.
737 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
738 flag from <resolv.h> have been removed. The option relied on a
739 backwards-incompatible DNS extension which was never deployed on the
742 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
743 RES_BLAST defined in the <resolv.h> header file have been deprecated.
744 They were already unimplemented.
746 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
747 _res.flags are deprecated. The flag was standardized in RFC 2133, but
748 removed again from the IETF name lookup interface specification in RFC
749 2553. Applications should use getaddrinfo instead.
751 * DNSSEC-related declarations and definitions have been removed from the
752 <arpa/nameser.h> header file, and libresolv will no longer attempt to
753 decode the data part of DNSSEC record types. Previous versions of glibc
754 only implemented minimal support for the previous version of DNSSEC, which
755 is incompatible with the currently deployed version.
757 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
758 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
759 <arpa/nameser.h> header file because the distinction between RR types and
760 meta-RR types is not officially standardized, subject to revision, and
761 thus not suitable for encoding in a macro.
763 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
764 and rhook members of the res_state type in <resolv.h> have been removed.
765 The glibc stub resolver did not support these hooks, but the header file
766 did not reflect that.
768 * For multi-arch support it is recommended to use a GCC which has
769 been built with support for GNU indirect functions. This ensures
770 that correct debugging information is generated for functions
771 selected by IFUNC resolvers. This support can either be enabled by
772 configuring GCC with '--enable-gnu-indirect-function', or by
773 enabling it by default by setting 'default_gnu_indirect_function'
774 variable for a particular architecture in the GCC source file
777 * GDB pretty printers have been added for mutex and condition variable
778 structures in POSIX Threads. When installed and loaded in gdb these pretty
779 printers show various pthread variables in human-readable form when read
780 using the 'print' or 'display' commands in gdb.
782 * Tunables feature added to allow tweaking of the runtime for an application
783 program. This feature can be enabled with the '--enable-tunables' configure
784 flag. The GNU C Library manual has details on usage and README.tunables has
785 instructions on adding new tunables to the library.
787 * A new version of condition variables functions have been implemented in
788 the NPTL implementation of POSIX Threads to provide stronger ordering
791 * A new version of pthread_rwlock functions have been implemented to use a more
792 scalable algorithm primarily through not using a critical section anymore to
795 Security related changes:
797 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
798 have been created with makecontext could fail to terminate due to a
799 missing .cantunwind annotation. This has been observed to lead to a hang
800 (denial of service) in some Go applications compiled with gccgo. Reported
801 by Andreas Schwab. (CVE-2016-6323)
803 * The DNS stub resolver functions would crash due to a NULL pointer
804 dereference when processing a query with a valid DNS question type which
805 was used internally in the implementation. The stub resolver now uses a
806 question type which is outside the range of valid question type values.
809 The following bugs are resolved with this release:
811 [4099] stdio: Overly agressive caching by stream i/o functions.
812 [7065] build: Support building glibc with -fstack-protector or -fstack-
814 [9842] localedata: en_CA: incorrect date format
815 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
816 before it started waiting
817 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
818 [15765] nptl: sem_open is wrongly a cancellation point
819 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
820 [16458] libc: endian.h and netinet/in.h byte order macros return values of
822 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
823 library linked with pthread
824 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
826 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
827 [17252] libc: getrandom and getentropy syscall
828 [17730] malloc: thread-local storage is sometimes improperly free()'d
829 after being __libc_memalign()'d
830 [18241] stdio: failed fseek on memstream does not set errno and can fail
832 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
834 [18463] nptl: pthread_cond_broadcast issue when surrounded by
835 PTHREAD_PRIO_INHERIT mutex on ARM
836 [18784] network: res_query and related function crash for special record
837 type queries (CVE-2015-5180)
838 [19380] math: strtod does not raise "inexact"
839 [19387] string: Integer overflow in memchr
840 [19390] string: Integer overflow in strncat
841 [19398] build: linknamespace tests fail with massively parallel build
842 [19402] nptl: Deadlock with robust shared mutex and asynchronous
844 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
845 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
846 [19514] libc: [PATCH] Fix spelling errors in spelling
847 "implement"/"implementation" in several places
848 [19582] network: Deprecate RES_USE_INET6
849 [19673] manual: clog10 docs appear to be erroneous
850 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
851 causes a segmentation fault
852 [19826] libc: invalid pointer returned from __tls_get_addr with static
854 [20016] network: resolv: Remove hooks support from the API
855 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
857 [20033] math: [x86_64] vectorized math function don't call the __finite
859 [20116] nptl: use after free in pthread_create
860 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
861 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
863 [20311] nptl: please install proc_service.h
864 [20366] libc: Compilation errors in installed headers in strict-compliance
866 [20370] malloc: malloc: Arena free list management is still racy
867 (incorrect fix in bug 19243)
868 [20386] libc: assert (X = 0) does not result in GCC warning
869 [20432] malloc: malloc: Minimize interface required for interposition
870 [20435] libc: Missing unwind info in __startcontext causes infinite loop
871 in _Unwind_Backtrace (CVE-2016-6323)
872 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
873 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
875 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
876 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
877 [20477] network: resolv: incorrect double-checked locking related to
879 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
881 [20495] math: x86_64 performance degradation due to AVX/SSE transition
883 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
884 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
885 cause transition penalty
886 [20517] math: sparcv9 missing fdiml compat symbol
887 [20524] manual: strverscmp is inconsistent
888 [20525] libc: <sys/quota.h> should be based on kernel headers
889 [20539] math: GCC 7 -static -lm fails to link at -O3
890 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
892 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
894 [20591] network: Remove obsolete DNSSEC support
895 [20592] network: DNS resource record type classification macros in
896 <arpa/nameser.h> are incorrect
897 [20593] network: Update DNS RR type definitions
898 [20611] network: getaddrinfo accepts invalid numeric scope IDs
899 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
901 [20629] network: libresolv: Remove support for bitstring labels
903 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
904 assertion: "maxidx >= 2"
905 [20660] math: [arm] Use VSQRT
906 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
907 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
908 [20689] libc: Test for FMA should also check for AVX.
909 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
910 [20715] math: iszero macro breaks existing code
911 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
912 [20728] libc: powerpc: Missing TOC stub in clone
913 [20729] build: glibc-2.24 fails to build for i486 with -Os
914 [20750] build: Build fails with default PIE enabled gcc-6.x
915 [20768] math: [alpha] sqrt fegetenv namespace
916 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
918 [20787] math: float_t is defined as float incorrectly on x86_64 even if
919 __FLT_EVAL_METHOD__ is 2
920 [20790] network: rpcgen buffer overrun in get_prog_declaration
921 [20822] nptl: race condition in __lll_unlock_elision on powerpc
922 [20829] libc: crypt snprintf namespace
923 [20847] libc: tst-vfork3 failure
924 [20855] math: Default bits/mathdef.h has inappropriate float_t
925 [20859] math: [sh4] FP_ILOGB0 invalid
926 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
927 U+20AC), not same as GBK
928 [20915] dynamic-link: global-dynamic TLS broken on aarch64
929 [20916] math: pow handling of sNaN arguments
930 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
931 [20919] math: Bad pow (qNaN, 0) result with -lieee
932 [20940] math: hypot sNaN handling
933 [20947] math: fmax, fmin sNaN handling
934 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
935 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
937 [20971] string: powerpc64/power7 memchr overflows internal pointer check
938 [20973] nptl: robust mutexes: Lost wake-ups
939 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
940 [20978] nis: strlen on null pointer in nss_nisplus
941 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
943 [21014] string: i686 memchr overflows internal pointer check
944 [21019] libc: [mips] n32 lseek incorrect on overflow
945 [21022] libc: [microblaze] __backtrace get_frame_size namespace
946 [21026] libc: [MIPS] readahead syscall is broken on n64
947 [21028] math: Fallback fesetexceptflag should always succeed
948 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
949 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
951 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
952 [21061] librt: [microblaze] librt lost clock_* exports
953 [21073] libc: tunables: insecure environment variables passed to
954 subprocesses with AT_SECURE
955 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
959 * The minimum Linux kernel version that this version of the GNU C Library
960 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
961 kernel version 2.6.32 or later suffices (on architectures that already
962 required kernel versions more recent than 3.2, those requirements remain
963 unchanged). Linux 3.2 or later kernel headers are required on all
966 * The pap_AN locale has been deleted. This has been deprecated for a long
967 time. It has been replaced by pap_AW & pap_CW, both of which have long
968 been included in previous releases.
970 * The readdir_r and readdir64_r functions have been deprecated. It is
971 recommended to use readdir and readdir64 instead.
973 * The type “union wait” has been removed. It was deprecated in the early
974 1990s and never part of POSIX. Application code should use the int type
975 instead of “union wait”.
977 * A new NSS action is added to facilitate large distributed system
978 administration. The action, MERGE, allows remote user stores like LDAP
979 to be merged into local user stores like /etc/groups in order to provide
980 easy to use, updated, and managed sets of merged credentials. The new
981 action can be used by configuring it in /etc/nsswitch.conf:
982 group: files [SUCCESS=merge] nis
983 Implemented by Stephen Gallagher (Red Hat).
985 * The deprecated __malloc_initialize_hook variable has been removed from the
988 * The long unused localedef --old-style option has been removed. It hasn't
989 done anything in over 16 years. Scripts using this option can safely
992 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
993 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
994 return the next representable value in the direction of positive infinity
995 and the nextdown functions return the next representable value in the
996 direction of negative infinity. These are currently enabled as GNU
999 Security related changes:
1001 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1002 could result in a stack overflow when getnetbyname was called with an
1003 overly long name. (CVE-2016-3075)
1005 * Previously, getaddrinfo copied large amounts of address data to the stack,
1006 even after the fix for CVE-2013-4458 has been applied, potentially
1007 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1008 instead. Reported by Michael Petlan. (CVE-2016-3706)
1010 * The glob function suffered from a stack-based buffer overflow when it was
1011 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1012 Reported by Alexander Cherepanov. (CVE-2016-1234)
1014 * The Sun RPC UDP client could exhaust all available stack space when
1015 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1016 alloca plugin for GCC. (CVE-2016-4429)
1018 * The IPv6 name server management code in libresolv could result in a memory
1019 leak for each thread which is created, performs a failing naming lookup,
1020 and exits. Over time, this could result in a denial of service due to
1021 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1023 The following bugs are resolved with this release:
1025 [1170] localedata: ne_NP: update Nepali locale definition file
1026 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1028 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1029 [6796] math: fdim() does not set errno on overflow
1030 [10354] libc: posix_spawn should use vfork() in more cases than presently
1031 [11213] localedata: localedata: add copyright disclaimer to locale files
1032 [12143] localedata: chr_US: new Cherokee locale
1033 [12450] localedata: sgs_LT: new locale
1034 [12676] localedata: ln_CD: new locale
1035 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1037 [13304] math: fma, fmaf, fmal produce wrong results
1038 [14259] build: --localedir arg to configure is ignored
1039 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1040 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1041 [14934] localedata: es_CL: wrong first weekday chilean locale
1042 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1044 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1046 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1047 [15368] nptl: raise() is not async-signal-safe
1048 [15479] math: ceil, floor, round and trunc raise inexact exception
1049 [15578] localedata: kk_KZ: various updates
1050 [16003] localedata: pap_AN: punt old locale
1051 [16137] localedata: iw_IL: punt old locale
1052 [16190] localedata: eo: new esperanto locale
1053 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1054 [16742] malloc: race condition: pthread_atfork() called before first
1055 malloc() results in unexpected locking behaviour/deadlocks
1056 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1058 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1059 [17565] localedata: pt_PT: wrong (work-)week start
1060 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1061 [17950] build: Build fails with -msse
1062 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1063 [18433] libc: posix_spawn does not return correctly upon failure to
1065 [18453] localedata: charmaps/IBM875: incorrect codes
1066 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1068 [18896] localedata: he_IL: improvements for currency
1069 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1070 Romanian locale data
1071 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1073 [19056] libc: Deprecate readdir_r
1074 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1076 [19198] localedata: nl_NL: small improvements for Dutch locales
1077 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1078 nameservers (CVE-2016-5417)
1079 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1081 [19400] locale: Language missing in "iso-639.def", trivial fix in
1083 [19431] malloc: Deadlock between fflush, getdelim, and fork
1084 [19505] libc: Incorrect file descriptor validity checks in
1085 posix_spawn_file_actions_add{open,close,dup2}
1086 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1087 when using RTLD_NEXT
1088 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1089 `intl/{gettextP,loadinfo}.h'
1090 [19534] libc: execle, execlp may use malloc
1091 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1092 [19573] network: res_nclose and __res_maybe_init disagree about name
1093 server initialization, breaking Hesiod
1094 [19575] localedata: Status of GB18030 tables
1095 [19581] localedata: sr_* date_fmt string contains additional newline
1096 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1098 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1099 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1100 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1101 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1103 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1105 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1107 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1108 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1109 [19633] locale: strfmon_l applies global locale to number formatting
1110 [19642] network: Memory leak in getnameinfo
1111 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1112 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1114 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1115 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1117 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1118 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1120 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1122 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1124 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1125 pointers and lengths in error-case.
1126 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1127 report errors on UTF-16 surrogates values.
1128 [19755] nscd: nscd assertion failure in gc
1129 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1130 [19759] libc: mempcpy shouldn't be inlined
1131 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1132 [19765] libc: s390 needs an optimized mempcpy
1133 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1134 NAME_MAX limit assumption (CVE-2016-1234)
1135 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1136 [19787] network: Missing and incorrect truncation checks in getnameinfo
1137 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1139 [19791] network: Assertion failure in res_query.c with un-connectable name
1141 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1142 [19822] math: libm.so install clobbers old version
1143 [19825] network: resolv: send_vc can return uninitialized data in second
1144 response to getaddrinfo
1145 [19830] network: nss_dns: should check RDATA length against buffer length
1146 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1147 confronted with A/AAAA records of invalid size
1148 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1149 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1150 [19853] stdio: Printing IBM long double in decimal with high precision is
1152 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1153 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1154 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1156 [19865] network: Assertion failure or memory leak in
1157 _nss_dns_getcanonname_r
1158 [19868] network: nss_dns: netent code does not skip over non-PTR records
1159 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1161 [19881] string: Improve x86-64 memset
1162 [19907] string: Incorrect memcpy tests
1163 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1165 [19925] libc: termios.h XCASE namespace
1166 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1167 [19929] libc: limits.h NL_NMAX namespace
1168 [19931] stdio: Memory leak in vfprintf
1169 [19957] libc: clone(CLONE_VM) access invalid parent memory
1170 [19963] localedata: en_IL: New locale
1171 [19989] stdio: stdio.h cuserid namespace
1172 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1174 [19996] locale: langinfo.h nl_langinfo_l namespace
1175 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1176 [20010] network: getaddrinfo: Stack overflow in hostent translation
1178 [20012] stdio: libio: fmemopen append mode failure
1179 [20014] stdio: stdio.h namespace for pre-threads POSIX
1180 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1181 [20023] libc: fcntl.h timespec namespace
1182 [20024] math: [x86_64] vectorized sincos trashes the stack
1183 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1184 [20041] time: sys/time.h timespec namespace
1185 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1186 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1187 [20051] libc: ttyslot in wrong header under wrong conditions
1188 [20054] libc: gethostname not declared for XPG4
1189 [20055] libc: termios.h missing tcgetsid for XPG4
1190 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1192 [20073] libc: sys/stat.h fchmod namespace
1193 [20074] libc: stdlib.h rand_r namespace
1194 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1195 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1197 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1198 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1200 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1201 [20119] libc: Wrong mask for processors level type from CPUID
1202 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1203 AS not supporting AVX512
1204 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1206 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1207 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1209 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1210 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1211 [20178] libc: posix_spawn{p} should not call exit
1212 [20191] stdio: libio: vtables hardening
1213 [20195] string: FMA4 detection requires CPUID execution with register
1215 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1216 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1218 [20212] math: acos (sNaN) returns sNaN
1219 [20213] math: asin (sNaN) returns sNaN
1220 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1221 [20218] math: [i386] asinhl (sNaN) returns sNaN
1222 [20219] math: [i386] atanhl (sNaN) returns sNaN
1223 [20222] stdio: fopencookie: Mangle function pointers
1224 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1225 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1226 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1228 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1229 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1230 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1231 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1232 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1233 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1234 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1235 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1236 [20240] math: modf (sNaN) returns sNaN
1237 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1238 [20250] math: frexp (sNaN) returns sNaN
1239 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1240 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1242 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1243 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1244 subscript is below array bounds [-Werror=array-bounds]
1245 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1246 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1248 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1249 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1251 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1252 "invalid" exceptions
1253 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1254 [20316] localedata: id_ID: Februari instead of Pebruari
1255 [20327] string: POWER8 strcasecmp returns incorrect result
1256 [20347] math: Failure: Test: j0_downward (0xap+0)
1257 [20348] libc: FAIL: misc/tst-preadvwritev64
1258 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1259 [20350] libc: There is no test for p{read,write}64
1260 [20357] math: Incorrect cos result for 1.5174239687223976
1261 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1265 * Unicode 8.0.0 Support: Character encoding, character type info, and
1266 transliteration tables are all updated to Unicode 8.0.0, using new
1267 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1268 These updates cause user visible changes, such as the fixes for bugs
1269 89, 16061, and 18568.
1271 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1272 kernel-internal CPU set size. This means that requests that change the
1273 CPU affinity which failed before (for example, an all-ones CPU mask) will
1274 now succeed. Applications that need to determine the effective CPU
1275 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1276 setting it because the kernel can adjust it (and the previous size check
1277 would not detect this in the majority of cases).
1279 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1280 the following new symbols are used: fts64_children, fts64_close,
1281 fts64_open, fts64_read and fts64_set.
1283 * getaddrinfo now detects certain invalid responses on an internal netlink
1284 socket. If such responses are received, an affected process will
1285 terminate with an error message of "Unexpected error <number> on netlink
1286 descriptor <number>" or "Unexpected netlink response of size <number> on
1287 descriptor <number>". The most likely cause for these errors is a
1288 multi-threaded application which erroneously closes and reuses the netlink
1289 file descriptor while it is used by getaddrinfo.
1291 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1292 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1293 unnecessary serialization of memory allocation requests across threads.
1294 The defect is now corrected. Users should see a substantial increase in
1295 the concurent throughput of allocation requests for applications which
1296 trigger this bug. Affected applications typically create and
1297 destroy threads frequently. (Bug 19048 was reported and analyzed by
1300 * There is now a --disable-timezone-tools configure option for disabling the
1301 building and installing of the timezone related utilities (zic, zdump, and
1302 tzselect). This is useful for people who build the timezone data and code
1303 independent of the GNU C Library.
1305 * The obsolete header <regexp.h> has been removed. Programs that require
1306 this header must be updated to use <regex.h> instead.
1308 * The obsolete functions bdflush, create_module, get_kernel_syms,
1309 query_module and uselib are no longer available to newly linked binaries;
1310 the header <sys/kdaemon.h> has been removed. These functions and header
1311 were specific to systems using the Linux kernel and could not usefully be
1312 used with the GNU C Library on systems with version 2.6 or later of the
1315 * Optimized string, wcsmbs and memory functions for IBM z13.
1316 Implemented by Stefan Liebler.
1318 * Newly linked programs that define a variable called signgam will no longer
1319 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1320 require signgam to be set by those functions must ensure that they use the
1321 variable provided by the GNU C Library and declared in <math.h>, without
1322 defining their own copy.
1324 * The minimum GCC version that can be used to build this version of the GNU
1325 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1326 still be used to compile programs using the GNU C Library.
1328 Security related changes:
1330 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1331 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1333 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1334 the pointer guard feature. It is always enabled. Previously,
1335 LD_POINTER_GUARD could be used to disable security hardening in binaries
1336 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1339 * An integer overflow in hcreate and hcreate_r could lead to an
1340 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1342 * The catopen function no longer has unbounded stack usage. Reported by
1343 Max. (CVE-2015-8779)
1345 * The nan, nanf and nanl functions no longer have unbounded stack usage
1346 depending on the length of the string passed as an argument to the
1347 functions. Reported by Joseph Myers. (CVE-2014-9761)
1349 * A stack-based buffer overflow was found in libresolv when invoked from
1350 libnss_dns, allowing specially crafted DNS responses to seize control
1351 of execution flow in the DNS client. The buffer overflow occurs in
1352 the functions send_dg (send datagram) and send_vc (send TCP) for the
1353 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1354 family. The use of AF_UNSPEC triggers the low-level resolver code to
1355 send out two parallel queries for A and AAAA. A mismanagement of the
1356 buffers used for those queries could result in the response of a query
1357 writing beyond the alloca allocated buffer created by
1358 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1359 the overflow. Thanks to the Google Security Team and Red Hat for
1360 reporting the security impact of this issue, and Robert Holiday of
1361 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1363 The following bugs are resolved with this release:
1365 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1366 [887] math: Math library function "logb" and "nextafter" inconsistent
1367 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1368 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1369 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1370 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1372 [4404] localedata: German translation of "Alarm clock" is misleading
1373 [6799] math: nextafter() and nexttoward() doen't set errno on
1374 overflow/underflow errors
1375 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1377 [10432] nis: _nss_nis_setnetgrent assertion failure
1378 [11460] libc: fts has no LFS support
1379 [12926] network: getaddrinfo()/make_request() may spin forever
1380 [13065] nptl: Race condition in pthread barriers
1381 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1382 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1384 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1385 [14912] libc: Rename non-installed bits/*.h headers
1386 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1387 [15367] math: Let gcc use __builtin_isinf
1388 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1389 [15421] math: lgamma wrongly sets signgam for ISO C
1390 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1391 FE_INVALID with argument out of range
1392 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1394 [15786] dynamic-link: ifunc resolver functions can smash function
1396 [15918] math: Unnecessary check for equality in hypotf()
1397 [16061] localedata: Review / update transliteration data
1398 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1400 [16141] time: strptime %z offset restriction
1401 [16171] math: drem should be alias of remainder
1402 [16296] math: fegetround is pure?
1403 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1404 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1406 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1407 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1409 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1410 "Exception "Inexact" set" on ppc32
1411 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1412 [16517] math: Missing underflow exception from tanf/tan/tanl
1413 [16519] math: Missing underflow exception from sinhf
1414 [16520] math: Missing underflow exception from tanhf
1415 [16521] math: Missing underflow exception from exp2
1416 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1418 [16734] stdio: fopen calls mmap to allocate its buffer
1419 [16961] math: nan function incorrect handling of bad sequences
1420 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1421 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1422 [16985] locale: localedef: confusing error message when opening output
1424 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1425 [17197] locale: Redundant shift character in iconv conversion output at
1427 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1428 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1429 [17250] dynamic-link: static linking breaks nss loading
1430 (getaddrinfo/getpwnam/etc...)
1431 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1433 [17441] math: isnan() should use __builtin_isnan() in GCC
1434 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1435 (related to lock elision)
1436 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1437 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1438 [17887] time: strptime should be able to parse "+01:00" style timezones
1439 [17905] libc: catopen() Multiple unbounded stack allocations
1441 [18084] libc: backtrace (..., 0) dumps core on x86
1442 [18086] libc: nice() sets errno to 0 on success
1443 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1444 is too large (CVE-2015-8778)
1445 [18251] dynamic-link: SONAME missing when audit modules provides path
1446 [18265] libc: add attributes for wchar string and memory functions
1447 [18370] math: csqrt missing underflows
1448 [18421] libc: [hppa] read-only segment has dynamic relocations
1449 [18472] libc: Obsolete syscall wrappers should be compat symbols
1450 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1451 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1452 [18525] localedata: Remove locale timezone information
1453 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1454 [18568] localedata: Update locale data to Unicode 8.0
1455 [18589] locale: sort-test.sh fails at random
1456 [18595] math: ctan, ctanh missing underflows
1457 [18604] libc: assert macro-expands its argument
1458 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1459 contains a vector instruction exception.
1460 [18611] math: j1, jn missing errno setting on underflow
1461 [18618] localedata: sync Chechen locale definitions with other *_RU
1463 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1464 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1465 [18665] network: In send_dg, the recvfrom function is NOT always using the
1466 buffer size of a newly created buffer (CVE-2015-7547)
1467 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1469 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1471 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1472 [18699] math: tilegx cproj() for various complex infinities does not yield
1474 [18724] libc: Harden put*ent functions against data injection
1475 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1476 [18755] build: build errors with -DNDEBUG
1477 [18757] stdio: fmemopen fails to set errno on failure
1478 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1480 [18781] libc: openat64 lacks O_LARGEFILE
1481 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1482 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1483 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1484 [18790] math: [ldbl-128ibm] tanhl inaccurate
1485 [18795] libc: stpncpy fortification misses buffer lengths that are
1486 statically too large
1487 [18796] build: build fails for --disable-mathvec
1488 [18803] math: hypot missing underflows
1489 [18820] stdio: fmemopen may leak memory on failure
1490 [18823] math: csqrt spurious underflows
1491 [18824] math: fma spurious underflows
1492 [18825] math: pow missing underflows
1493 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1494 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1495 [18870] build: sem_open.c fails to compile with missing symbol
1497 [18872] stdio: Fix memory leak in printf_positional
1498 [18873] libc: posix_fallocate overflow check ineffective
1499 [18875] math: Excess precision leads incorrect libm
1500 [18877] libc: arm: mmap offset regression
1501 [18887] libc: memory corruption when using getmntent on blank lines
1502 [18918] localedata: hu_HU: change time to HH:MM:SS format
1503 [18921] libc: Regression: extraneous stat() and fstat() performed by
1505 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1506 binaries (CVE-2015-8777)
1507 [18951] math: tgamma missing underflows
1508 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1510 [18953] localedata: lt_LT: change currency symbol to the euro
1511 [18956] math: powf inaccuracy
1512 [18961] math: [i386] exp missing underflows
1513 [18966] math: [i386] exp10 missing underflows
1514 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1515 [18969] build: multiple string test failures due to missing locale
1517 [18970] libc: Reference of pthread_setcancelstate in libc.a
1518 [18977] math: float / long double Bessel functions not in XSI POSIX
1519 [18980] math: i386 libm functions return with excess range and precision
1520 [18981] math: i386 scalb*, ldexp return with excess range and precision
1521 [18982] stdio: va_list and vprintf
1522 [18985] time: Passing out of range data to strftime() causes a segfault
1524 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1525 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1526 [19012] locale: iconv_open leaks memory on error path
1527 [19016] math: clog, clog10 inaccuracy
1528 [19018] nptl: Mangle function pointers in tls_dtor_list
1529 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1530 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1531 [19048] malloc: malloc: arena free list can become cyclic, increasing
1533 [19049] math: [powerpc] erfc incorrect zero sign
1534 [19050] math: [powerpc] log* incorrect zero sign
1535 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1536 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1537 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1538 [19074] network: Data race in _res_hconf_reorder_addrs
1539 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1540 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1541 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1543 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1545 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1546 [19086] manual: posix_fallocate64 documented argument order is wrong.
1547 [19088] math: lround, llround missing exceptions close to overflow
1549 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1550 [19095] math: dbl-64 lrint incorrect for 64-bit long
1551 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1552 [19124] dynamic-link: ld.so failed to build with older assmebler
1553 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1554 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1555 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1556 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1557 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1558 pthread_setaffinity_np
1559 [19156] math: [ldbl-128] j0l spurious underflows
1560 [19164] nptl: tst-getcpu fails with many possible CPUs
1561 [19168] math: math/test-ildoubl and math/test-ldouble failure
1562 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1563 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1565 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1566 (FE_NOMASK_ENV) do not clear SSE exceptions
1567 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1568 _int_new_arena/reused_arena
1569 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1570 [19201] math: dbl-64 remainder incorrect sign of zero result
1571 [19205] math: bits/math-finite.h conditions do not match math.h and
1573 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1574 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1575 for C99-based standards
1576 [19212] libc: features.h not -Wundef clean
1577 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1579 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1580 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1581 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1583 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1584 "inexact" exceptions
1585 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1587 [19242] libc: strtol incorrect in Turkish locales
1588 [19243] malloc: reused_arena can pick an arena on the free list, leading
1589 to an assertion failure and reference count corruption
1590 [19253] time: tzset() ineffective when temporary TZ did not include DST
1592 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1593 [19270] math: [hppa] Shared libm missing __isnanl
1594 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1595 MAP_HUGETLB and MAP_STACK defines
1596 [19313] nptl: Wrong __cpu_mask for x32
1597 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1598 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1599 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1600 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1601 [19363] time: x32: times() return value wrongly truncates/sign extends
1603 [19367] dynamic-link: Improve branch prediction on Silvermont
1604 [19369] network: Default domain name not reset by res_ninit when "search"
1605 / "domain" entry is removed from resolv.conf
1606 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1608 [19385] localedata: bg_BG: time separator should be colon, not comma
1609 [19408] libc: linux personality syscall wrapper may erroneously return an
1610 error on 32-bit architectures
1611 [19415] libc: dladdr returns wrong names on hppa
1612 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1613 IBM905, IBM907, and IBM909
1614 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1615 [19443] build: build failures with -DDEBUG
1616 [19451] build: Make check fails on test-double-vlen2
1617 [19462] libc: Glibc failed to build with -Os
1618 [19465] math: Wrong code with -Os
1619 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1621 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1623 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1624 [19486] math: S390: Math tests fail with "Exception Inexact set".
1625 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1626 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1628 [19590] math: Fail to build shared objects that use libmvec.so functions.
1632 * The following bugs are resolved with this release:
1634 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1635 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1636 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1637 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1638 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1639 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1640 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1641 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1642 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1643 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1644 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1645 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1646 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1647 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1648 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1649 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1650 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1651 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1652 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1653 18657, 18676, 18694, 18696, 18887.
1655 * Cache information can be queried via sysconf() function on s390 e.g. with
1656 _SC_LEVEL1_ICACHE_SIZE as argument.
1658 * A buffer overflow in gethostbyname_r and related functions performing DNS
1659 requests has been fixed. If the NSS functions were called with a
1660 misaligned buffer, the buffer length change due to pointer alignment was
1661 not taken into account. This could result in application crashes or,
1662 potentially arbitrary code execution, using crafted, but syntactically
1663 valid DNS responses. (CVE-2015-1781)
1665 * The time zone file parser has been made more robust against crafted time
1666 zone files, avoiding heap buffer overflows related to the processing of
1667 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1668 large time zone data files. Overly long time zone specifiers in the TZ
1669 variable no longer result in stack overflows and crashes.
1671 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1672 for LD and GD on x86 and x86-64, has been implemented. You will need
1673 binutils-2.24 or later to enable this optimization.
1675 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1676 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1677 Hat). These updates cause user visible changes, such as the fix for bug
1680 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1681 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1682 condition in some applications.
1684 * Added vector math library named libmvec with the following vectorized x86_64
1685 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1687 The library can be disabled with --disable-mathvec. Use of the functions is
1688 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1689 Shared library libmvec.so is linked in as needed when using -lm (no need to
1690 specify -lmvec explicitly for not static builds).
1691 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1693 * A new fmemopen implementation has been added with the goal of POSIX
1694 compliance. The new implementation fixes the following long-standing
1695 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1696 old implementation is still present for use be by existing binaries.
1698 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1699 releases. It has been fixed to match 2.19 and older, but binaries built
1700 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1702 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1703 Contributed by Roland McGrath (Google).
1705 * The header <regexp.h> is deprecated, and will be removed in a future
1706 release. Use of this header will trigger a deprecation warning.
1707 Application developers should update their code to use <regex.h> instead.
1709 This header was formerly part of SUS, but was deprecated in 1994 and
1710 removed from the standard in 2001. Also, the glibc implementation
1711 leaks memory. See BZ#18681 for more details.
1715 * The following bugs are resolved with this release:
1717 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1718 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1719 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1720 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1721 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1722 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1723 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1724 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1725 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1726 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1729 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1730 too little memory for the to-be-scanned arguments and overflow the
1731 allocated buffer. The implementation now correctly computes the required
1732 buffer size when using malloc, and switches to malloc from alloca as
1735 * A new semaphore algorithm has been implemented in generic C code for all
1736 machines. Previous custom assembly implementations of semaphore were
1737 difficult to reason about or ensure that they were safe. The new version
1738 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1739 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1740 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1742 * Port to Altera Nios II has been contributed by Mentor Graphics.
1744 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1745 implementations for powerpc64/powerpc64le.
1746 Implemented by Adhemerval Zanella (IBM).
1748 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1749 and powerpc64le. This may improve lock scaling of existing programs on
1750 HTM capable systems. The lock elision code is only enabled with
1751 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1752 powerpc will issue a transaction abort on every syscall to avoid side
1753 effects being visible outside transactions.
1755 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1756 AArch64. Contributed by ARM Ltd.
1758 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1760 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1761 under certain input conditions resulting in the execution of a shell for
1762 command substitution when the application did not request it. The
1763 implementation now checks WRDE_NOCMD immediately before executing the
1764 shell and returns the error WRDE_CMDSUB as expected.
1766 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1767 processing format strings with a large number of format specifiers.
1769 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1770 infinite loop if the DNS response contained a PTR record of an unexpected
1773 * The minimum GCC version that can be used to build this version of the GNU
1774 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1775 still be used to compile programs using the GNU C Library.
1777 * The GNU C Library is now built with -Werror by default. This can be
1778 disabled by configuring with --disable-werror.
1780 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1782 * The obsolete sigvec function has been removed. This was the original
1783 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1784 programs have been using instead for about 25 years. Of course, ABI
1785 compatibility for old binaries using sigvec remains intact.
1787 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1788 with newer versions of bison.
1790 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1791 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1792 registers overlay two consecutive single-precision registers. MIPS32R2
1793 introduced a new FPU mode (FR=1) where double-precision registers extend the
1794 corresponding single-precision registers which is incompatible with the
1795 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1796 require the use of FR=1 making a transition necessary. New o32 ABI
1797 extensions enable users to migrate over time from the original o32 ABI
1798 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1799 tracks the ABI of any loaded object and verifies that new objects are
1800 compatible. Mode transitions will also be requested as required and
1801 unsupportable objects will be rejected. The ABI checks include both soft and
1802 hard float ABIs for o32, n32 and n64.
1804 GCC 5 with GNU binutils 2.25 onwards:
1805 It is strongly recommended that all o32 system libraries are built using the
1806 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1807 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1808 compiler using --with-fp-32=xx to set this by default.
1812 * The following bugs are resolved with this release:
1814 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1815 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1816 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1817 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1818 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1819 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1820 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1821 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1822 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1823 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1824 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1825 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1826 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1827 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1828 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1830 * Reverted change of ABI data structures for s390 and s390x:
1831 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1832 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1833 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1834 are preserved pointing straight to the same implementation as the old ones.
1835 Given that, new callers will simply provide a too-big buffer to these
1836 functions. Any applications/libraries out there that embed jmp_buf or
1837 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1838 against 2.19 headers will have to rebuilt again. This is necessary in any
1839 case to revert the breakage in their ABI caused by the glibc change.
1841 * Support for file description locks is added to systems running the
1842 Linux kernel. The standard file locking interfaces are extended to
1843 operate on file descriptions, not file descriptors, via the use of
1844 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1845 are associated with an open file instead of a process.
1847 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1849 * The minimum Linux kernel version that this version of the GNU C Library
1850 can be used with is 2.6.32.
1852 * Running the testsuite no longer terminates as soon as a test fails.
1853 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1854 with PASS or FAIL lines for individual tests. A summary of the results is
1855 printed, including a list of failing lists, and "make check" exits with
1856 error status if there were any unexpected failures. "make check
1857 stop-on-test-failure=y" may be used to keep the old behavior.
1859 * The am33 port, which had not worked for several years, has been removed
1862 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1863 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1864 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1865 conflicted with POSIX (support for which was removed in 2.19), the
1866 interfaces those macros enabled remain available when compiling with
1867 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1868 test macros defined.
1870 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1872 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1873 This may improve lock scaling of existing programs on TX capable systems.
1874 The lock elision code is only built with --enable-lock-elision=yes and
1875 then requires a GCC version supporting the TX builtins. With lock elision
1876 default mutexes are elided via __builtin_tbegin, if the cpu supports
1877 transactions. By default lock elision is not enabled and the elision code
1880 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1881 copy the path argument. This allowed programs to cause posix_spawn to
1882 deference a dangling pointer, or use an unexpected pathname argument if
1883 the string was modified after the posix_spawn_file_actions_addopen
1886 * All supported architectures now use the main glibc sysdeps directory
1887 instead of some being in a separate "ports" directory (which was
1888 distributed separately before glibc 2.17).
1890 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1891 On configurations that support it (all Linux configurations), it's now
1892 used regardless of the --enable-add-ons switch to configure. It is no
1893 longer possible to build such configurations without pthreads support.
1895 * Locale names, including those obtained from environment variables (LANG
1896 and the LC_* variables), are more tightly checked for proper syntax.
1897 setlocale will now fail (with EINVAL) for locale names that are overly
1898 long, contain slashes without starting with a slash, or contain ".." path
1899 components. (CVE-2014-0475) Previously, some valid locale names were
1900 silently replaced with the "C" locale when running in AT_SECURE mode
1901 (e.g., in a SUID program). This is no longer necessary because of the
1904 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1905 with application code using Intel MPX instructions. (With all previous
1906 versions, the MPX register state could be clobbered when making calls
1907 into or out of a shared library.) Note that while the new dynamic
1908 linker is compatible with all known x86 hardware whether or not it
1909 supports Intel MPX, some x86 instruction-set emulators might fail to
1910 handle the new instruction encodings. This is known to affect Valgrind
1911 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1912 release), and might affect other tools that do instruction emulation.
1914 * Support for loadable gconv transliteration modules has been removed.
1915 The support for transliteration modules has been non-functional for
1916 over a decade, and the removal is prompted by security defects. The
1917 normal gconv conversion modules are still supported. Transliteration
1918 with //TRANSLIT is still possible, and the //IGNORE specifier
1919 continues to be supported. (CVE-2014-5119)
1921 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1922 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1923 resulting a denial-of-service security vulnerability in applications which
1924 use functions related to iconv. (CVE-2014-6040)
1928 * The following bugs are resolved with this release:
1930 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1931 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1932 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1933 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1934 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1935 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1936 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1937 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1938 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1939 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1940 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1941 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1942 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1943 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1944 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1945 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1946 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1947 16431, 16453, 16474, 16506, 16510, 16529
1949 * Slovenian translations for glibc messages have been contributed by the
1950 Translation Project's Slovenian team of translators.
1952 * The public headers no longer use __unused nor __block. This change is to
1953 support compiling programs that are derived from BSD sources and use
1954 __unused internally, and to support compiling with Clang's -fblock
1955 extension which uses __block.
1957 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1958 large collation sequences to optimize multiple passes. This cache
1959 computation may overflow for large collation sequences and may cause a
1960 stack or buffer overflow. This is now fixed to use a slower algorithm
1961 which does not use a cache if there is an integer overflow.
1963 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1964 rules for large collation sequences to optimize multiple passes and falls
1965 back to alloca if malloc fails, resulting in a possible stack overflow.
1966 The implementation now falls back to an uncached collation sequence lookup
1969 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1970 initialized for static applications resulting in the security feature
1971 being disabled. The pointer guard is now correctly initialized to a
1972 random value for static applications. Existing static applications need
1973 to be recompiled to take advantage of the fix (bug 15754).
1975 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1976 to the d_name member of struct dirent, or omit the terminating NUL
1977 character. (Bugzilla #14699).
1979 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1980 aligned_alloc functions could allocate too few bytes or corrupt the
1981 heap when passed very large allocation size values (Bugzilla #15855,
1984 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1985 for AF_INET6 has been fixed (Bugzilla #16072).
1987 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1988 pap_CW, quz_PE, the_NP.
1990 * Substantially revised locales: gd_GB, ht_HT
1992 * The LC_ADDRESS field was updated to support country_car for almost all
1995 * ISO 1427 definitions were updated.
1997 * ISO 3166 definitions were updated.
1999 * The localedef utility now supports --big-endian and --little-endian
2000 command-line options to generate locales for a different system from that
2001 for which the C library was built.
2003 * Binary locale files now only depend on the endianness of the system for
2004 which they are generated and not on other properties of that system. As a
2005 consequence, binary files generated with new localedef may be incompatible
2006 with old versions of the GNU C Library, and binary files generated with
2007 old localedef may be incompatible with this version of the GNU C Library,
2008 in the following circumstances:
2010 + Locale files may be incompatible on m68k systems.
2012 + Locale archive files (but not separate files for individual locales) may
2013 be incompatible on systems where plain "char" is signed.
2015 * The configure option --disable-versioning has been removed. Builds with
2016 --disable-versioning had not worked for several years.
2018 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2019 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2021 * SystemTap probes for malloc have been introduced.
2023 * SystemTap probes for slow multiple precision fallback paths of
2024 transcendental functions have been introduced.
2026 * Support for powerpc64le has been added.
2028 * The soft-float powerpc port now supports e500 processors.
2030 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2032 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2033 set of header declarations that are enabled by default, even when other
2034 feature test macros or compiler options such as -std=c99 would otherwise
2035 disable some of those declarations.
2037 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2038 conflict with POSIX. The libbsd-compat library (which was a dummy library
2039 that did nothing) has also been removed.
2041 * Preliminary documentation about Multi-Thread, Async-Signal and
2042 Async-Cancel Safety has been added.
2044 * Change of ABI data structures for s390 and s390x:
2045 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2046 allow for future hardware extensions. All library functions that accept or
2047 return these structures were versioned in order to provide backwards
2048 compatibility with existing code. However, not all cases can be handled
2049 with interface versioning. If an application embeds either structure into
2050 another structure and that structure is passed to another compilation unit
2051 which expects the newer larger-sized object then any access to the new
2052 fields will result in undefined behaviour. Similarly any access to fields
2053 that were moved by the enlarging of the structures will cause undefined
2054 behaviour. To fix the undefined behaviour all objects that are part of the
2055 final in-memory image must agree on the size of structures, and this may
2056 require recompilation.
2060 * The following bugs are resolved with this release:
2062 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2063 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2064 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2065 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2066 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2067 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2068 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2069 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2070 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2071 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2072 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2073 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2074 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2077 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2078 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2079 Distributions can re-enable building and using pt_chown via the new configure
2080 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2081 considerable security risks and should only be used if the distribution
2082 understands and accepts the risks.
2084 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2087 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2088 fixed (Bugzilla #15330).
2090 * Add support for calling C++11 thread_local object destructors on thread
2091 and program exit. This needs compiler support for offloading C++11
2092 destructor calls to glibc.
2094 * Improved worst case performance of libm functions with double inputs and
2097 * Support for priority inherited mutexes in pthread condition variables on
2098 non-x86 architectures.
2100 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2102 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2104 * Optimized string functions for ARM. Implemented by Will Newton and
2107 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2109 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2112 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2113 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2115 * Added a benchmark framework to track performance of functions in glibc.
2117 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2118 It is based on draft TS 18661 and currently enabled as a GNU extension.
2120 * On Linux, the clock function now uses the clock_gettime system call
2121 for improved precision, rather than old times system call.
2123 * Added support for version-3 tz format files. This is needed when using
2124 the tz database release 2013e or later, and affects a few unusual cases --
2125 currently only TZ='America/Godthab' for time stamps after 2037.
2127 * Added new API functions pthread_getattr_default_np and
2128 pthread_setattr_default_np to get and set the default pthread
2129 attributes of a process.
2131 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2132 This may improve lock scaling of existing programs on TSX capable systems.
2133 When the --enable-lock-elision=yes parameter is specified at configure
2134 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2137 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2138 availability). Implemented by Ryan S. Arnold.
2140 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2144 * The following bugs are resolved with this release:
2146 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2147 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2148 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2149 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2150 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2151 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2152 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2153 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2154 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2155 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2156 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2157 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2158 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2159 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2160 14879, 14889, 14893, 14898, 14914.
2162 * Optimization of memcpy for MIPS.
2164 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2165 EMFILE has been fixed (Bugzilla #14889).
2167 * The add-on ports collection is now distributed in the "ports" subdirectory
2168 of the main GNU C Library distribution, rather than separately.
2170 * Port to ARM AArch64 contributed by Linaro.
2172 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2173 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2175 Implemented by Andreas Krebbel.
2177 * The new function secure_getenv allows secure access to the environment,
2178 returning NULL if running in a SUID/SGID process. This function replaces
2179 the internal function __secure_getenv.
2181 * SystemTap static probes have been added into the dynamic linker.
2182 Implemented by Gary Benson.
2184 * Optimizations of string functions strstr, strcasestr and memmem.
2185 Implemented by Maxim Kuvyrkov.
2187 * The minimum Linux kernel version that this version of the GNU C Library
2188 can be used with is 2.6.16.
2190 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2191 powerpc POWER7. Implemented by Will Schmidt.
2193 * New configure option --disable-nscd builds the C library such that it
2194 never attempts to contact the Name Service Caching Daemon (nscd).
2195 New configure option --disable-build-nscd avoids building nscd itself;
2196 this is the default if --disable-nscd is used.
2198 * Improved support for cross-compilation, including cross-testing and
2199 bootstrap builds without a previously built glibc.
2201 * Several testsuite tests are now able to test multiple IFUNC variants of an
2202 interface, rather than just testing the one that would be chooen by
2205 * New configure options --with-bugurl and --with-pkgversion, for
2206 distributors to use to embed their bug-reporting and package version
2207 information in --help and --version output.
2209 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2210 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2211 allows creation of chroots without the procfs mounted on /proc.
2213 * The `crypt' function now fails if passed salt bytes that violate the
2214 specification for those values. On Linux, the `crypt' function will
2215 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2216 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2217 when the mode is enabled.
2219 * The `clock_*' suite of functions (declared in <time.h>) is now available
2220 directly in the main C library. Previously it was necessary to link with
2221 -lrt to use these functions. This change has the effect that a
2222 single-threaded program that uses a function such as `clock_gettime' (and
2223 is not linked with -lrt) will no longer implicitly load the pthreads
2224 library at runtime and so will not suffer the overheads associated with
2225 multi-thread support in other code such as the C++ runtime library.
2227 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2232 * The following bugs are resolved with this release:
2234 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2235 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2236 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2237 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2238 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2239 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2240 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2241 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2242 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2243 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2244 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2245 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2246 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2247 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2248 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2249 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2250 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2251 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2252 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2253 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2254 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2255 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2258 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2259 configuring glibc with:
2260 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2261 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2262 Implemented by H.J. Lu.
2266 + define static_assert
2268 + do not declare gets
2270 + declare at_quick_exit and quick_exit also for ISO C11
2272 + aligned_alloc. NB: The code is deliberately allows the size parameter
2273 to not be a multiple of the alignment. This is a moronic requirement
2274 in the standard but it is only a requirement on the caller, not the
2277 + timespec_get added
2279 + uchar.h support added
2281 + CMPLX, CMPLXF, CMPLXL added
2283 Implemented by Ulrich Drepper.
2285 * Support for the IA-64 has been moved to ports.
2287 * Remove support for anything but ELF binary format
2289 * Checking versions of poll, ppoll added.
2290 Implemented by Ulrich Drepper.
2292 * More generic and 64-bit performance optimizations to math functions.
2293 Implemented by Ulrich Drepper.
2295 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2296 headers and functions available at compile time as they were before
2297 version 2.14. This option will be removed at some time in the future
2298 after the TI-RPC library becomes fully sufficient for the needs of
2299 existing applications.
2301 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2302 Note that glibc is not expected to work with any Linux kernel version
2305 * New header <sys/auxv.h> and function getauxval allowing easy access to
2306 the AT_* key-value pairs passed from the Linux kernel. The header also
2307 defines the HWCAP_* bits associated with the AT_HWCAP key.
2309 * New locales: mag_IN
2311 * New configure option --enable-systemtap builds SystemTap static probes
2312 into libc for setjmp and longjmp and into libpthread for various operations.
2313 So far the setjmp/longjmp probes and some of the libpthread probes are
2314 provided only for i*86 and x86_64.
2315 Implemented by Roland McGrath and Rayson Ho.
2317 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2319 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2322 * More optimized functions for SPARC. Implemented by David S. Miller.
2324 * Improved support for cross-compilation, especially bootstrap builds
2325 without a previously built glibc.
2327 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2328 Chris Metcalf from Tilera.
2330 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2331 now supported for ARM processors.
2333 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2334 as the name of the dynamic linker, to distinguish it from the
2335 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2337 * Support for CRIS has been removed from ports.
2339 * A new class of installed header has been documented for low-level
2340 platform-specific functionality. PowerPC added the first instance with a
2341 function to provide time base register access. Contributed by Tulio
2342 Magno Quites Machado Filho.
2344 * ix86 configurations now install header files that are consistent with
2345 what x86-64 configurations install. These same header files can be used
2346 for -m32, -m64, or -mx32 builds.
2347 Contributed by H.J. Lu.
2349 * Math library bug fixes. A thorough audit of all open math library bugs was
2350 conducted by Joseph Myers. Significant progress was made on many math
2351 library bugs resulting in more accurate exceptions and function results.
2352 Many thanks to all those that contributed including Andreas Jaeger for his
2353 patch review and work on the x87 trigonometric instruction issues.
2355 * Timezone data is no longer installed. Timezone-related binaries and scripts
2356 will continue to be installed. Users should obtain their timezone data from
2357 their distribution provider or from the tzdata package at
2358 <ftp://munnari.oz.au/pub/>.
2362 * The following bugs are resolved with this release:
2364 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2365 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2366 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2367 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2368 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2369 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2370 13484, 13506, 13515, 13523, 13524, 13538, 13540
2372 * New program pldd to list loaded object of a process
2373 Implemented by Ulrich Drepper.
2375 * Add nss_db support back to glibc. No more dependency on Berkeley db
2376 and support for initgroups lookups.
2377 Implemented by Ulrich Drepper.
2379 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2380 Contributed by HJ Lu.
2382 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2383 Contributed by HJ Lu.
2385 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2386 on x86-32 and x86-64.
2387 Contributed by Liubov Dmitrieva.
2389 * Optimized strchr and strrchr for SSE on x86-32.
2390 Contributed by Liubov Dmitrieva.
2392 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2393 for x86-64 and x86-32.
2394 Contributed by Liubov Dmitrieva.
2396 * New interfaces: scandirat, scandirat64
2397 Implemented by Ulrich Drepper.
2399 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2400 Implemented by Ulrich Drepper.
2402 * nscd now also caches the netgroup database.
2403 Implemented by Ulrich Drepper.
2405 * Integrate libm with gcc's -ffinite-math-only option.
2406 Implemented by Ulrich Drepper.
2408 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2409 to math functions. Implemented by Ulrich Drepper.
2411 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2412 Implemented by Ulrich Drepper.
2414 * New Linux interfaces: process_vm_readv, process_vm_writev
2416 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2417 Implemented by Ulrich Drepper.
2419 * Optimized nearbyint and strcasecmp for PPC.
2420 Implemented by Adhemerval Zanella.
2422 * New locales: bho_IN, unm_US, es_CU, ta_LK
2426 * The following bugs are resolved with this release:
2428 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2429 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2430 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2431 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2432 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2433 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2434 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2435 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2436 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2437 12795, 12811, 12813, 12814, 12841
2439 * The RPC implementation in libc is obsoleted. Old programs keep working
2440 but new programs cannot be linked with the routines in libc anymore.
2441 Programs in need of RPC functionality must be linked against TI-RPC.
2442 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2444 Visible changes of this change include (obviously) the inability to link
2445 programs using RPC functions without referencing the TI-RPC library and the
2446 removal of the RPC headers from the glibc headers.
2447 Implemented by Ulrich Drepper.
2449 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2450 syncfs, setns, sendmmsg
2452 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2453 yue_HK, lij_IT, mhr_RU
2455 * New encodings: CP770, CP771, CP772, CP773, CP774
2457 * New program sotruss to trace calls through PLTs
2458 Implemented by Ulrich Drepper.
2460 * The malloc hook implementation is marked deprecated and will be removed
2461 from the default implementation in the next version. The design never
2462 worked ever since the introduction of threads. Even programs which do
2463 not create threads themselves can use multiple threads created internally.
2467 * The following bugs are resolved with this release:
2469 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2470 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2471 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2472 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2475 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2477 * POWER7 optimizations: memset, memcmp, strncmp
2479 * New optimized string functions for x86-64: strnlen (SSE2),
2480 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2481 Implemented by Ulrich Drepper.
2485 * The following bugs are resolved with this release:
2487 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2488 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2489 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2490 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2491 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2492 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2493 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2494 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2496 * New interfaces: pthread_getname_np, pthread_setname_np
2498 * New Linux interface: recvmmsg
2500 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2502 * The dynamic linker now recognizes supported ABI versions from the
2503 EI_ABIVERSION field in the ELF header.
2504 Implemented by Ulrich Drepper.
2506 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2507 will not be used to fill in password fields in the passwd.byname replies.
2508 Instead it is used to synthesize the shadow.byname table, should it be
2509 missing. This is a useful mode in some installations involving Solaris.
2510 Implemented by Ulrich Drepper.
2512 * New locales: kok_IN, sq_MK, cv_RU
2516 * The following bugs are resolved with this release:
2518 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2519 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2520 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2521 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2522 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2523 10789, 10817, 10823, 10840, 10847
2525 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2527 Implemented by Ulrich Drepper.
2529 * Checking version of longjmp added that fails if an uninitialized stack
2530 frame would be created. Implemented by Ulrich Drepper.
2532 * STT_GNU_IFUNC is now supported in static executables.
2533 Implemented by H.J. Lu.
2535 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2537 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2538 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2539 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2540 Contributed by H.J. Lu.
2542 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2543 Implemented by Ulrich Drepper.
2545 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2547 Contributed by H.J. Lu.
2549 * Support for fma instruction in AVX on x86-64.
2550 Implemented by H.J. Lu and Ulrich Drepper.
2552 * AVX support in x86-64 auditing support in ld.so.
2553 Implemented by H.J. Lu.
2555 * STB_GNU_UNIQUE support added.
2556 Implemented by Ulrich Drepper.
2558 * Implement second fallback mode for optimized DNS lookup for even more
2559 broken environments. If two requests from the same port are not
2560 handled correctly close the socket and open a new one before sending
2561 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2562 can be used to select this mode right away, instead of rediscovering the
2563 necessity is every process again.
2564 Implemented by Ulrich Drepper.
2566 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2567 Implemented by Adam Tkac.
2569 * Optimized iconv conversions for S390x.
2570 Implemented by Andreas Krebbel.
2572 * Using condvars with PI mutexes is now more efficient due to kernel
2573 support for requeueing to PI futexes. NPTL support added for x86-64.
2574 Implemented by Ulrich Drepper.
2576 * New locales: ps_AF, my_MM
2580 * The following bugs are resolved with this release:
2582 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2583 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2584 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2585 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2586 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2588 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2589 Implemented by Ulrich Drepper.
2591 * New Linux interfaces: accept4, fallocate, fallocate64.
2592 Implemented by Ulrich Drepper.
2594 * Correct declarations of string function when used in C++ code. This
2595 could lead to compile errors for invalid C++ code.
2597 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2600 * New POSIX 2008 interface: psiginfo
2601 Implemented by Ulrich Drepper.
2603 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2604 Implemented by Ulrich Drepper.
2606 * Support for selecting between multiple function definitions at runtime
2607 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2609 * The libcrypt library can now use the hash function implementations in
2610 NSS. Implemented by Ulrich Drepper.
2612 * The malloc implementation can be compiled to be less memory efficient
2613 but higher performing in multi-threaded programs.
2614 Implemented by Ulrich Drepper.
2616 * New locales: nan_TW@latin, ks_IN
2618 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2619 Implemented by Ulrich Drepper.
2621 * Extended printf hook support. It is possible to use user-defined types
2622 and extend existing format specifiers.
2623 Implemented by Ulrich Drepper.
2625 * Handling for group shadow files has been added.
2626 Implemented by Ulrich Drepper.
2628 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2629 servers (the case, e.g., for some people using the built-in DNS
2630 server in ADSL modems/routers). There is a once-per-process timeout
2631 in case of a broken server. To avoid it, users can run nscd or put
2632 'options single-request' in /etc/resolv.conf.
2633 Implemented by Ulrich Drepper.
2637 * The following bugs are resolved with this release:
2639 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2640 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2641 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2642 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2643 6968, 6974, 6980, 6995, 7008, 7009, 7029
2645 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2646 up at the same time. Implemented by Ulrich Drepper.
2648 * TLS descriptors for LD and GD on x86 and x86-64.
2649 Implemented by Alexandre Oliva.
2651 * getaddrinfo now handles DCCP and UDPlite.
2652 Implemented by Ulrich Drepper.
2654 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2655 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2656 Implemented by Ulrich Drepper.
2658 * New implementation of memmem, strstr, and strcasestr which is O(n).
2659 Implemented by Eric Blake.
2661 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2663 * Implement "e" option for popen to open file descriptor with the
2664 close-on-exec flag set. Implemented by Ulrich Drepper.
2666 * Implement "b" mode for fmemopen. In this mode writes writes don't
2667 implicitly add a NUL byte and seeks from the end of the buffer really
2668 use the buffer end, not the string length as the basis.
2669 Implemented by Ulrich Drepper.
2671 * Many functions, exported and internal, now atomically set the close-on-exec
2672 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2674 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2676 Implemented by Pravin Satpute.
2678 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2682 * The following bugs are resolved with this release:
2684 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2685 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2686 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2687 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2688 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2689 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2690 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2692 * New locales: bo_CN, bo_IN, shs_CA.
2694 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2696 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2697 Implemented by Pravin Satpute.
2699 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2701 * nscd caches now all timeouts for DNS entries
2702 Implemented by Ulrich Drepper.
2704 * nscd is more efficient and wakes up less often.
2705 Implemented by Ulrich Drepper.
2707 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2708 vdprintf, and obstack_vprintf.
2709 Implemented by Jakub Jelinek.
2711 * Faster memset for x86-64.
2712 Implemented by Harsha Jagasia and H.J. Lu.
2714 * Faster memcpy on x86.
2715 Implemented by Ulrich Drepper.
2717 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2718 Implemented by Ulrich Drepper.
2720 * Faster sqrt and sqrtf implemention for some PPC variants.
2721 Implemented by Steven Munroe.
2725 * The following bugs are resolved with this release:
2727 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2728 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2729 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2730 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2731 5071, 5103, 5104, 5112, 5113, 5184, 5186
2733 * More checking functions: fread, fread_unlocked, open*, mq_open.
2734 Implemented by Jakub Jelinek and Ulrich Drepper.
2736 * Extend fortification to C++. Implemented by Jakub Jelinek.
2738 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2739 by not recognizing 'a' as a modifier when those specs are requested.
2740 Implemented by Jakub Jelinek.
2742 * PPC optimizations to math and string functions.
2743 Implemented by Steven Munroe.
2745 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2746 options to be passed. Implemented by Ulrich Drepper.
2748 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2750 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2751 Implemented by Ulrich Drepper.
2753 * Handle private futexes in the NPTL implementation.
2754 Implemented by Jakub Jelinek and Ulrich Drepper.
2756 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2757 Implemented by Roland McGrath and Ulrich Drepper.
2759 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2761 * SHA-256 and SHA-512 based password encryption.
2762 Implemented by Ulrich Drepper.
2764 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2765 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2768 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2769 Implemented by Ulrich Drepper.
2773 * The following bugs are resolved with this release:
2775 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2776 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2777 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2778 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2779 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2780 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2781 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2784 * New Linux interfaces: epoll_pwait, sched_getcpu.
2786 * New generic interfaces: strerror_l.
2788 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2792 * The following bugs are resolved with this release:
2794 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2795 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2796 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2797 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2798 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2799 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2800 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2801 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2802 3225, 3251, 3252, 3253, 3273
2804 * For Linux, the sorting of addresses returned by getaddrinfo now also
2805 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2806 Implemented by Ulrich Drepper.
2808 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2809 Implemented by Ulrich Drepper.
2811 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2813 * New iconv module for MIK. Contributed by Alexander Shopov.
2815 * For sites with broken group and/or passwd database, the auto-propagate
2816 option of nscd can prevent creating ID lookup entries from the results
2817 of a name lookup and vice versa. This usually is no problem but some
2818 site might have problems with the default behavior.
2819 Implemented by Ulrich Drepper.
2821 * Iterating over entire database in NIS can be slow. With the
2822 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2823 to trade time for memory. The entire database will be read at once.
2824 Implemented by Ulrich Drepper.
2826 * The interfaces introduced in RFC 3542 have been implemented by
2829 * Support for the new ELF hash table format was added by Ulrich Drepper.
2831 * Support for priority inheritance mutexes added by Jakub Jelinek and
2834 * Support for priority protected mutexes added by Jakub Jelinek.
2836 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2837 or_IN, csb_PL, fy_NL, sr_ME.
2841 * More overflow detection functions.
2843 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2844 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2846 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2847 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2848 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2849 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2850 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2851 by Masahide Washizawa.
2853 * It is now possible to install audit libraries for the dynamic linker using
2854 LD_AUDIT. Implemented by Ulrich Drepper.
2856 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2857 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2858 For a libc and libpthread that works well on Linux 2.4 kernels, we
2859 recommend using the stable 2.3 branch.
2861 * The new function open_wmemstream acts like open_memstream,
2862 but uses a wchar_t wide character buffer.
2864 * The new function ppoll is an improved version of poll, similar to pselect.
2866 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2867 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2868 renameat, unlinkat, linkat, symlinkat, readlinkat.
2870 * New Linux kernel system calls: unshare,
2871 inotify_init, inotify_add_watch, inotify_rm_watch.
2873 * The euidaccess function is now also known by the alias eaccess,
2874 for compatibility with some other systems.
2876 * Timezone data updated to 2006b version.
2880 * The following bugs are resolved with this release:
2882 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2883 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2884 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2885 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2886 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2887 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2889 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2891 * As of this release, GCC 4 can be used to compile the C Library.
2893 * Timezone data updated to 2005m version.
2897 * The following bugs are resolved with this release:
2899 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2900 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2901 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2902 777, 787, 821, 822, 823, 825
2904 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2908 * Support for RFC 3678. Real implementations exist only for Linux so far.
2909 Implemented by Ulrich Drepper.
2911 * nscd can now cache entries persistently. Expiring entries are reloaded.
2912 For speedups the cache can be shared in memory with client processes.
2913 Implemented by Ulrich Drepper.
2915 * nscd can now perform SELinux checks.
2916 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2918 * getaddrinfo queries are now cached. Canonical name lookup is performed
2920 Implemented by Ulrich Drepper.
2922 * The nothrow function attribute is used when headers are used by gcc when
2923 compiling C code. This can avoid significant amounts of exception
2926 * The malloc functions perform more error checking and are stricter when
2927 it comes to reacting on errors. The default action is to terminate
2928 the process after showing an error message. Implemented by Ulrich Drepper.
2930 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2931 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2933 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2934 namespaces using the new function dlmopen(). This feature is of course,
2935 like most other dynamic loading functionality, not available in statically
2936 linked applications. Implemented by Ulrich Drepper.
2938 * Low-overhead boundary checking variants of string and some stdio functions
2939 were added. These are to be used in conjunction with a gcc patch by
2940 Jakub Jelinek which adds calls to these functions if possible.
2941 Implemented by Jakub Jelinek and Ulrich Drepper.
2943 * Old code for several operating systems and machine architectures that
2944 have not been in working condition in a long time have been removed from
2945 the main source tree maintained by the GNU C Library's maintainers.
2946 These files are now reside in the separate `ports' source module
2947 that is usable as an add-on when building the library.
2951 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2952 interrogate the dynamic linker, compatible with the Solaris interface.
2954 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2955 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2957 * getifaddrs now uses the netlink interface on Linux to get its information.
2958 Implemented by Thorsten Kukuk.
2960 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2961 Implemented by Ulrich Drepper.
2963 * support for non-executable stacks on x86 has been added. Changes mostly
2966 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2969 * getaddrinfo now performs destination address selection according to
2974 * Thread-safe interfaces for many functions that access locale data
2975 were added in version 2.3, but these features were omitted from NEWS.
2976 Many functions have variants with an `_l' suffix that take a `locale_t'
2977 object as a parameter rather than consulting the current locale.
2978 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2979 create and maintain `locale_t' objects. Additionally, the new function
2980 `uselocale' sets "the current locale" (as used by functions not so
2981 parameterized) set for an individual thread. These features were added
2982 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2984 * The functions getresuid, getresgid, setresuid, and setresgid, which
2985 have long been available on Linux, are now declared in <unistd.h>
2986 and are now also available on the Hurd.
2988 * ELF thread-local storage support (TLS) now works on x86-64.
2990 * The new dynamic string token $LIB is expanded in shared library names.
2991 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2993 * Aldy Hernandez contributed complete software floating point support for
2994 PowerPC machines with no FPU.
2996 * fexecve is implemented on Linux.
2998 * The `btowc' function should work at least twice as fast due to
2999 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3001 * With approriate thread add-ons cancelable functions are now implemented
3002 in libc.so as well. No need to call the function in libpthread. This
3003 change allowed to finally disable the incorrect and expensive handling
3004 of weak definition in ld.so.
3006 * Yet more PLT entries in libc.so have been removed. We finally arrived
3007 at the bare minimum. Startup times improved appropriately.
3009 * Support for the new Linux/x86 system call interface was added. The
3010 AT_SYSINFO auxiliary vector entry is recognized and handled.
3014 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3017 * iconv (the program and the interface) now accepts empty names (excluding
3018 options like //TRANSLIT) to mean "use charset of current locale".
3020 * localedef can now transliterate characters in strings which are not in
3021 the provided charmap. The information from the input locale is used.
3023 * Prelinking support was added for ELF targets. This requires additional
3024 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3026 * Read-only stdio streams now use mmap to speed up operation by eliminating
3027 copying and buffer underflows. To use add 'm' to the mode string of
3028 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3030 * The malloc functions were completely rewritten by Wolfram Gloger based
3031 on Doug Lea's malloc-2.7.0.c.
3033 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3034 implementation of regex.
3036 * Bruno Haible upgraded the iconv and locale implementation to support
3039 * Contents of the LC_* and LANG environment variables in the CEN style are
3040 not recognized anymore. It never was used. Change by Ulrich Drepper.
3042 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3043 thread-local storage (TLS) ABI on some platforms.
3044 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3046 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3047 EUC-JISX0213, and TSCII.
3049 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3050 BSD-compatible interface for getting all network interface addresses.
3051 Implementation for IPv4 by Roland McGrath.
3053 * Loading of locale data is faster due to the introduction of a locale
3054 archive. Implemented by Roland McGrath and Ulrich Drepper.
3056 * Startup times are significantly reduced by not using exported functions
3057 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3060 * Steven Munroe contributed a port to PowerPC64/Linux.
3064 * The Hurd now uses the GNU libio implementation of stdio rather than the
3065 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3067 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3068 with the same behavior as the Linux system call of the same name.
3072 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3073 128-bit long double format.
3075 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3076 IBM1161, and IBM1162 charsets.
3078 * Andreas Jaeger contributed a port to x86-64/Linux.
3080 * Peter Bruin contributed a port to PowerPC/Hurd.
3082 * libc functions using I/O streams now can handle wide-oriented streams
3085 * optimizations in the dynamic linker. Binaries created by recent binutils
3086 versions start up quicker due to reduced time spent on relocations.
3088 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3092 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3093 asinh, atanh, j0 for the 128-bit long double format.
3095 * Bruno Haible updated all the code handling Unicode in some form to
3096 support Unicode 3.1.
3098 * Speed of regex for single-byte locales is back to previous levels.
3099 Patch by Isamu Hasegawa.
3101 * Alpha, SPARC, and IA-64 now also using floating stacks.
3103 * Startup time of internationalized applications greatly improved through
3104 iconv cache. Use iconvconfig to generate the cache file.
3105 Contributed by Ulrich Drepper.
3107 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3108 severe performance problems. Patch by David Mosberger.
3110 * The Hurd port got a lot more functionality like AIO, various stdio
3111 extensions, etc. Mainly done by Roland McGrath.
3113 * mtrace can now lookup symbols in shared libraries.
3117 * Intel's IA-64 math library is largely integrated. It provides fast and
3118 accurate implementatations for most basic and standard math functions
3119 in float, double, and long double format.
3121 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3122 and asin for the 96-bit long double format and asin, log, tan for the
3123 128-bit long double format.
3125 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3126 The basic double functions exist today. Contributed by Abraham Ziv
3127 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3128 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3130 * An asynchronous name lookup library was added. The interface is designed
3131 after POSIX AIO. The proposal was circulated beforehand to get comments.
3132 No negative ones came in. Implemented by Ulrich Drepper.
3134 * Port to S390/64bit contributed by Martin Schwidefsky
3135 <schwidefsky@de.ibm.com>.
3137 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3138 of functions for Linux/IA-64.
3140 * The RPC code is now thread safe. Threads can now use the same service
3141 of different services at the same time. Patch by Eric Norum
3142 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3144 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3145 family of functions for Linux/S390.
3147 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3148 of functions for Linux/x86.
3150 * Port to Linux/CRIS contributed by Axis Communications.
3154 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3155 we can now check for namespace violations and missing declarations. The
3156 result is that almost all headers are now Unix-compliant (as defined in
3157 the upcoming XPG6). The negative side is that some programs might need
3158 corrections, too, if they depend on the incorrect form of the headers in
3159 previous versions which defined too many symbols and included too many
3162 * regex now handles multibyte character sets correctly.
3163 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3165 * iconv (the program) does now conform to the upcoming XPG6 and handles
3166 charmaps. Instead of the charset names the path of charmaps can be
3167 provided and the conversion happens based on this data.
3168 Contributed by Ulrich Drepper.
3170 * The locale program now provides detailed information about the installed
3175 only lists the names of the supported locales
3179 provides details such as country, language, and codeset name.
3180 Contributed by Ulrich Drepper.
3184 * The gencat program now parses the input file according to the charset
3185 selected by the LC_CTYPE category. This is important for stateful
3186 character sets. To make generating catalogs easier there is a way
3187 to overwrite the charset selected by the locale: before the first
3188 message or $ quote line the catalog can contain a line like
3190 $ codeset=ISO-8859-2
3192 to select the charset (ISO-8859-2 in this case).
3194 Implemented by Shinya Hanataka and Ulrich Drepper.
3196 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3197 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3198 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3200 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3201 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3203 * The LANGUAGE environment variable is now ignored unless the locale is
3204 changed from the default "C" locale.
3206 * The usual bug fixes.
3210 * Greg McGary added runtime support for bounds checking using gcc's
3211 new -fbounded-pointers option. ix86 target is complete. PowerPC
3214 * Thorsten Kukuk added secure mode support to nscd.
3216 * The Berkeley db libraries have been removed.
3218 Related, the nss_db module is now in a separate package since it
3219 obviously requires a database library being available.
3221 * Wide character I/O streams implemented by Ulrich Drepper.
3223 * Functions from the extended socket API added by Ulrich Drepper.
3225 * Functions feenableexcept and fedisableexcept to control the
3226 behaviour of individual exceptions have been added by Andreas Jaeger.
3228 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3230 * The resolver code has been updated from bind 8.2.3-T5B which supports
3231 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3234 This change could in some situations effect backward compatibility. Since
3235 now `_res' is a thread-local instead of a global variable, modifying it
3236 in one thread does not have any effect in other threads.
3238 The resolver library was also extended to allow IPv6 as the transport
3239 protocol for the requests. This work was done by Stig Venaas.
3241 * Compatibility code for K&R C compilers has been removed from the
3242 header files. A ISO C compiler is needed to use the library
3243 (conforming to either C89 or C99 standard).
3245 * Complete rewrite of the localedef program to support multibyte character
3246 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3247 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3248 Implemented by Ulrich Drepper.
3250 Bruno Haible significantly improved the generation and use of the data
3251 structures for the wide character tables.
3253 * Plural handling in gettext implemented by Ulrich Drepper.
3255 * The utmp daemon has been removed.
3257 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3259 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3262 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3264 * POSIX barriers implemented by Kaz Kylheku.
3266 * POSIX spawn function family implemented by Ulrich Drepper.
3268 * POSIX spinlocks are now available.
3270 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3272 * the configure option --enable-kernel=X.Y.Z allows to strip out
3273 compatibility for kernel versions before X.Y.Z. This is currently only
3274 implemented for Linux.
3276 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3277 field sin6_scope_id. This means that all programs using IPv6 should be
3278 recompiled. Don't expect binary compatibility with previous glibc
3281 * various conversion modules for IBM character sets contributed by
3284 * IA-64 port by Jes Sorensen and HJ Lu.
3298 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3300 * New xdr functions are added; some rpc functions are now 64bit clean.
3302 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3304 * Fixed known incompatibilities with glibc 2.0.
3306 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3308 * Optimized a number of functions (especially the ELF dynamic loader).
3310 * Update timezone data files.
3312 * lots of charmaps corrections
3314 * some new locale definitions and charmaps
3319 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3320 conform to POSIX member type requirements. Symbol versions have been
3321 adjusted as needed within the library, and for direct use by applications,
3322 but there is potential for problems if third-party libraries use
3323 struct timeval as part of their interface. This does not present
3324 a problem for X and other "essential" system libraries.
3326 * An additional locale model to support C++ Standard Library locale
3327 model and probably more was implemented by Ulrich Drepper.
3329 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3332 * Miles Bader provided the `argp' function family to support hierachical
3333 command line argument parsing, layered on top of getopt.
3335 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3337 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3340 * scanf recognizes the %a and %A format for scanning floating point numbers.
3342 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3343 information and interfaces for the available integer types.
3345 * about 130 new math functions were added to implement the ISO C9x math
3348 * the new header <complex.h> contains definitions of the complex math
3349 functions from ISO C 9X.
3351 * the new header <tgmath.h> defines generic macros to use complex or
3352 real valued functions.
3354 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3356 * Andreas Jaeger provided a test suite for the math library.
3358 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3360 * Ulrich Drepper added character set conversion functions (iconv).
3362 * Optimized string functions have been added.
3364 * The localedata addon is now part of glibc.
3366 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3368 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3369 daemon for NSS (nscd).
3371 Missing a better place here are some numbers on improvements. Under
3372 Linux 2.1.125 un-tar-ing the kernel sources takes
3376 using local files 12.19s 6.88s 22.91s
3378 using NIS 13.92s 8.91s 26.34s
3380 using NIS & nscd 10.37s 7.34s 25.30s
3382 using NIS+ 27.57s 30.37s 640.46s
3384 using NIS+ & nscd 10.25s 7.83s 26.51s
3386 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3388 Keep in mind that non-namelookup related operations dominate above times.
3389 It was just a common complain that using NIS+ unpacking the kernel is
3392 [1] The old nscd implementation is not available anymore since it was
3393 distributed with glibc up to version 2.0.98 and thus is now replaced.
3395 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3397 * Mark Kettenis provided a Hesiod NSS module.
3399 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3400 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3402 * rcmd can now handle netgroups (Dick Streefland).
3404 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3407 * Support for the IPv6 protocol has been added to the socket API, as per the
3408 latest draft standards.
3410 * Support for Linux 2.2 has been added.
3412 * Interface changes relative to the latest 2.0.x release:
3413 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3414 addseverity NEW: Unix98
3415 alphasort64 NEW: LFS
3416 argp_err_exit_status NEW: argp, GNU ext
3417 argp_error NEW: argp, GNU ext
3418 argp_failure NEW: argp, GNU ext
3419 argp_help NEW: argp, GNU ext
3420 argp_parse NEW: argp, GNU ext
3421 argp_program_bug_address NEW: argp, GNU ext
3422 argp_program_version NEW: argp, GNU ext
3423 argp_program_version_hook NEW: argp, GNU ext
3424 argp_state_help NEW: argp, GNU ext
3425 argp_usage NEW: argp, GNU ext
3426 authdes_create NEW: Secure RPC
3427 authdes_getucred NEW: Secure RPC
3428 authdes_pk_create NEW: Secure RPC
3429 backtrace NEW: GNU ext.
3430 backtrace_symbols NEW: GNU ext.
3431 backtrace_symbols_fd NEW: GNU ext.
3433 cacosf NEW: ISO C 9x
3434 cacosh NEW: ISO C 9x
3435 cacoshf NEW: ISO C 9x
3436 cacoshl NEW: ISO C 9x
3437 cacosl NEW: ISO C 9x
3444 casinf NEW: ISO C 9x
3445 casinh NEW: ISO C 9x
3446 casinhf NEW: ISO C 9x
3447 casinhl NEW: ISO C 9x
3448 casinl NEW: ISO C 9x
3450 catanf NEW: ISO C 9x
3451 catanh NEW: ISO C 9x
3452 catanhf NEW: ISO C 9x
3453 catanhl NEW: ISO C 9x
3454 catanl NEW: ISO C 9x
3455 cbc_crypt NEW: Secure RPC
3459 ccoshf NEW: ISO C 9x
3460 ccoshl NEW: ISO C 9x
3466 cimagf NEW: ISO C 9x
3467 cimagl NEW: ISO C 9x
3468 clearerr_locked REMOVED
3469 clntunix_create NEW: sunrpc ext
3471 clog10 NEW: ISO C 9x
3472 clog10f NEW: ISO C 9x
3473 clog10l NEW: ISO C 9x
3483 cprojf NEW: ISO C 9x
3484 cprojl NEW: ISO C 9x
3486 crealf NEW: ISO C 9x
3487 creall NEW: ISO C 9x
3492 csinhf NEW: ISO C 9x
3493 csinhl NEW: ISO C 9x
3496 csqrtf NEW: ISO C 9x
3497 csqrtl NEW: ISO C 9x
3501 ctanhf NEW: ISO C 9x
3502 ctanhl NEW: ISO C 9x
3504 des_setparity NEW: Secure RPC
3505 ecb_crypt NEW: Secure RPC
3506 endutxent NEW: Unix98
3508 exp10f NEW: ISO C 9x
3509 exp10l NEW: ISO C 9x
3513 fattach NEW: STREAMS
3514 fdetach NEW: STREAMS
3518 feclearexcept NEW: ISO C 9x
3519 fegetenv NEW: ISO C 9x
3520 fegetexceptflag NEW: ISO C 9x
3521 fegetround NEW: ISO C 9x
3522 feholdexcept NEW: ISO C 9x
3524 feraiseexcept NEW: ISO C 9x
3525 ferror_locked REMOVED
3526 fesetenv NEW: ISO C 9x
3527 fesetexceptflag NEW: ISO C 9x
3528 fesetround NEW: ISO C 9x
3529 fetestexcept NEW: ISO C 9x
3530 feupdateenv NEW: ISO C 9x
3531 fflush_locked REMOVED
3535 fgets_unlocked NEW: GNU ext.
3536 fileno_locked REMOVED
3548 fputc_locked REMOVED
3549 fputs_unlocked NEW: GNU ext.
3550 fread_unlocked NEW: GNU ext.
3555 fstatvfs NEW: Unix98
3559 ftruncate64 NEW: LFS
3561 fwrite_unlocked NEW: GNU ext.
3562 gai_strerror NEW: IPv6
3566 getchar_locked REMOVED
3568 getdate_err NEW: Unix98
3569 getdate_r NEW: GNU ext.
3571 getnameinfo NEW: IPv6
3572 getnetname NEW: Secure RPC
3573 getpmsg NEW: STREAMS
3574 getpt NEW: Unix98 PTY
3575 getrlimit64 NEW: LFS
3576 getutxent NEW: Unix98
3577 getutxid NEW: Unix98
3578 getutxline NEW: Unix98
3579 glob64 NEW: GNU ext.
3580 globfree64 NEW: GNU ext.
3581 gnu_get_libc_release NEW: GNU ext.
3582 gnu_get_libc_version NEW: GNU ext.
3583 grantpt NEW: Unix98 PTY
3584 host2netname NEW: Secure RPC
3586 iconv_close NEW: iconv
3587 iconv_open NEW: iconv
3588 if_freenameindex NEW: IPv6
3589 if_indextoname NEW: IPv6
3590 if_nameindex NEW: IPv6
3591 if_nametoindex NEW: IPv6
3592 in6addr_any NEW: IPv6
3593 in6addr_loopback NEW: IPv6
3594 inet6_isipv4mapped NEW: IPv6
3595 isastream NEW: STREAMS
3596 iswblank NEW: GNU ext.
3597 key_decryptsession NEW: Secure RPC
3598 key_decryptsession_pk NEW: Secure RPC
3599 key_encryptsession NEW: Secure RPC
3600 key_encryptsession_pk NEW: Secure RPC
3601 key_gendes NEW: Secure RPC
3602 key_get_conv NEW: Secure RPC
3603 key_secretkey_is_set NEW: Secure RPC
3604 key_setnet NEW: Secure RPC
3605 key_setsecret NEW: Secure RPC
3606 llrint NEW: ISO C 9x
3607 llrintf NEW: ISO C 9x
3608 llrintl NEW: ISO C 9x
3609 llround NEW: ISO C 9x
3610 llroundf NEW: ISO C 9x
3611 llroundl NEW: ISO C 9x
3616 lrintf NEW: ISO C 9x
3617 lrintl NEW: ISO C 9x
3618 lround NEW: ISO C 9x
3619 lroundf NEW: ISO C 9x
3620 lroundl NEW: ISO C 9x
3622 makecontext NEW: Unix98
3623 mempcpy NEW: GNU ext.
3626 modify_ldt NEW: kernel
3630 nearbyint NEW: ISO C 9x
3631 nearbyintf NEW: ISO C 9x
3632 nearbyintl NEW: ISO C 9x
3633 netname2host NEW: Secure RPC
3634 netname2user NEW: Secure RPC
3635 nexttoward NEW: ISO C 9x
3636 nexttowardf NEW: ISO C 9x
3637 nexttowardl NEW: ISO C 9x
3641 passwd2des NEW: Secure RPC
3643 pow10f NEW: GNU ext.
3644 pow10l NEW: GNU ext.
3647 printf_size NEW: GNU ext.
3648 printf_size_info NEW: GNU ext.
3649 profil_counter REMOVED
3650 pthread_mutexattr_getkind_np REPLACED
3651 pthread_mutexattr_setkind_np REPLACED
3652 ptsname NEW: Unix98 PTY
3653 ptsname_r NEW: Unix98 PTY
3655 putchar_locked REMOVED
3656 putgrent NEW: GNU ext.
3658 putpmsg NEW: STREAMS
3659 pututxline NEW: Unix98
3663 readdir64_r NEW: LFS
3664 remquo NEW: ISO C 9x
3665 remquof NEW: ISO C 9x
3666 remquol NEW: ISO C 9x
3668 roundf NEW: ISO C 9x
3669 roundl NEW: ISO C 9x
3671 scalbln NEW: ISO C 9x
3672 scalblnf NEW: ISO C 9x
3673 scalblnl NEW: ISO C 9x
3675 sendfile NEW: kernel
3676 setcontext NEW: Unix98
3677 setrlimit64 NEW: LFS
3678 setutxent NEW: Unix98
3680 sigignore NEW: Unix98
3681 sigqueue NEW: POSIX.1b
3682 sigrelse NEW: Unix98
3683 sigset NEW: POSIX.1b
3684 sigtimedwait NEW: POSIX.1b
3685 sigwaitinfo NEW: POSIX.1b
3686 sincos NEW: GNU ext.
3687 sincosf NEW: GNU ext.
3688 sincosl NEW: GNU ext.
3692 strcasestr NEW: GNU ext.
3693 strtoimax NEW: ISO C 9x
3694 strtoumax NEW: ISO C 9x
3695 strverscmp NEW: GNU ext.
3696 svcauthdes_stats NEW: Secure RPC
3697 svcunix_create NEW: sunrpc etx
3698 svcunixfd_create NEW: sunrpc ext
3699 swapcontext NEW: Unix98
3700 tcgetsid NEW: Unix98 PTY
3701 tdestroy NEW: GNU ext.
3702 tgamma NEW: ISO C 9x
3703 tgammaf NEW: ISO C 9x
3704 tgammal NEW: ISO C 9x
3708 truncf NEW: ISO C 9x
3709 truncl NEW: ISO C 9x
3711 unlockpt NEW: Unix98 PTY
3712 updwtmpx NEW: Unix98
3713 user2netname NEW: Secure RPC
3714 utmpxname NEW: Unix98
3715 versionsort NEW: GNU ext.
3716 versionsort64 NEW: GNU ext.
3718 wcscasecmp NEW: GNU ext.
3719 wcsncasecmp NEW: GNU ext.
3720 wcsnlen NEW: GNU ext.
3721 wcstoimax NEW: ISO C 9x
3722 wcstoll NEW: ISO C 9x
3723 wcstoull NEW: ISO C 9x
3724 wcstoumax NEW: ISO C 9x
3726 wordexp NEW: POSIX.2
3727 wordfree NEW: POSIX.2
3728 write_profiling REMOVED
3729 xdecrypt NEW: Secure RPC
3730 xdr_authdes_cred NEW: Secure RPC
3731 xdr_authdes_verf NEW: Secure RPC
3732 xdr_cryptkeyarg NEW: Secure RPC
3733 xdr_cryptkeyarg2 NEW: Secure RPC
3734 xdr_cryptkeyres NEW: Secure RPC
3735 xdr_getcredres NEW: Secure RPC
3736 xdr_key_netstarg NEW: Secure RPC
3737 xdr_key_netstres NEW: Secure RPC
3738 xdr_keybuf NEW: Secure RPC
3739 xdr_keystatus NEW: Secure RPC
3740 xdr_netnamestr NEW: Secure RPC
3741 xdr_sizeof NEW: Secure RPC
3742 xdr_unixcred NEW: sunrpc ext
3743 xencrypt NEW: Secure RPC
3744 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3755 * inet_ntoa is thread-safe
3757 * updwtmp is moved from libutil to libc
3759 * rewrite of cbrt function
3761 * update of timezone data
3775 * add atoll function
3777 * fix complex problems in Berkeley DB code
3779 * fix math functions
3783 * fixed lots of header problems (especially Linux/GNU specific)
3785 * dynamic loader preserves all registers
3787 * Roland McGrath provided support for handling of auxiliary objects in
3788 the ELF dynamic loader.
3790 * support for parallel builds is improved
3794 * GNU extensions are no longer declared by default. To enable them you
3795 must define the macro `_GNU_SOURCE' in your program or compile with
3798 * The library has changed from using GNU ld symbol aliases to using weak
3799 symbols where available. The ELF object file format supports weak
3800 symbols; GNU ld also supports weak symbols in the a.out format. (There
3801 is also now support for other GNU ld extensions in ELF. Use the
3802 `--with-elf' option to configure to indicate you have ELF, and
3803 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3804 of many files which contained only symbol aliases, reducing the size of
3805 the source and the compiled library; many other files were renamed to
3806 less cryptic names previously occupied by the symbol alias files.
3807 There is a new header file <elf.h> for programs which operate on
3808 files in the ELF format.
3810 * Converted to Autoconf version 2, so `configure' has more options.
3811 Run `configure --help' to see the details.
3813 * The library can now be configured to build profiling, highly-optimized
3814 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3815 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3816 `configure' enable building these extra libraries. The shared library is
3817 built by default when using both ELF and GNU ld. When shared libraries
3818 are enabled, the new library `-ldl' is available for arbitrary run-time
3819 loading of shared objects; its interface is defined in <dlfcn.h>. The
3820 new header file <link.h> gives access to the internals of the run-time
3821 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3822 application of same name on other systems and it provides information
3823 about dynamically linked binaries.
3825 * The C library now provides the run-time support code for profiling
3826 executables compiled with `-pg'. Programs can control the profiling code
3827 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3828 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3829 the support for this file format was contributed by David Mosberger-Tang.
3831 * The math code has been replaced with a math library based on fdlibm from
3832 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3833 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3834 functions. The math functions now reside in a separate library, so
3835 programs using them will need to use `-lm' their linking commands.
3837 * John C. Bowman contributed optimized ix87 assembler inline functions.
3839 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3840 mechanism similar to that found in Solaris 2. This is now used for the
3841 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3842 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3843 file controls what services are used for each individual database. This
3844 works by loading shared libraries with names specified in `nsswitch.conf',
3845 so service modules can be changed or added at any time without even
3846 relinking any program. Currently there are the file, db, and NIS based
3847 NSS services available.
3849 * The new functions `strtoq' and `strtouq' parse integer values from
3850 strings, like `strtol' and `strtoul', but they return `long long int' and
3851 `unsigned long long int' values, respectively (64-bit quantities).
3853 * The new functions `strtof' and `strtold' parse floating-point values from
3854 strings, like `strtod', but they return `float' and `long double' values,
3855 respectively (on some machines `double' and `long double' are the same).
3857 * Ulrich Drepper has contributed new implementations of the floating-point
3858 printing and reading code used in the `printf' family of functions and
3859 `strtod', `strtof', and `strtold'. These new functions are perfectly
3860 accurate, and much faster than the old ones.
3862 * The implementation of the POSIX locale model was completely rewritten by
3863 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3864 to compile the POSIX locale definition.
3866 * The former dummy implementations of the strcoll and strxfrm function are
3867 now replaced by fully functional code contributed by Ulrich Drepper. The
3868 collation information comes from the POSIX locale definitions.
3870 * The new header <langinfo.h> defines an interface for accessing
3871 various locale-dependent data (using the locale chosen with `setlocale').
3873 * Ulrich Drepper has contributed a new suite of functions for operation on
3874 wide-character and multibyte-character strings, in <wchar.h>;
3875 and classification and case conversion of wide characters, in <wctype.h>.
3876 These new functions are conforming to the ISO C, Amendement 1 specification.
3878 * There is now a second implementation of the standard I/O library available.
3879 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3880 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3882 * You can now use positional parameter specifications in format strings
3883 for the `printf' and `scanf' families of functions. For example,
3884 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3885 ``Number 6, Mr Jones''. This is mainly useful when providing different
3886 format strings for different languages, whose grammars may dictate
3887 different orderings of the values being printed. To support this
3888 feature, the interface for `register_printf_handler' has changed; see
3889 the header file <printf.h> for details.
3891 * The `printf' and `scanf' families of functions now understand a new
3892 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3893 to group numbers as indicated by the locale; for `scanf' and friends, this
3894 says to accept as valid only a number with all the proper grouping
3895 separators in the right places. In the default "C" locale, numbers are
3896 not grouped; but locales for specific countries will define the usual
3897 conventions (i.e. separate thousands with `,' in the US locale).
3899 * The pgrp functions have been regularized, slightly incompatibly but much
3900 less confusingly. The core functions are now `getpgid' and `setpgid',
3901 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3902 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3903 provided for compatibility. There is no longer an incompatible `getpgrp'
3904 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3906 * The new header file <fts.h> and suite of functions simplify programs that
3907 operate on directory trees. This code comes from 4.4 BSD.
3909 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3910 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3911 NSS scheme used in glibc.
3913 * There is a new malloc debugging hook `__memalign_hook'.
3915 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3916 `unsigned int' in <sys/types.h>. These are for compatibility only and
3917 their use is discouraged.
3919 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3920 done differently, so that it works even without GNU ld.
3922 * New function `euidaccess' checks allowed access to a file like `access',
3923 but using the effective IDs instead of the real IDs.
3925 * The time zone data files have been updated for the latest and greatest
3926 local time conventions of the countries of the world.
3928 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3931 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3932 for formatting floating-point numbers. They are provided only for
3933 compatibility; new programs should use `sprintf' instead. There are
3934 also equivalent function for the `long double' floating-point type and
3935 all functions also exist in a reentrant form.
3937 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3938 functions for maintaining the login-record files (primarily of use to
3939 system programs such as `login'), and convenient functions for
3940 allocating and initializing a pseudo-terminal (pty) device.
3942 * Ulrich Drepper has contributed new support for System V style
3943 shared memory and IPC on systems that support it.
3945 * Ulrich Drepper has contributed several miscellaneous new functions found
3946 in System V: The `hsearch' family of functions provide an effective
3947 implementation of hash tables; `a64l' and `l64a' provide a very simple
3948 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3951 * Ulrich Drepper has contributed new reentrant counterparts for the
3952 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3954 * Ulrich Drepper has contributed new, highly-optimized versions of several
3955 string functions for the i486/Pentium family of processors.
3957 * Ulrich Drepper has updated the Linux-specific code, based largely
3958 on work done in Hongjiu Lu's version of GNU libc for Linux.
3959 The GNU library now supports Linux versions 2.0.10 and later,
3960 using the ELF object file format (i[3456]86-*-linux).
3962 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3964 * David Mosberger-Tang and Richard Henderson have ported the C library
3965 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3966 dynamic linking support for ELF/Alpha.
3968 * Richard Henderson contributed several Alpha optimized assembler function
3969 for arithmetic and string handling.
3971 * Ulrich Drepper has contributed a new set of message catalog functions to
3972 support multiple languages using the <libintl.h> interface, for use with
3973 his new package GNU gettext. Translation volunteers have contributed
3974 catalogs of the library's messages in Spanish, German, and Korean.
3976 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3977 program and the `catgets' function for reading the catalog files it
3978 creates. (The <libintl.h> interface is preferred; we include the
3979 <nl_types.h> interface using `catgets' only for source compatibility with
3980 programs already written to use it.)
3982 * New header file <values.h> gives SVID-compatible names for <limits.h>
3985 * Various new macros, declarations, and small header files for compatibility
3988 * New function `group_member' is a convenient way to check if a process has
3989 a given effective group ID.
3991 * When using GCC 2.7 and later, the socket functions are now declared in a
3992 special way so that passing an argument of type `struct sockaddr_in *',
3993 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3994 `struct sockaddr *' type, does not generate a type-clash warning.
3996 * New function `error' declared in header file <error.h> is a convenient
3997 function for printing error messages and optionally exiting; this is the
3998 canonical function used in GNU programs. The new functions `err', `warn',
3999 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4000 doing the same thing.
4002 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4003 POSIX.2 `glob' function to do ~ and {...} expansion.
4005 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4006 `clearenv' which is used in POSIX.9 is also available.
4008 * New function `getsid' returns session ID number on systems that support it.
4010 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4011 files <db.h> and <mpool.h> provide a rich set of functions for several
4012 types of simple databases stored in memory and in files, and <ndbm.h> is
4013 an old `ndbm'-compatible interface using the `db' functions. Link with
4014 `-ldb' to get these functions.
4016 * New macro `strdupa' copies a string like `strdup', but uses local stack
4017 space from `alloca' instead of dynamic heap space from `malloc'.
4019 * New function `strnlen' is like `strlen' but searches only a given maximum
4020 number of characters for the null terminator. `stpncpy', `strndup' and
4021 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4024 * New function `statfs' in header <sys/statfs.h>.
4026 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4027 provide convenient functions for operating on blocks of null-terminated
4030 * A new suite of functions in <utmp.h> handle all the details of reading
4031 and writing the utmp file.
4033 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4036 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4037 completely POSIX compliant and also implemented the extended functionality
4038 to handle alternate digit representation and alternate era date formats.
4040 * Ulrich Drepper provided an implementation of the `strptime' function
4041 defined in XPG4.2 which transforms a string into a `struct tm' value.
4043 * Paul Eggert provided the tzselect shell script as part of the timezone
4044 code. The shell script makes it easy to select the correct timezone
4047 * The implementation of the malloc family of functions is completely replaced
4048 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4049 The implementation uses the mmap function (if available) and it is
4050 optimized for the use in multi threaded programs.
4052 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4053 functions. This new functionality is usable by specifying a special
4054 salt string and it is compatible with implementation on *BSD systems.
4056 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4057 `getsubopt' to handle second level command line options, `bsd_signal'
4058 to access BSD style `signal' functionality, the obsolete `regexp' style
4061 * the `lchown' function is available on system which support this
4064 * The implementation of the shadow password handling function was contributed
4067 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4069 * POSIX.1g support was added. The <sys/select.h> header is available,
4070 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4071 implementation of `getaddrinfo'.
4075 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4077 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4078 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4080 * The new function `daemon' from 4.4 BSD is useful for server programs that
4081 want to put themselves in the background.
4083 * Joel Sherrill has contributed support for several standalone boards that
4084 run without an operating system.
4086 * `printf', `scanf' and friends now accept a `q' type modifier for long
4087 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4089 * All of the code taken from BSD (notably most of the math and networking
4090 routines) has been updated from the BSD 4.4-Lite release.
4092 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4094 * The new functions `getdomainname' and `setdomainname' fetch or change the
4095 YP/NIS domain name. These are system calls which exist on systems which
4098 * The time zone data files have been updated for the latest international
4101 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4102 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4106 * The C library now includes support for Sun RPC, from Sun's free
4107 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4108 are included. (There is still no support for YP.)
4110 * Tom Quinn has contributed a port of the C library to SGI machines running
4111 Irix 4 (mips-sgi-irix4).
4113 * The new `lockf' function is a simplified interface to the locking
4114 facilities of `fcntl', included for compatibility.
4116 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4118 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4120 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4123 * The error message printed by `assert' for a failed assertion now includes
4124 the name of the program (if using GNU ld) and the name of the calling
4125 function (with versions of GCC that support this).
4127 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4129 * The library now includes the <sys/mman.h> header file and memory
4130 management functions `mmap', `munmap', `mprotect', `msync', and
4131 `madvise', on systems that support those facilities.
4133 * The interface for `mcheck' has changed slightly: the function called to
4134 abort the program when an allocation inconsistency is detected now takes
4135 an argument that indicates the type of failure. The new function
4136 `mprobe' lets you request a consistency check for a particular block at
4137 any time (checks are normally done only when you call `free' or `realloc'
4140 * It is now possible to easily cross-compile the C library, building on one
4141 system a library to run on another machine and/or operating system. All
4142 you need to do is set the variable `HOST_CC' in `configparms' to the
4143 native compiler for programs to run on the machine you are building on (a
4144 few generator programs are used on Unix systems); set `CC' to the
4147 * The new function `fexecve' (only implemented on the GNU system) executes
4148 a program file given a file descriptor already open on the file.
4152 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4153 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4155 * You can set the variable `libprefix' in `configparms' to specify a prefix
4156 to be prepended to installed library files; this makes it easy to install
4157 the GNU C library to be linked as `-lgnuc' or whatever.
4159 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4160 copies a limited number of characters from a string, and returns the
4161 address of the last character written.
4163 * You no longer need to check for whether the installed `stddef.h' is
4164 compatible with the GNU C library. configure now checks for you.
4166 * You can now define a per-stream `fileno' function to convert the
4167 stream's cookie into an integral file descriptor.
4169 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4170 allocates zero bytes of storage, and returns a unique pointer which
4171 you can pass to `realloc' or `free'. The behavior is undefined if
4172 you dereference this pointer.
4174 * The C library now runs on Sony NEWS m68k machines running either
4175 NewsOS 3 or NewsOS 4.
4177 * The new `syscall' function is a system-dependent primitive function
4178 for invoking system calls. It has the canonical behavior on Unix
4179 systems, including unreliable return values for some calls (such as
4180 `pipe', `fork' and `getppid').
4182 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4183 to `EAGAIN', which is the preferred name. On systems whose kernels
4184 use two distinct codes, the C library now translates EWOULDBLOCK to
4185 EAGAIN in every system call function.
4189 * The GNU C Library Reference Manual is now distributed with the library.
4190 `make dvi' will produce a DVI file of the printed manual.
4191 `make info' will produce Info files that you can read on line using C-h i
4192 in Emacs or the `info' program.
4193 Please send comments on the manual to bug-glibc-manual@gnu.org.
4195 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4197 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4199 * Jason Merrill has contributed a port to the Sequent Symmetry running
4200 Dynix version 3 (i386-sequent-dynix).
4202 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4203 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4205 * New function `memory_warnings' lets you arrange to get warnings when
4206 malloc is running out of memory to allocate, like Emacs gives you.
4208 * The C library now contains the relocating allocator used in Emacs 19 for
4209 its editing buffers. This allocator (ralloc) minimizes allocation
4210 overhead and fragmentation by moving allocated regions around whenever it
4211 needs to. You always refer to a ralloc'd region with a "handle" (a
4212 pointer to a pointer--an object of type `void **').
4214 * There is a new `printf' format: `%m' gives you the string corresponding
4215 to the error code in `errno'.
4217 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4218 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4219 pass, the `a' modifier says to fill in a `char **' you pass with a
4222 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4223 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4224 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4226 * `mkstemp' is a traditional Unix function to atomically create and open a
4227 uniquely-named temporary file.
4231 * The standard location for the file that says what the local timezone is
4232 has changed again. It is now `/usr/local/etc/localtime' (or more
4233 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4235 * The distribution no longer contains any files with names longer than 14
4238 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4239 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4241 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4245 * The configuration process has changed quite a bit. The `configure'
4246 script is now used just like the configuration scripts for other GNU
4247 packages. The `sysdeps' directory hierarchy is much rearranged.
4248 The file `INSTALL' explains the new scheme in detail.
4250 * The header files no longer need to be processed into ANSI C and
4251 traditional C versions. There is just one set of files to install, and
4252 it will work with ANSI or old C compilers (including `gcc -traditional').
4254 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4255 MIPS DECStation running Ultrix 4.
4257 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4258 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4259 made itself into a shared library.
4261 * Yet further improved support for the i386, running 4.3 BSD-like systems
4262 (such as Mach 3 with the Unix single-server), or System V.
4264 * New function `strncasecmp' to do case-insensitive string comparison
4265 with limited length.
4267 * New function `strsep' is a reentrant alternative to `strtok'.
4269 * New functions `scandir' and `alphasort' for searching directories.
4271 * New function `setenv' is a better interface to `putenv'.
4273 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4274 function for traversing a directory tree.
4276 * The GNU obstack package is now also part of the C library.
4277 The new function `open_obstack_stream' creates a stdio stream that
4278 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4279 formatted output directly to an obstack.
4281 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4282 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4284 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4286 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4287 things to your strings.
4289 * There are some new test programs: test-fseek, testmb, and testrand.
4291 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4292 These ports are not finished, but are a good starting place for really
4293 supporting those systems.
4295 * `/etc/localtime' is now the standard location for the file that says what
4296 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4297 This follows the general principle that `/etc' is the place for all local
4298 configuration files.
4300 * The C library header files now use `extern "C"' when used by the C++
4301 compiler, so the C library should now work with C++ code.
4303 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4304 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4307 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4308 function which uses the `merge sort' algorithm, and is said to be
4309 significantly faster than the old GNU `qsort' function. Merge sort is
4310 now the standard `qsort' function. The new algorithm can require a lot
4311 of temporary storage; so, the old sorting function is called when the
4312 required storage is not available.
4314 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4315 provides the Unix `crypt' function, plus some other entry points.
4317 * The code and header files taken from 4.4 BSD have been updated with the
4318 latest files released from Berkeley.
4320 ----------------------------------------------------------------------
4321 Copyright information:
4323 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4325 Permission is granted to anyone to make or distribute verbatim copies
4326 of this document as received, in any medium, provided that the
4327 copyright notice and this permission notice are preserved,
4328 thus giving the recipient permission to redistribute in turn.
4330 Permission is granted to distribute modified versions
4331 of this document, or of portions of it,
4332 under the above conditions, provided also that they
4333 carry prominent notices stating who last changed them.
4336 version-control: never