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 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
66 * The ldconfig utility now processes `include' directives using the C/POSIX
67 collation ordering. Previous glibc versions used locale-specific
68 ordering, the change might break systems that relied on that.
70 Deprecated and removed features, and other changes affecting compatibility:
72 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
73 defined by <sys/ptrace.h>.
75 * libm no longer supports SVID error handling (calling a user-provided
76 matherr function on error) or the _LIB_VERSION variable to control error
77 handling. (SVID error handling and the _LIB_VERSION variable still work
78 for binaries linked against older versions of the GNU C Library.) The
79 libieee.a library is no longer provided. math.h no longer defines struct
80 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
81 TLOSS, PLOSS and HUGE.
83 * The libm functions pow10, pow10f and pow10l are no longer supported for
84 new programs. Programs should use the standard names exp10, exp10f and
85 exp10l for these functions instead.
87 * The mcontext_t type is no longer the same as struct sigcontext. On
88 platforms where it was previously the same, this changes the C++ name
89 mangling for interfaces involving this type.
91 * The add-ons mechanism for building additional packages at the same time as
92 glibc has been removed. The --enable-add-ons configure option is now
95 * The --without-fp configure option is now ignored. Whether hardware
96 floating-point instructions are used is now configured based on whether
97 the compiler used at configure time (without any options implied by a
98 --with-cpu= configure option) uses such instructions.
100 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
101 the specified string can be parsed as a domain name.
103 * In the malloc_info output, the <heap> element may contain another <aspace>
104 element, "subheaps", which contains the number of sub-heaps.
106 * The libresolv function p_secstodate is no longer supported for new
109 * The tilepro-*-linux-gnu configuration is no longer supported.
111 Changes to build and runtime requirements:
113 * bison version 2.7 or later is required to generate code in the 'intl'
116 Security related changes:
118 CVE-2009-5064: The ldd script would sometimes run the program under
119 examination directly, without preventing code execution through the
120 dynamic linker. (The glibc project disputes that this is a security
121 vulnerability; only trusted binaries must be examined using the ldd
124 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
125 suffered from a one-byte overflow during ~ operator processing (either
126 on the stack or the heap, depending on the length of the user name).
127 Reported by Tim Rühsen.
129 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
130 would sometimes fail to free memory allocated during ~ operator
131 processing, leading to a memory leak and, potentially, to a denial
134 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
135 without GLOB_NOESCAPE, could write past the end of a buffer while
136 unescaping user names. Reported by Tim Rühsen.
138 CVE-2017-17426: The malloc function, when called with an object size near
139 the value SIZE_MAX, would return a pointer to a buffer which is too small,
140 instead of NULL. This was a regression introduced with the new malloc
141 thread cache in glibc 2.26. Reported by Iain Buclaw.
143 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
144 to the allocation of too much memory. (This is not a security bug per se,
145 it is mentioned here only because of the CVE assignment.) Reported by
148 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
149 of the number of search path components. (This is not a security
150 vulnerability per se because no trust boundary is crossed if the fix for
151 CVE-2017-1000366 has been applied, but it is mentioned here only because
152 of the CVE assignment.) Reported by Qualys.
154 The following bugs are resolved with this release:
156 [The release manager will add the list generated by
157 scripts/list-fixed-bugs.py just before the release.]
164 * A per-thread cache has been added to malloc. Access to the cache requires
165 no locks and therefore significantly accelerates the fast path to allocate
166 and free small amounts of memory. Refilling an empty cache requires locking
167 the underlying arena. Performance measurements show significant gains in a
168 wide variety of user workloads. Workloads were captured using a special
169 instrumented malloc and analyzed with a malloc simulator. Contributed by
170 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
172 * Unicode 10.0.0 Support: Character encoding, character type info, and
173 transliteration tables are all updated to Unicode 10.0.0, using
174 generator scripts contributed by Mike FABIAN (Red Hat).
175 These updates cause user visible changes, especially the changes in
176 wcwidth for many emoji characters cause problems when emoji sequences
177 are rendered with pango, see for example:
178 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
180 * Collation of Hungarian has been overhauled and is now consistent with "The
181 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
184 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
186 - The GNU C Library will now detect when /etc/resolv.conf has been
187 modified and reload the changed configuration. The new resolver option
188 “no-reload” (RES_NORELOAD) disables this behavior.
190 - The GNU C Library now supports an arbitrary number of search domains
191 (configured using the “search” directive in /etc/resolv.conf);
192 previously, there was a hard limit of six domains. For backward
193 compatibility, applications that directly modify the ‘_res’ global
194 object are still limited to six search domains.
196 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
197 Library will now randomly pick a name server from the configuration as a
198 starting point. (Previously, the second name server was always used.)
200 * The tunables feature is now enabled by default. This allows users to tweak
201 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
203 * New function reallocarray, which resizes an allocated block (like realloc)
204 to the product of two sizes, with a guaranteed clean failure upon integer
205 overflow in the multiplication. Originally from OpenBSD, contributed by
206 Dennis Wölfing and Rüdiger Sonderfeld.
208 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
209 These are extended versions of preadv and pwritev, respectively, taking an
210 additional flags argument. The set of supported flags depends on the
211 running kernel; full support currently requires kernel 4.7 or later.
213 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
214 create a new session ID for the spawned process. This feature is
215 scheduled to be added to the next major revision of POSIX; for the time
216 being, it is available under _GNU_SOURCE.
218 * errno.h is now safe to use from C-preprocessed assembly language on all
219 supported operating systems. In this context, it will only define the
220 Exxxx constants, as preprocessor macros expanding to integer literals.
222 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
223 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
224 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
225 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
227 To compile programs that use this feature, the compiler must support
228 128-bit floating point with the type name _Float128 (as defined by TS
229 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
230 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
231 must be defined to make the new interfaces visible.
233 The new functions and macros correspond to those present for other
234 floating-point types (except for a few obsolescent interfaces not
235 supported for the new type), with F128 or f128 suffixes; for example,
236 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
237 printf or scanf formats for the new type; the strfromf128 and strtof128
238 interfaces should be used instead.
240 Deprecated and removed features, and other changes affecting compatibility:
242 * The synchronization that pthread_spin_unlock performs has been changed to
243 now be equivalent to a C11 atomic store with release memory order to the
244 spin lock's memory location. Previously, several (but not all)
245 architectures used stronger synchronization (e.g., containing what is
246 often called a full barrier). This change can improve performance, but
247 may affect odd fringe uses of spin locks that depend on the previous
248 behavior (e.g., using spin locks as atomic variables to try to implement
249 Dekker's mutual exclusion algorithm).
251 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
254 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
255 will only be built and installed when the GNU C Library is configured with
256 --enable-obsolete-rpc. This allows alternative RPC implementations, such
257 as TIRPC or rpcsvc-proto, to be used.
259 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
260 libnss_compat, are deprecated, and will not be built or installed by
263 The NIS(+) support library, libnsl, is also deprecated. By default, a
264 compatibility shared library will be built and installed, but not headers
265 or development libraries. Only a few NIS-related programs require this
266 library. (In particular, the GNU C Library has never required programs
267 that use 'gethostbyname' to be linked with libnsl.)
269 Replacement implementations based on TIRPC, which additionally support
270 IPv6, are available from <https://github.com/thkukuk/>. The configure
271 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
272 name service modules, to be built and installed.
274 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
275 support is enabled, the configured recursive resolver must support EDNS.
276 (Responding to EDNS-enabled queries with responses which are not
277 EDNS-enabled is fine, but FORMERR responses are not.)
279 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
280 servers have not supported this opcode for a long time.
282 * The _res_opcodes variable has been removed from libresolv. It had been
283 exported by accident.
285 * <string.h> no longer includes inline versions of any string functions,
286 as this kind of optimization is better done by the compiler. The macros
287 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
289 * The nonstandard header <xlocale.h> has been removed. Most programs should
290 use <locale.h> instead. If you have a specific need for the definition of
291 locale_t with no other declarations, please contact
292 libc-alpha@sourceware.org and explain.
294 * The obsolete header <sys/ultrasound.h> has been removed.
296 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
298 * The obsolete function cfree has been removed. Applications should use
301 * The stack_t type no longer has the name struct sigaltstack. This changes
302 the C++ name mangling for interfaces involving this type.
304 * The ucontext_t type no longer has the name struct ucontext. This changes
305 the C++ name mangling for interfaces involving this type.
307 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
308 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
309 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
310 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
311 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
312 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
313 fpregset_t type no longer has the name struct fpu. This changes the C++
314 name mangling for interfaces involving those types.
316 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
317 synced with the kernel:
319 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
320 are not supported on this architecture and have been removed.
322 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
323 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
324 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
326 Programs that assume the GET/SETREGS ptrace requests are universally
327 available will now fail to build, instead of malfunctioning at runtime.
329 Changes to build and runtime requirements:
331 * Linux kernel 3.2 or later is required at runtime, on all architectures
332 supported by that kernel. (This is a change from version 2.25 only for
335 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
337 * On most architectures, GCC 4.9 or later is required to build the GNU C
338 Library. On powerpc64le, GCC 6.2 or later is required.
340 Older GCC versions and non-GNU compilers are still supported when
341 compiling programs that use the GNU C Library. (We do not know exactly
342 how old, and some GNU extensions to C may be _de facto_ required. If you
343 are interested in helping us make this statement less vague, please
344 contact libc-alpha@sourceware.org.)
346 Security related changes:
348 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
349 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
351 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
352 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
354 * Avoid printing a backtrace from the __stack_chk_fail function since it is
355 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
358 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
359 fixed (CVE-2017-12133).
361 The following bugs are resolved with this release:
363 [984] network: Respond to changed resolv.conf in gethostbyname
364 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
365 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
367 [12189] libc: __stack_chk_fail should not attempt a backtrace
369 [14096] time: Race condition on timezone/tst-timezone.out
370 [14172] localedata: az_IR: new locale
371 [14995] build: glibc fails to build if gold is the default linker, even if
373 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
374 [16637] network: inet_pton function is accepting IPv6 with bad format
375 [16640] string: string/strtok.c: undefined behaviour inconsistent between
376 x86 and other generic code
377 [16875] localedata: ko_KR: fix lang_name
378 [17225] localedata: ar_SY: localized month names for May and June are
380 [17297] localedata: da_DK: wrong date_fmt string
381 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
382 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
383 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
384 [19066] localedata: ar_SA abbreviated day and month names are in English
385 [19569] network: resolv: Support an arbitrary number of search domains
386 [19570] network: Implement random DNS server selection in the stub
388 [19838] locale: localedef fails on PA-RISC
389 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
390 order of 0D36 and 0D37
391 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
393 [20098] libc: FAIL: debug/backtrace-tst on hppa
394 [20257] network: sunrpc: clntudp_call does not enforce timeout when
396 [20275] localedata: locale day/abday/mon/abmon should not have trailing
398 [20313] localedata: Update locale data to Unicode 9.0
399 [20424] manual: Document how to provide a malloc replacement
400 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
402 [20686] locale: Add el_GR@euro to SUPPORTED.
403 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
404 failures consistently
405 [21015] dynamic-link: Document and fix --enable-bind-now
406 [21016] nptl: pthread_cond support is broken on hppa
407 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
408 frame-pointer on i386
409 [21049] libc: segfault in longjmp_chk() due to clobbered processor
411 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
412 [21088] libc: Build fails with --enable-static-nss
413 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
414 generic c code is used
415 [21109] libc: Tunables broken on big-endian
416 [21112] math: powf has large ulp errors with base close to 1 and exponent
418 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
420 [21120] malloc: glibc malloc is incompatible with GCC 7
421 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
423 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
424 (-0.0) when linking with -lieee
425 [21171] math: log10, log2 and lgamma return incorrect results
426 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
427 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
428 [21207] localedata: ce_RU: update weekdays from CLDR
429 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
430 [21217] localedata: Update months from CLDR-31
431 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
432 [21243] libc: support_delete_temp_file should issue warning for failed
434 [21244] libc: support resolv_test_start() socket fd close should be
436 [21253] libc: localedef randomly segfaults when using -fstack-check due to
437 new posix_spawn implementation
438 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
439 leads to lower CPU frequency
440 [21259] libc: [alpha] termios.h missing IXANY for POSIX
441 [21261] libc: [sparc64] bits/setjmp.h namespace
442 [21267] network: [mips] bits/socket.h IOC* namespace
443 [21268] libc: [alpha] termios.h NL2, NL3 namespace
444 [21270] libc: mmap64 silently truncates large offset values
445 [21275] libc: posix_spawn always crashes on ia64 now
446 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
447 [21280] math: [powerpc] logbl for POWER7 return incorrect results
448 [21289] libc: Incorrect declaration for 32-bit platforms with
449 _FILE_OFFSET_BITS=64 causes build error
450 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
451 not support gethostbyname4_r
452 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
454 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
456 [21340] libc: Support POSIX_SPAWN_SETSID
457 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
458 [21359] network: ns_name_pack needs additional byte in destination buffer
459 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
460 against fragmentation attacks (CVE-2017-12132)
461 [21369] network: resolv: Remove EDNS fallback
462 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
464 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
465 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
466 [21393] stdio: Missing dup3 error check in freopen, freopen64
467 [21396] libc: Use AVX2 memcpy/memset on Skylake server
468 [21399] localedata: Bad description for U00EC in
469 localedata/charmaps/CP1254
470 [21411] malloc: realloc documentation error
471 [21426] network: sys/socket.h uio.h namespace
472 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
473 [21445] libc: signal.h bsd_signal namespace
474 [21455] network: Network headers stdint.h namespace
475 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
477 [21475] network: resolv: Overlong search path is truncated mid-label
478 [21511] libc: sigstack namespace
479 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
480 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
481 bits/syscall.h: No such file or directory
482 [21517] libc: struct sigaltstack namespace
483 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
484 [21533] localedata: Update locale data to Unicode 10.0
486 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
487 junk at end of line, first unrecognized character is `@'
488 [21538] libc: SIG_HOLD missing for XPG4
489 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
490 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
491 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
492 [21543] libc: sigevent namespace
493 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
494 size for general purpose registers in mcontext_t structure
495 [21550] libc: sigwait namespace
496 [21552] libc: XPG4 bsd_signal namespace
497 [21554] libc: sigpause namespace
498 [21560] libc: sys/wait.h signal.h namespace
499 [21561] libc: waitid namespace
500 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
501 '/usr/include/stdlib.h': Permission denied
502 [21575] libc: sys/wait.h missing struct rusage definition
503 [21584] libc: sigaltstack etc namespace
504 [21597] libc: siginterrupt namespace
505 [21607] math: hppa: FAIL: math/test-tgmath
506 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
508 [21622] libc: [tile] missing SA_* for POSIX.1:2008
509 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
510 stack and heap (CVE-2017-1000366)
511 [21625] libc: wait3 namespace
512 [21654] nss: Incorrect pointer alignment in NSS group merge result
514 [21657] network: Parse interface zone id for node-local multicast
515 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
517 [21666] libc: .symver is used on common symbol
518 [21668] network: resolv: res_init cross-thread broadcast introduces race
520 [21687] math: tgmath.h totalorder, totalordermag return type
521 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
523 [21696] libc: Incorrect assumption of of __cpu_mask in
524 posix/sched_cpucount.c
525 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
526 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
528 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
529 leading to relocation crash
530 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
531 `resolv_conf_matches (resp, conf)' failed.
532 [21710] localedata: Added Samoan language locale for Samoa
533 [21711] localedata: Pashto yesstr/nostr locale are missing
534 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
535 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
536 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
537 [21723] localedata: yesstr/nostr missing for Chinese language locale
538 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
540 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
542 [21728] localedata: New Locale for Tongan language
543 [21729] localedata: incorrect LC_NAME fields for hi_IN
544 [21733] localedata: yesstr and nostr are missing for zh_HK
545 [21734] localedata: Missing yesstr and nostr are for kw_GB
546 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
547 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
548 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
549 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
551 [21744] libc: Tests failing on --enable-tunables --enable-stack-
553 [21749] localedata: Wrong abbreviated day name (“abday”) for
555 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
556 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
557 [21759] localedata: missing yesstr and nostr for Tigrinya
558 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
559 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
560 [21767] localedata: Missing Bislama locales
561 [21768] localedata: Missing yesstr and nostr for aa_ET
562 [21770] localedata: Missing Field in li_NL
563 [21778] nptl: Robust mutex may deadlock
564 [21779] libc: MicroBlaze segfaults when loading libpthread
565 [21783] localedata: Fix int_select international_call_prefixes
566 [21784] localedata: Inconsistency in country_isbn
567 [21788] localedata: Missing Country Postal Abbreviations
568 [21794] localedata: Added-country_isbn-for-Italy
569 [21795] localedata: Add/Fix country_isbn for France
570 [21796] localedata: Added country_isbn for Republic of Korea
571 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
572 [21799] localedata: Added int_select international_call_prefixes
573 [21801] localedata: Added int_select international_call_prefixes
574 [21804] nptl: Double semicolon in thread-shared-types.h
575 [21807] localedata: LC_ADDRESS fix for pap_CW
576 [21808] localedata: Fix LC_ADDRESS for pap_AW
577 [21821] localedata: Added country_name in mai_IN
578 [21822] localedata: Fix LC_TIME for mai_IN
579 [21823] localedata: missing yesstr, nostr for sa_IN
580 [21825] localedata: Fix name_mrs for mag_IN
581 [21828] localedata: 2.26 changelog should mention user visible changes
583 [21835] localedata: Added Maithili language locale for Nepal
584 [21838] localedata: Removed redundant data for the_NP
585 [21839] localedata: Fix LC_MONETARY for ta_LK
586 [21844] localedata: Fix Latin characters and Months Sequence.
587 [21848] localedata: Fix mai_NP Title Name
592 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
593 24731-2:2010, is supported to enable declarations of functions from that
594 TR. Note that not all functions from that TR are supported by the GNU C
597 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
598 18661-1:2014, is supported to enable declarations of functions and macros
599 from that TS. Note that not all features from that TS are supported by
602 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
603 18661-4:2015, is supported to enable declarations of functions and macros
604 from that TS. Note that most features from that TS are not supported by
607 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
608 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
609 Since the GNU C Library defaults to a much newer revision of POSIX, this
610 will only affect programs that specifically request an old conformance
611 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
612 see a change in the visible declarations, but a program compiled with
613 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
616 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
617 defined by all multithreaded code, but glibc has not required this for
620 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
621 means that in a future release, the macros “major”, “minor”, and “makedev”
622 will only be available from <sys/sysmacros.h>.
624 These macros are not part of POSIX nor XSI, and their names frequently
625 collide with user code; see for instance glibc bug 19239 and Red Hat bug
626 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
627 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
630 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
631 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
632 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
634 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
635 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
636 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
637 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
638 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
639 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
640 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
641 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
642 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
643 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
644 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
647 * New <math.h> features are added from TS 18661-1:2014:
649 - Signaling NaN macros: SNANF, SNAN, SNANL.
651 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
652 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
653 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
655 - llogb functions: the llogb, llogbf and llogbl functions, and the
656 FP_LLOGB0 and FP_LLOGBNAN macros.
658 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
661 - Comparison macros: iseqsig.
663 - Classification macros: iscanonical, issubnormal, iszero.
665 - Total order functions: totalorder, totalorderf, totalorderl,
666 totalordermag, totalordermagf, totalordermagl.
668 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
670 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
671 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
673 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
674 are added to libc. They convert a floating-point number into string.
676 * Most of glibc can now be built with the stack smashing protector enabled.
677 It is recommended to build glibc with --enable-stack-protector=strong.
678 Implemented by Nick Alcock (Oracle).
680 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
681 intended to be used instead of memset() to erase sensitive data after use;
682 the compiler will not optimize out calls to explicit_bzero even if they
683 are "unnecessary" (in the sense that no _correct_ program can observe the
684 effects of the memory clear).
686 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
687 to float instead of double. This does not affect the ABI of any libraries
688 that are part of the GNU C Library, but may affect the ABI of other
689 libraries that use this type in their interfaces.
691 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
692 float_t and double_t types are now defined to long double instead of float
693 and double. These options are not the default, and this does not affect
694 the ABI of any libraries that are part of the GNU C Library, but it may
695 affect the ABI of other libraries that use this type in their interfaces,
696 if they are compiled or used with those options.
698 * The getentropy and getrandom functions, and the <sys/random.h> header file
701 * The buffer size for byte-oriented stdio streams is now limited to 8192
702 bytes by default. Previously, on Linux, the default buffer size on most
703 file systems was 4096 bytes (and thus remains unchanged), except on
704 network file systems, where the buffer size was unpredictable and could be
705 as large as several megabytes.
707 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
708 for the Linux quota interface which predates kernel version 2.4.22 has
711 * The malloc_get_state and malloc_set_state functions have been removed.
712 Already-existing binaries that dynamically link to these functions will
713 get a hidden implementation in which malloc_get_state is a stub. As far
714 as we know, these functions are used only by GNU Emacs and this change
715 will not adversely affect already-built Emacs executables. Any undumped
716 Emacs executables, which normally exist only during an Emacs build, should
717 be rebuilt by re-running “./configure; make” in the Emacs build tree.
719 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
720 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
721 “no-ip6-dotint” had already been the default, and support for the
722 “ip6-dotint” option was removed from the Internet in 2006.
724 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
725 flag from <resolv.h> have been removed. The option relied on a
726 backwards-incompatible DNS extension which was never deployed on the
729 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
730 RES_BLAST defined in the <resolv.h> header file have been deprecated.
731 They were already unimplemented.
733 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
734 _res.flags are deprecated. The flag was standardized in RFC 2133, but
735 removed again from the IETF name lookup interface specification in RFC
736 2553. Applications should use getaddrinfo instead.
738 * DNSSEC-related declarations and definitions have been removed from the
739 <arpa/nameser.h> header file, and libresolv will no longer attempt to
740 decode the data part of DNSSEC record types. Previous versions of glibc
741 only implemented minimal support for the previous version of DNSSEC, which
742 is incompatible with the currently deployed version.
744 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
745 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
746 <arpa/nameser.h> header file because the distinction between RR types and
747 meta-RR types is not officially standardized, subject to revision, and
748 thus not suitable for encoding in a macro.
750 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
751 and rhook members of the res_state type in <resolv.h> have been removed.
752 The glibc stub resolver did not support these hooks, but the header file
753 did not reflect that.
755 * For multi-arch support it is recommended to use a GCC which has
756 been built with support for GNU indirect functions. This ensures
757 that correct debugging information is generated for functions
758 selected by IFUNC resolvers. This support can either be enabled by
759 configuring GCC with '--enable-gnu-indirect-function', or by
760 enabling it by default by setting 'default_gnu_indirect_function'
761 variable for a particular architecture in the GCC source file
764 * GDB pretty printers have been added for mutex and condition variable
765 structures in POSIX Threads. When installed and loaded in gdb these pretty
766 printers show various pthread variables in human-readable form when read
767 using the 'print' or 'display' commands in gdb.
769 * Tunables feature added to allow tweaking of the runtime for an application
770 program. This feature can be enabled with the '--enable-tunables' configure
771 flag. The GNU C Library manual has details on usage and README.tunables has
772 instructions on adding new tunables to the library.
774 * A new version of condition variables functions have been implemented in
775 the NPTL implementation of POSIX Threads to provide stronger ordering
778 * A new version of pthread_rwlock functions have been implemented to use a more
779 scalable algorithm primarily through not using a critical section anymore to
782 Security related changes:
784 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
785 have been created with makecontext could fail to terminate due to a
786 missing .cantunwind annotation. This has been observed to lead to a hang
787 (denial of service) in some Go applications compiled with gccgo. Reported
788 by Andreas Schwab. (CVE-2016-6323)
790 * The DNS stub resolver functions would crash due to a NULL pointer
791 dereference when processing a query with a valid DNS question type which
792 was used internally in the implementation. The stub resolver now uses a
793 question type which is outside the range of valid question type values.
796 The following bugs are resolved with this release:
798 [4099] stdio: Overly agressive caching by stream i/o functions.
799 [7065] build: Support building glibc with -fstack-protector or -fstack-
801 [9842] localedata: en_CA: incorrect date format
802 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
803 before it started waiting
804 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
805 [15765] nptl: sem_open is wrongly a cancellation point
806 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
807 [16458] libc: endian.h and netinet/in.h byte order macros return values of
809 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
810 library linked with pthread
811 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
813 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
814 [17252] libc: getrandom and getentropy syscall
815 [17730] malloc: thread-local storage is sometimes improperly free()'d
816 after being __libc_memalign()'d
817 [18241] stdio: failed fseek on memstream does not set errno and can fail
819 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
821 [18463] nptl: pthread_cond_broadcast issue when surrounded by
822 PTHREAD_PRIO_INHERIT mutex on ARM
823 [18784] network: res_query and related function crash for special record
824 type queries (CVE-2015-5180)
825 [19380] math: strtod does not raise "inexact"
826 [19387] string: Integer overflow in memchr
827 [19390] string: Integer overflow in strncat
828 [19398] build: linknamespace tests fail with massively parallel build
829 [19402] nptl: Deadlock with robust shared mutex and asynchronous
831 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
832 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
833 [19514] libc: [PATCH] Fix spelling errors in spelling
834 "implement"/"implementation" in several places
835 [19582] network: Deprecate RES_USE_INET6
836 [19673] manual: clog10 docs appear to be erroneous
837 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
838 causes a segmentation fault
839 [19826] libc: invalid pointer returned from __tls_get_addr with static
841 [20016] network: resolv: Remove hooks support from the API
842 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
844 [20033] math: [x86_64] vectorized math function don't call the __finite
846 [20116] nptl: use after free in pthread_create
847 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
848 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
850 [20311] nptl: please install proc_service.h
851 [20366] libc: Compilation errors in installed headers in strict-compliance
853 [20370] malloc: malloc: Arena free list management is still racy
854 (incorrect fix in bug 19243)
855 [20386] libc: assert (X = 0) does not result in GCC warning
856 [20432] malloc: malloc: Minimize interface required for interposition
857 [20435] libc: Missing unwind info in __startcontext causes infinite loop
858 in _Unwind_Backtrace (CVE-2016-6323)
859 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
860 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
862 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
863 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
864 [20477] network: resolv: incorrect double-checked locking related to
866 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
868 [20495] math: x86_64 performance degradation due to AVX/SSE transition
870 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
871 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
872 cause transition penalty
873 [20517] math: sparcv9 missing fdiml compat symbol
874 [20524] manual: strverscmp is inconsistent
875 [20525] libc: <sys/quota.h> should be based on kernel headers
876 [20539] math: GCC 7 -static -lm fails to link at -O3
877 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
879 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
881 [20591] network: Remove obsolete DNSSEC support
882 [20592] network: DNS resource record type classification macros in
883 <arpa/nameser.h> are incorrect
884 [20593] network: Update DNS RR type definitions
885 [20611] network: getaddrinfo accepts invalid numeric scope IDs
886 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
888 [20629] network: libresolv: Remove support for bitstring labels
890 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
891 assertion: "maxidx >= 2"
892 [20660] math: [arm] Use VSQRT
893 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
894 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
895 [20689] libc: Test for FMA should also check for AVX.
896 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
897 [20715] math: iszero macro breaks existing code
898 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
899 [20728] libc: powerpc: Missing TOC stub in clone
900 [20729] build: glibc-2.24 fails to build for i486 with -Os
901 [20750] build: Build fails with default PIE enabled gcc-6.x
902 [20768] math: [alpha] sqrt fegetenv namespace
903 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
905 [20787] math: float_t is defined as float incorrectly on x86_64 even if
906 __FLT_EVAL_METHOD__ is 2
907 [20790] network: rpcgen buffer overrun in get_prog_declaration
908 [20822] nptl: race condition in __lll_unlock_elision on powerpc
909 [20829] libc: crypt snprintf namespace
910 [20847] libc: tst-vfork3 failure
911 [20855] math: Default bits/mathdef.h has inappropriate float_t
912 [20859] math: [sh4] FP_ILOGB0 invalid
913 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
914 U+20AC), not same as GBK
915 [20915] dynamic-link: global-dynamic TLS broken on aarch64
916 [20916] math: pow handling of sNaN arguments
917 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
918 [20919] math: Bad pow (qNaN, 0) result with -lieee
919 [20940] math: hypot sNaN handling
920 [20947] math: fmax, fmin sNaN handling
921 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
922 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
924 [20971] string: powerpc64/power7 memchr overflows internal pointer check
925 [20973] nptl: robust mutexes: Lost wake-ups
926 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
927 [20978] nis: strlen on null pointer in nss_nisplus
928 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
930 [21014] string: i686 memchr overflows internal pointer check
931 [21019] libc: [mips] n32 lseek incorrect on overflow
932 [21022] libc: [microblaze] __backtrace get_frame_size namespace
933 [21026] libc: [MIPS] readahead syscall is broken on n64
934 [21028] math: Fallback fesetexceptflag should always succeed
935 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
936 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
938 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
939 [21061] librt: [microblaze] librt lost clock_* exports
940 [21073] libc: tunables: insecure environment variables passed to
941 subprocesses with AT_SECURE
942 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
946 * The minimum Linux kernel version that this version of the GNU C Library
947 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
948 kernel version 2.6.32 or later suffices (on architectures that already
949 required kernel versions more recent than 3.2, those requirements remain
950 unchanged). Linux 3.2 or later kernel headers are required on all
953 * The pap_AN locale has been deleted. This has been deprecated for a long
954 time. It has been replaced by pap_AW & pap_CW, both of which have long
955 been included in previous releases.
957 * The readdir_r and readdir64_r functions have been deprecated. It is
958 recommended to use readdir and readdir64 instead.
960 * The type “union wait” has been removed. It was deprecated in the early
961 1990s and never part of POSIX. Application code should use the int type
962 instead of “union wait”.
964 * A new NSS action is added to facilitate large distributed system
965 administration. The action, MERGE, allows remote user stores like LDAP
966 to be merged into local user stores like /etc/groups in order to provide
967 easy to use, updated, and managed sets of merged credentials. The new
968 action can be used by configuring it in /etc/nsswitch.conf:
969 group: files [SUCCESS=merge] nis
970 Implemented by Stephen Gallagher (Red Hat).
972 * The deprecated __malloc_initialize_hook variable has been removed from the
975 * The long unused localedef --old-style option has been removed. It hasn't
976 done anything in over 16 years. Scripts using this option can safely
979 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
980 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
981 return the next representable value in the direction of positive infinity
982 and the nextdown functions return the next representable value in the
983 direction of negative infinity. These are currently enabled as GNU
986 Security related changes:
988 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
989 could result in a stack overflow when getnetbyname was called with an
990 overly long name. (CVE-2016-3075)
992 * Previously, getaddrinfo copied large amounts of address data to the stack,
993 even after the fix for CVE-2013-4458 has been applied, potentially
994 resulting in a stack overflow. getaddrinfo now uses a heap allocation
995 instead. Reported by Michael Petlan. (CVE-2016-3706)
997 * The glob function suffered from a stack-based buffer overflow when it was
998 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
999 Reported by Alexander Cherepanov. (CVE-2016-1234)
1001 * The Sun RPC UDP client could exhaust all available stack space when
1002 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1003 alloca plugin for GCC. (CVE-2016-4429)
1005 * The IPv6 name server management code in libresolv could result in a memory
1006 leak for each thread which is created, performs a failing naming lookup,
1007 and exits. Over time, this could result in a denial of service due to
1008 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1010 The following bugs are resolved with this release:
1012 [1170] localedata: ne_NP: update Nepali locale definition file
1013 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1015 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1016 [6796] math: fdim() does not set errno on overflow
1017 [10354] libc: posix_spawn should use vfork() in more cases than presently
1018 [11213] localedata: localedata: add copyright disclaimer to locale files
1019 [12143] localedata: chr_US: new Cherokee locale
1020 [12450] localedata: sgs_LT: new locale
1021 [12676] localedata: ln_CD: new locale
1022 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1024 [13304] math: fma, fmaf, fmal produce wrong results
1025 [14259] build: --localedir arg to configure is ignored
1026 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1027 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1028 [14934] localedata: es_CL: wrong first weekday chilean locale
1029 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1031 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1033 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1034 [15368] nptl: raise() is not async-signal-safe
1035 [15479] math: ceil, floor, round and trunc raise inexact exception
1036 [15578] localedata: kk_KZ: various updates
1037 [16003] localedata: pap_AN: punt old locale
1038 [16137] localedata: iw_IL: punt old locale
1039 [16190] localedata: eo: new esperanto locale
1040 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1041 [16742] malloc: race condition: pthread_atfork() called before first
1042 malloc() results in unexpected locking behaviour/deadlocks
1043 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1045 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1046 [17565] localedata: pt_PT: wrong (work-)week start
1047 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1048 [17950] build: Build fails with -msse
1049 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1050 [18433] libc: posix_spawn does not return correctly upon failure to
1052 [18453] localedata: charmaps/IBM875: incorrect codes
1053 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1055 [18896] localedata: he_IL: improvements for currency
1056 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1057 Romanian locale data
1058 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1060 [19056] libc: Deprecate readdir_r
1061 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1063 [19198] localedata: nl_NL: small improvements for Dutch locales
1064 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1065 nameservers (CVE-2016-5417)
1066 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1068 [19400] locale: Language missing in "iso-639.def", trivial fix in
1070 [19431] malloc: Deadlock between fflush, getdelim, and fork
1071 [19505] libc: Incorrect file descriptor validity checks in
1072 posix_spawn_file_actions_add{open,close,dup2}
1073 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1074 when using RTLD_NEXT
1075 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1076 `intl/{gettextP,loadinfo}.h'
1077 [19534] libc: execle, execlp may use malloc
1078 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1079 [19573] network: res_nclose and __res_maybe_init disagree about name
1080 server initialization, breaking Hesiod
1081 [19575] localedata: Status of GB18030 tables
1082 [19581] localedata: sr_* date_fmt string contains additional newline
1083 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1085 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1086 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1087 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1088 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1090 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1092 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1094 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1095 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1096 [19633] locale: strfmon_l applies global locale to number formatting
1097 [19642] network: Memory leak in getnameinfo
1098 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1099 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1101 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1102 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1104 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1105 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1107 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1109 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1111 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1112 pointers and lengths in error-case.
1113 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1114 report errors on UTF-16 surrogates values.
1115 [19755] nscd: nscd assertion failure in gc
1116 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1117 [19759] libc: mempcpy shouldn't be inlined
1118 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1119 [19765] libc: s390 needs an optimized mempcpy
1120 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1121 NAME_MAX limit assumption (CVE-2016-1234)
1122 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1123 [19787] network: Missing and incorrect truncation checks in getnameinfo
1124 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1126 [19791] network: Assertion failure in res_query.c with un-connectable name
1128 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1129 [19822] math: libm.so install clobbers old version
1130 [19825] network: resolv: send_vc can return uninitialized data in second
1131 response to getaddrinfo
1132 [19830] network: nss_dns: should check RDATA length against buffer length
1133 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1134 confronted with A/AAAA records of invalid size
1135 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1136 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1137 [19853] stdio: Printing IBM long double in decimal with high precision is
1139 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1140 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1141 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1143 [19865] network: Assertion failure or memory leak in
1144 _nss_dns_getcanonname_r
1145 [19868] network: nss_dns: netent code does not skip over non-PTR records
1146 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1148 [19881] string: Improve x86-64 memset
1149 [19907] string: Incorrect memcpy tests
1150 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1152 [19925] libc: termios.h XCASE namespace
1153 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1154 [19929] libc: limits.h NL_NMAX namespace
1155 [19931] stdio: Memory leak in vfprintf
1156 [19957] libc: clone(CLONE_VM) access invalid parent memory
1157 [19963] localedata: en_IL: New locale
1158 [19989] stdio: stdio.h cuserid namespace
1159 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1161 [19996] locale: langinfo.h nl_langinfo_l namespace
1162 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1163 [20010] network: getaddrinfo: Stack overflow in hostent translation
1165 [20012] stdio: libio: fmemopen append mode failure
1166 [20014] stdio: stdio.h namespace for pre-threads POSIX
1167 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1168 [20023] libc: fcntl.h timespec namespace
1169 [20024] math: [x86_64] vectorized sincos trashes the stack
1170 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1171 [20041] time: sys/time.h timespec namespace
1172 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1173 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1174 [20051] libc: ttyslot in wrong header under wrong conditions
1175 [20054] libc: gethostname not declared for XPG4
1176 [20055] libc: termios.h missing tcgetsid for XPG4
1177 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1179 [20073] libc: sys/stat.h fchmod namespace
1180 [20074] libc: stdlib.h rand_r namespace
1181 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1182 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1184 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1185 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1187 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1188 [20119] libc: Wrong mask for processors level type from CPUID
1189 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1190 AS not supporting AVX512
1191 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1193 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1194 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1196 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1197 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1198 [20178] libc: posix_spawn{p} should not call exit
1199 [20191] stdio: libio: vtables hardening
1200 [20195] string: FMA4 detection requires CPUID execution with register
1202 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1203 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1205 [20212] math: acos (sNaN) returns sNaN
1206 [20213] math: asin (sNaN) returns sNaN
1207 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1208 [20218] math: [i386] asinhl (sNaN) returns sNaN
1209 [20219] math: [i386] atanhl (sNaN) returns sNaN
1210 [20222] stdio: fopencookie: Mangle function pointers
1211 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1212 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1213 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1215 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1216 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1217 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1218 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1219 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1220 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1221 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1222 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1223 [20240] math: modf (sNaN) returns sNaN
1224 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1225 [20250] math: frexp (sNaN) returns sNaN
1226 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1227 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1229 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1230 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1231 subscript is below array bounds [-Werror=array-bounds]
1232 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1233 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1235 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1236 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1238 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1239 "invalid" exceptions
1240 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1241 [20316] localedata: id_ID: Februari instead of Pebruari
1242 [20327] string: POWER8 strcasecmp returns incorrect result
1243 [20347] math: Failure: Test: j0_downward (0xap+0)
1244 [20348] libc: FAIL: misc/tst-preadvwritev64
1245 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1246 [20350] libc: There is no test for p{read,write}64
1247 [20357] math: Incorrect cos result for 1.5174239687223976
1248 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1252 * Unicode 8.0.0 Support: Character encoding, character type info, and
1253 transliteration tables are all updated to Unicode 8.0.0, using new
1254 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1255 These updates cause user visible changes, such as the fixes for bugs
1256 89, 16061, and 18568.
1258 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1259 kernel-internal CPU set size. This means that requests that change the
1260 CPU affinity which failed before (for example, an all-ones CPU mask) will
1261 now succeed. Applications that need to determine the effective CPU
1262 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1263 setting it because the kernel can adjust it (and the previous size check
1264 would not detect this in the majority of cases).
1266 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1267 the following new symbols are used: fts64_children, fts64_close,
1268 fts64_open, fts64_read and fts64_set.
1270 * getaddrinfo now detects certain invalid responses on an internal netlink
1271 socket. If such responses are received, an affected process will
1272 terminate with an error message of "Unexpected error <number> on netlink
1273 descriptor <number>" or "Unexpected netlink response of size <number> on
1274 descriptor <number>". The most likely cause for these errors is a
1275 multi-threaded application which erroneously closes and reuses the netlink
1276 file descriptor while it is used by getaddrinfo.
1278 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1279 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1280 unnecessary serialization of memory allocation requests across threads.
1281 The defect is now corrected. Users should see a substantial increase in
1282 the concurent throughput of allocation requests for applications which
1283 trigger this bug. Affected applications typically create and
1284 destroy threads frequently. (Bug 19048 was reported and analyzed by
1287 * There is now a --disable-timezone-tools configure option for disabling the
1288 building and installing of the timezone related utilities (zic, zdump, and
1289 tzselect). This is useful for people who build the timezone data and code
1290 independent of the GNU C Library.
1292 * The obsolete header <regexp.h> has been removed. Programs that require
1293 this header must be updated to use <regex.h> instead.
1295 * The obsolete functions bdflush, create_module, get_kernel_syms,
1296 query_module and uselib are no longer available to newly linked binaries;
1297 the header <sys/kdaemon.h> has been removed. These functions and header
1298 were specific to systems using the Linux kernel and could not usefully be
1299 used with the GNU C Library on systems with version 2.6 or later of the
1302 * Optimized string, wcsmbs and memory functions for IBM z13.
1303 Implemented by Stefan Liebler.
1305 * Newly linked programs that define a variable called signgam will no longer
1306 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1307 require signgam to be set by those functions must ensure that they use the
1308 variable provided by the GNU C Library and declared in <math.h>, without
1309 defining their own copy.
1311 * The minimum GCC version that can be used to build this version of the GNU
1312 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1313 still be used to compile programs using the GNU C Library.
1315 Security related changes:
1317 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1318 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1320 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1321 the pointer guard feature. It is always enabled. Previously,
1322 LD_POINTER_GUARD could be used to disable security hardening in binaries
1323 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1326 * An integer overflow in hcreate and hcreate_r could lead to an
1327 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1329 * The catopen function no longer has unbounded stack usage. Reported by
1330 Max. (CVE-2015-8779)
1332 * The nan, nanf and nanl functions no longer have unbounded stack usage
1333 depending on the length of the string passed as an argument to the
1334 functions. Reported by Joseph Myers. (CVE-2014-9761)
1336 * A stack-based buffer overflow was found in libresolv when invoked from
1337 libnss_dns, allowing specially crafted DNS responses to seize control
1338 of execution flow in the DNS client. The buffer overflow occurs in
1339 the functions send_dg (send datagram) and send_vc (send TCP) for the
1340 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1341 family. The use of AF_UNSPEC triggers the low-level resolver code to
1342 send out two parallel queries for A and AAAA. A mismanagement of the
1343 buffers used for those queries could result in the response of a query
1344 writing beyond the alloca allocated buffer created by
1345 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1346 the overflow. Thanks to the Google Security Team and Red Hat for
1347 reporting the security impact of this issue, and Robert Holiday of
1348 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1350 The following bugs are resolved with this release:
1352 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1353 [887] math: Math library function "logb" and "nextafter" inconsistent
1354 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1355 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1356 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1357 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1359 [4404] localedata: German translation of "Alarm clock" is misleading
1360 [6799] math: nextafter() and nexttoward() doen't set errno on
1361 overflow/underflow errors
1362 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1364 [10432] nis: _nss_nis_setnetgrent assertion failure
1365 [11460] libc: fts has no LFS support
1366 [12926] network: getaddrinfo()/make_request() may spin forever
1367 [13065] nptl: Race condition in pthread barriers
1368 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1369 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1371 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1372 [14912] libc: Rename non-installed bits/*.h headers
1373 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1374 [15367] math: Let gcc use __builtin_isinf
1375 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1376 [15421] math: lgamma wrongly sets signgam for ISO C
1377 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1378 FE_INVALID with argument out of range
1379 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1381 [15786] dynamic-link: ifunc resolver functions can smash function
1383 [15918] math: Unnecessary check for equality in hypotf()
1384 [16061] localedata: Review / update transliteration data
1385 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1387 [16141] time: strptime %z offset restriction
1388 [16171] math: drem should be alias of remainder
1389 [16296] math: fegetround is pure?
1390 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1391 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1393 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1394 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1396 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1397 "Exception "Inexact" set" on ppc32
1398 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1399 [16517] math: Missing underflow exception from tanf/tan/tanl
1400 [16519] math: Missing underflow exception from sinhf
1401 [16520] math: Missing underflow exception from tanhf
1402 [16521] math: Missing underflow exception from exp2
1403 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1405 [16734] stdio: fopen calls mmap to allocate its buffer
1406 [16961] math: nan function incorrect handling of bad sequences
1407 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1408 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1409 [16985] locale: localedef: confusing error message when opening output
1411 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1412 [17197] locale: Redundant shift character in iconv conversion output at
1414 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1415 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1416 [17250] dynamic-link: static linking breaks nss loading
1417 (getaddrinfo/getpwnam/etc...)
1418 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1420 [17441] math: isnan() should use __builtin_isnan() in GCC
1421 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1422 (related to lock elision)
1423 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1424 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1425 [17887] time: strptime should be able to parse "+01:00" style timezones
1426 [17905] libc: catopen() Multiple unbounded stack allocations
1428 [18084] libc: backtrace (..., 0) dumps core on x86
1429 [18086] libc: nice() sets errno to 0 on success
1430 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1431 is too large (CVE-2015-8778)
1432 [18251] dynamic-link: SONAME missing when audit modules provides path
1433 [18265] libc: add attributes for wchar string and memory functions
1434 [18370] math: csqrt missing underflows
1435 [18421] libc: [hppa] read-only segment has dynamic relocations
1436 [18472] libc: Obsolete syscall wrappers should be compat symbols
1437 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1438 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1439 [18525] localedata: Remove locale timezone information
1440 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1441 [18568] localedata: Update locale data to Unicode 8.0
1442 [18589] locale: sort-test.sh fails at random
1443 [18595] math: ctan, ctanh missing underflows
1444 [18604] libc: assert macro-expands its argument
1445 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1446 contains a vector instruction exception.
1447 [18611] math: j1, jn missing errno setting on underflow
1448 [18618] localedata: sync Chechen locale definitions with other *_RU
1450 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1451 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1452 [18665] network: In send_dg, the recvfrom function is NOT always using the
1453 buffer size of a newly created buffer (CVE-2015-7547)
1454 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1456 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1458 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1459 [18699] math: tilegx cproj() for various complex infinities does not yield
1461 [18724] libc: Harden put*ent functions against data injection
1462 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1463 [18755] build: build errors with -DNDEBUG
1464 [18757] stdio: fmemopen fails to set errno on failure
1465 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1467 [18781] libc: openat64 lacks O_LARGEFILE
1468 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1469 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1470 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1471 [18790] math: [ldbl-128ibm] tanhl inaccurate
1472 [18795] libc: stpncpy fortification misses buffer lengths that are
1473 statically too large
1474 [18796] build: build fails for --disable-mathvec
1475 [18803] math: hypot missing underflows
1476 [18820] stdio: fmemopen may leak memory on failure
1477 [18823] math: csqrt spurious underflows
1478 [18824] math: fma spurious underflows
1479 [18825] math: pow missing underflows
1480 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1481 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1482 [18870] build: sem_open.c fails to compile with missing symbol
1484 [18872] stdio: Fix memory leak in printf_positional
1485 [18873] libc: posix_fallocate overflow check ineffective
1486 [18875] math: Excess precision leads incorrect libm
1487 [18877] libc: arm: mmap offset regression
1488 [18887] libc: memory corruption when using getmntent on blank lines
1489 [18918] localedata: hu_HU: change time to HH:MM:SS format
1490 [18921] libc: Regression: extraneous stat() and fstat() performed by
1492 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1493 binaries (CVE-2015-8777)
1494 [18951] math: tgamma missing underflows
1495 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1497 [18953] localedata: lt_LT: change currency symbol to the euro
1498 [18956] math: powf inaccuracy
1499 [18961] math: [i386] exp missing underflows
1500 [18966] math: [i386] exp10 missing underflows
1501 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1502 [18969] build: multiple string test failures due to missing locale
1504 [18970] libc: Reference of pthread_setcancelstate in libc.a
1505 [18977] math: float / long double Bessel functions not in XSI POSIX
1506 [18980] math: i386 libm functions return with excess range and precision
1507 [18981] math: i386 scalb*, ldexp return with excess range and precision
1508 [18982] stdio: va_list and vprintf
1509 [18985] time: Passing out of range data to strftime() causes a segfault
1511 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1512 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1513 [19012] locale: iconv_open leaks memory on error path
1514 [19016] math: clog, clog10 inaccuracy
1515 [19018] nptl: Mangle function pointers in tls_dtor_list
1516 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1517 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1518 [19048] malloc: malloc: arena free list can become cyclic, increasing
1520 [19049] math: [powerpc] erfc incorrect zero sign
1521 [19050] math: [powerpc] log* incorrect zero sign
1522 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1523 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1524 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1525 [19074] network: Data race in _res_hconf_reorder_addrs
1526 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1527 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1528 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1530 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1532 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1533 [19086] manual: posix_fallocate64 documented argument order is wrong.
1534 [19088] math: lround, llround missing exceptions close to overflow
1536 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1537 [19095] math: dbl-64 lrint incorrect for 64-bit long
1538 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1539 [19124] dynamic-link: ld.so failed to build with older assmebler
1540 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1541 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1542 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1543 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1544 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1545 pthread_setaffinity_np
1546 [19156] math: [ldbl-128] j0l spurious underflows
1547 [19164] nptl: tst-getcpu fails with many possible CPUs
1548 [19168] math: math/test-ildoubl and math/test-ldouble failure
1549 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1550 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1552 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1553 (FE_NOMASK_ENV) do not clear SSE exceptions
1554 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1555 _int_new_arena/reused_arena
1556 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1557 [19201] math: dbl-64 remainder incorrect sign of zero result
1558 [19205] math: bits/math-finite.h conditions do not match math.h and
1560 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1561 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1562 for C99-based standards
1563 [19212] libc: features.h not -Wundef clean
1564 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1566 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1567 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1568 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1570 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1571 "inexact" exceptions
1572 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1574 [19242] libc: strtol incorrect in Turkish locales
1575 [19243] malloc: reused_arena can pick an arena on the free list, leading
1576 to an assertion failure and reference count corruption
1577 [19253] time: tzset() ineffective when temporary TZ did not include DST
1579 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1580 [19270] math: [hppa] Shared libm missing __isnanl
1581 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1582 MAP_HUGETLB and MAP_STACK defines
1583 [19313] nptl: Wrong __cpu_mask for x32
1584 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1585 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1586 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1587 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1588 [19363] time: x32: times() return value wrongly truncates/sign extends
1590 [19367] dynamic-link: Improve branch prediction on Silvermont
1591 [19369] network: Default domain name not reset by res_ninit when "search"
1592 / "domain" entry is removed from resolv.conf
1593 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1595 [19385] localedata: bg_BG: time separator should be colon, not comma
1596 [19408] libc: linux personality syscall wrapper may erroneously return an
1597 error on 32-bit architectures
1598 [19415] libc: dladdr returns wrong names on hppa
1599 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1600 IBM905, IBM907, and IBM909
1601 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1602 [19443] build: build failures with -DDEBUG
1603 [19451] build: Make check fails on test-double-vlen2
1604 [19462] libc: Glibc failed to build with -Os
1605 [19465] math: Wrong code with -Os
1606 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1608 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1610 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1611 [19486] math: S390: Math tests fail with "Exception Inexact set".
1612 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1613 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1615 [19590] math: Fail to build shared objects that use libmvec.so functions.
1619 * The following bugs are resolved with this release:
1621 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1622 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1623 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1624 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1625 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1626 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1627 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1628 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1629 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1630 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1631 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1632 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1633 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1634 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1635 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1636 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1637 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1638 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1639 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1640 18657, 18676, 18694, 18696, 18887.
1642 * Cache information can be queried via sysconf() function on s390 e.g. with
1643 _SC_LEVEL1_ICACHE_SIZE as argument.
1645 * A buffer overflow in gethostbyname_r and related functions performing DNS
1646 requests has been fixed. If the NSS functions were called with a
1647 misaligned buffer, the buffer length change due to pointer alignment was
1648 not taken into account. This could result in application crashes or,
1649 potentially arbitrary code execution, using crafted, but syntactically
1650 valid DNS responses. (CVE-2015-1781)
1652 * The time zone file parser has been made more robust against crafted time
1653 zone files, avoiding heap buffer overflows related to the processing of
1654 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1655 large time zone data files. Overly long time zone specifiers in the TZ
1656 variable no longer result in stack overflows and crashes.
1658 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1659 for LD and GD on x86 and x86-64, has been implemented. You will need
1660 binutils-2.24 or later to enable this optimization.
1662 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1663 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1664 Hat). These updates cause user visible changes, such as the fix for bug
1667 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1668 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1669 condition in some applications.
1671 * Added vector math library named libmvec with the following vectorized x86_64
1672 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1674 The library can be disabled with --disable-mathvec. Use of the functions is
1675 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1676 Shared library libmvec.so is linked in as needed when using -lm (no need to
1677 specify -lmvec explicitly for not static builds).
1678 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1680 * A new fmemopen implementation has been added with the goal of POSIX
1681 compliance. The new implementation fixes the following long-standing
1682 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1683 old implementation is still present for use be by existing binaries.
1685 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1686 releases. It has been fixed to match 2.19 and older, but binaries built
1687 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1689 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1690 Contributed by Roland McGrath (Google).
1692 * The header <regexp.h> is deprecated, and will be removed in a future
1693 release. Use of this header will trigger a deprecation warning.
1694 Application developers should update their code to use <regex.h> instead.
1696 This header was formerly part of SUS, but was deprecated in 1994 and
1697 removed from the standard in 2001. Also, the glibc implementation
1698 leaks memory. See BZ#18681 for more details.
1702 * The following bugs are resolved with this release:
1704 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1705 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1706 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1707 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1708 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1709 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1710 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1711 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1712 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1713 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1716 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1717 too little memory for the to-be-scanned arguments and overflow the
1718 allocated buffer. The implementation now correctly computes the required
1719 buffer size when using malloc, and switches to malloc from alloca as
1722 * A new semaphore algorithm has been implemented in generic C code for all
1723 machines. Previous custom assembly implementations of semaphore were
1724 difficult to reason about or ensure that they were safe. The new version
1725 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1726 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1727 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1729 * Port to Altera Nios II has been contributed by Mentor Graphics.
1731 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1732 implementations for powerpc64/powerpc64le.
1733 Implemented by Adhemerval Zanella (IBM).
1735 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1736 and powerpc64le. This may improve lock scaling of existing programs on
1737 HTM capable systems. The lock elision code is only enabled with
1738 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1739 powerpc will issue a transaction abort on every syscall to avoid side
1740 effects being visible outside transactions.
1742 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1743 AArch64. Contributed by ARM Ltd.
1745 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1747 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1748 under certain input conditions resulting in the execution of a shell for
1749 command substitution when the application did not request it. The
1750 implementation now checks WRDE_NOCMD immediately before executing the
1751 shell and returns the error WRDE_CMDSUB as expected.
1753 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1754 processing format strings with a large number of format specifiers.
1756 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1757 infinite loop if the DNS response contained a PTR record of an unexpected
1760 * The minimum GCC version that can be used to build this version of the GNU
1761 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1762 still be used to compile programs using the GNU C Library.
1764 * The GNU C Library is now built with -Werror by default. This can be
1765 disabled by configuring with --disable-werror.
1767 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1769 * The obsolete sigvec function has been removed. This was the original
1770 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1771 programs have been using instead for about 25 years. Of course, ABI
1772 compatibility for old binaries using sigvec remains intact.
1774 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1775 with newer versions of bison.
1777 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1778 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1779 registers overlay two consecutive single-precision registers. MIPS32R2
1780 introduced a new FPU mode (FR=1) where double-precision registers extend the
1781 corresponding single-precision registers which is incompatible with the
1782 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1783 require the use of FR=1 making a transition necessary. New o32 ABI
1784 extensions enable users to migrate over time from the original o32 ABI
1785 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1786 tracks the ABI of any loaded object and verifies that new objects are
1787 compatible. Mode transitions will also be requested as required and
1788 unsupportable objects will be rejected. The ABI checks include both soft and
1789 hard float ABIs for o32, n32 and n64.
1791 GCC 5 with GNU binutils 2.25 onwards:
1792 It is strongly recommended that all o32 system libraries are built using the
1793 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1794 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1795 compiler using --with-fp-32=xx to set this by default.
1799 * The following bugs are resolved with this release:
1801 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1802 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1803 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1804 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1805 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1806 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1807 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1808 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1809 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1810 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1811 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1812 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1813 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1814 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1815 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1817 * Reverted change of ABI data structures for s390 and s390x:
1818 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1819 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1820 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1821 are preserved pointing straight to the same implementation as the old ones.
1822 Given that, new callers will simply provide a too-big buffer to these
1823 functions. Any applications/libraries out there that embed jmp_buf or
1824 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1825 against 2.19 headers will have to rebuilt again. This is necessary in any
1826 case to revert the breakage in their ABI caused by the glibc change.
1828 * Support for file description locks is added to systems running the
1829 Linux kernel. The standard file locking interfaces are extended to
1830 operate on file descriptions, not file descriptors, via the use of
1831 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1832 are associated with an open file instead of a process.
1834 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1836 * The minimum Linux kernel version that this version of the GNU C Library
1837 can be used with is 2.6.32.
1839 * Running the testsuite no longer terminates as soon as a test fails.
1840 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1841 with PASS or FAIL lines for individual tests. A summary of the results is
1842 printed, including a list of failing lists, and "make check" exits with
1843 error status if there were any unexpected failures. "make check
1844 stop-on-test-failure=y" may be used to keep the old behavior.
1846 * The am33 port, which had not worked for several years, has been removed
1849 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1850 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1851 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1852 conflicted with POSIX (support for which was removed in 2.19), the
1853 interfaces those macros enabled remain available when compiling with
1854 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1855 test macros defined.
1857 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1859 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1860 This may improve lock scaling of existing programs on TX capable systems.
1861 The lock elision code is only built with --enable-lock-elision=yes and
1862 then requires a GCC version supporting the TX builtins. With lock elision
1863 default mutexes are elided via __builtin_tbegin, if the cpu supports
1864 transactions. By default lock elision is not enabled and the elision code
1867 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1868 copy the path argument. This allowed programs to cause posix_spawn to
1869 deference a dangling pointer, or use an unexpected pathname argument if
1870 the string was modified after the posix_spawn_file_actions_addopen
1873 * All supported architectures now use the main glibc sysdeps directory
1874 instead of some being in a separate "ports" directory (which was
1875 distributed separately before glibc 2.17).
1877 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1878 On configurations that support it (all Linux configurations), it's now
1879 used regardless of the --enable-add-ons switch to configure. It is no
1880 longer possible to build such configurations without pthreads support.
1882 * Locale names, including those obtained from environment variables (LANG
1883 and the LC_* variables), are more tightly checked for proper syntax.
1884 setlocale will now fail (with EINVAL) for locale names that are overly
1885 long, contain slashes without starting with a slash, or contain ".." path
1886 components. (CVE-2014-0475) Previously, some valid locale names were
1887 silently replaced with the "C" locale when running in AT_SECURE mode
1888 (e.g., in a SUID program). This is no longer necessary because of the
1891 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1892 with application code using Intel MPX instructions. (With all previous
1893 versions, the MPX register state could be clobbered when making calls
1894 into or out of a shared library.) Note that while the new dynamic
1895 linker is compatible with all known x86 hardware whether or not it
1896 supports Intel MPX, some x86 instruction-set emulators might fail to
1897 handle the new instruction encodings. This is known to affect Valgrind
1898 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1899 release), and might affect other tools that do instruction emulation.
1901 * Support for loadable gconv transliteration modules has been removed.
1902 The support for transliteration modules has been non-functional for
1903 over a decade, and the removal is prompted by security defects. The
1904 normal gconv conversion modules are still supported. Transliteration
1905 with //TRANSLIT is still possible, and the //IGNORE specifier
1906 continues to be supported. (CVE-2014-5119)
1908 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1909 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1910 resulting a denial-of-service security vulnerability in applications which
1911 use functions related to iconv. (CVE-2014-6040)
1915 * The following bugs are resolved with this release:
1917 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1918 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1919 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1920 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1921 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1922 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1923 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1924 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1925 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1926 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1927 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1928 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1929 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1930 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1931 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1932 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1933 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1934 16431, 16453, 16474, 16506, 16510, 16529
1936 * Slovenian translations for glibc messages have been contributed by the
1937 Translation Project's Slovenian team of translators.
1939 * The public headers no longer use __unused nor __block. This change is to
1940 support compiling programs that are derived from BSD sources and use
1941 __unused internally, and to support compiling with Clang's -fblock
1942 extension which uses __block.
1944 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1945 large collation sequences to optimize multiple passes. This cache
1946 computation may overflow for large collation sequences and may cause a
1947 stack or buffer overflow. This is now fixed to use a slower algorithm
1948 which does not use a cache if there is an integer overflow.
1950 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1951 rules for large collation sequences to optimize multiple passes and falls
1952 back to alloca if malloc fails, resulting in a possible stack overflow.
1953 The implementation now falls back to an uncached collation sequence lookup
1956 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1957 initialized for static applications resulting in the security feature
1958 being disabled. The pointer guard is now correctly initialized to a
1959 random value for static applications. Existing static applications need
1960 to be recompiled to take advantage of the fix (bug 15754).
1962 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1963 to the d_name member of struct dirent, or omit the terminating NUL
1964 character. (Bugzilla #14699).
1966 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1967 aligned_alloc functions could allocate too few bytes or corrupt the
1968 heap when passed very large allocation size values (Bugzilla #15855,
1971 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1972 for AF_INET6 has been fixed (Bugzilla #16072).
1974 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1975 pap_CW, quz_PE, the_NP.
1977 * Substantially revised locales: gd_GB, ht_HT
1979 * The LC_ADDRESS field was updated to support country_car for almost all
1982 * ISO 1427 definitions were updated.
1984 * ISO 3166 definitions were updated.
1986 * The localedef utility now supports --big-endian and --little-endian
1987 command-line options to generate locales for a different system from that
1988 for which the C library was built.
1990 * Binary locale files now only depend on the endianness of the system for
1991 which they are generated and not on other properties of that system. As a
1992 consequence, binary files generated with new localedef may be incompatible
1993 with old versions of the GNU C Library, and binary files generated with
1994 old localedef may be incompatible with this version of the GNU C Library,
1995 in the following circumstances:
1997 + Locale files may be incompatible on m68k systems.
1999 + Locale archive files (but not separate files for individual locales) may
2000 be incompatible on systems where plain "char" is signed.
2002 * The configure option --disable-versioning has been removed. Builds with
2003 --disable-versioning had not worked for several years.
2005 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2006 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2008 * SystemTap probes for malloc have been introduced.
2010 * SystemTap probes for slow multiple precision fallback paths of
2011 transcendental functions have been introduced.
2013 * Support for powerpc64le has been added.
2015 * The soft-float powerpc port now supports e500 processors.
2017 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2019 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2020 set of header declarations that are enabled by default, even when other
2021 feature test macros or compiler options such as -std=c99 would otherwise
2022 disable some of those declarations.
2024 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2025 conflict with POSIX. The libbsd-compat library (which was a dummy library
2026 that did nothing) has also been removed.
2028 * Preliminary documentation about Multi-Thread, Async-Signal and
2029 Async-Cancel Safety has been added.
2031 * Change of ABI data structures for s390 and s390x:
2032 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2033 allow for future hardware extensions. All library functions that accept or
2034 return these structures were versioned in order to provide backwards
2035 compatibility with existing code. However, not all cases can be handled
2036 with interface versioning. If an application embeds either structure into
2037 another structure and that structure is passed to another compilation unit
2038 which expects the newer larger-sized object then any access to the new
2039 fields will result in undefined behaviour. Similarly any access to fields
2040 that were moved by the enlarging of the structures will cause undefined
2041 behaviour. To fix the undefined behaviour all objects that are part of the
2042 final in-memory image must agree on the size of structures, and this may
2043 require recompilation.
2047 * The following bugs are resolved with this release:
2049 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2050 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2051 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2052 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2053 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2054 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2055 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2056 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2057 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2058 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2059 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2060 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2061 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2064 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2065 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2066 Distributions can re-enable building and using pt_chown via the new configure
2067 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2068 considerable security risks and should only be used if the distribution
2069 understands and accepts the risks.
2071 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2074 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2075 fixed (Bugzilla #15330).
2077 * Add support for calling C++11 thread_local object destructors on thread
2078 and program exit. This needs compiler support for offloading C++11
2079 destructor calls to glibc.
2081 * Improved worst case performance of libm functions with double inputs and
2084 * Support for priority inherited mutexes in pthread condition variables on
2085 non-x86 architectures.
2087 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2089 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2091 * Optimized string functions for ARM. Implemented by Will Newton and
2094 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2096 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2099 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2100 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2102 * Added a benchmark framework to track performance of functions in glibc.
2104 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2105 It is based on draft TS 18661 and currently enabled as a GNU extension.
2107 * On Linux, the clock function now uses the clock_gettime system call
2108 for improved precision, rather than old times system call.
2110 * Added support for version-3 tz format files. This is needed when using
2111 the tz database release 2013e or later, and affects a few unusual cases --
2112 currently only TZ='America/Godthab' for time stamps after 2037.
2114 * Added new API functions pthread_getattr_default_np and
2115 pthread_setattr_default_np to get and set the default pthread
2116 attributes of a process.
2118 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2119 This may improve lock scaling of existing programs on TSX capable systems.
2120 When the --enable-lock-elision=yes parameter is specified at configure
2121 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2124 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2125 availability). Implemented by Ryan S. Arnold.
2127 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2131 * The following bugs are resolved with this release:
2133 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2134 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2135 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2136 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2137 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2138 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2139 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2140 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2141 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2142 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2143 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2144 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2145 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2146 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2147 14879, 14889, 14893, 14898, 14914.
2149 * Optimization of memcpy for MIPS.
2151 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2152 EMFILE has been fixed (Bugzilla #14889).
2154 * The add-on ports collection is now distributed in the "ports" subdirectory
2155 of the main GNU C Library distribution, rather than separately.
2157 * Port to ARM AArch64 contributed by Linaro.
2159 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2160 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2162 Implemented by Andreas Krebbel.
2164 * The new function secure_getenv allows secure access to the environment,
2165 returning NULL if running in a SUID/SGID process. This function replaces
2166 the internal function __secure_getenv.
2168 * SystemTap static probes have been added into the dynamic linker.
2169 Implemented by Gary Benson.
2171 * Optimizations of string functions strstr, strcasestr and memmem.
2172 Implemented by Maxim Kuvyrkov.
2174 * The minimum Linux kernel version that this version of the GNU C Library
2175 can be used with is 2.6.16.
2177 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2178 powerpc POWER7. Implemented by Will Schmidt.
2180 * New configure option --disable-nscd builds the C library such that it
2181 never attempts to contact the Name Service Caching Daemon (nscd).
2182 New configure option --disable-build-nscd avoids building nscd itself;
2183 this is the default if --disable-nscd is used.
2185 * Improved support for cross-compilation, including cross-testing and
2186 bootstrap builds without a previously built glibc.
2188 * Several testsuite tests are now able to test multiple IFUNC variants of an
2189 interface, rather than just testing the one that would be chooen by
2192 * New configure options --with-bugurl and --with-pkgversion, for
2193 distributors to use to embed their bug-reporting and package version
2194 information in --help and --version output.
2196 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2197 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2198 allows creation of chroots without the procfs mounted on /proc.
2200 * The `crypt' function now fails if passed salt bytes that violate the
2201 specification for those values. On Linux, the `crypt' function will
2202 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2203 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2204 when the mode is enabled.
2206 * The `clock_*' suite of functions (declared in <time.h>) is now available
2207 directly in the main C library. Previously it was necessary to link with
2208 -lrt to use these functions. This change has the effect that a
2209 single-threaded program that uses a function such as `clock_gettime' (and
2210 is not linked with -lrt) will no longer implicitly load the pthreads
2211 library at runtime and so will not suffer the overheads associated with
2212 multi-thread support in other code such as the C++ runtime library.
2214 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2219 * The following bugs are resolved with this release:
2221 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2222 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2223 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2224 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2225 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2226 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2227 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2228 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2229 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2230 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2231 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2232 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2233 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2234 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2235 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2236 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2237 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2238 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2239 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2240 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2241 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2242 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2245 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2246 configuring glibc with:
2247 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2248 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2249 Implemented by H.J. Lu.
2253 + define static_assert
2255 + do not declare gets
2257 + declare at_quick_exit and quick_exit also for ISO C11
2259 + aligned_alloc. NB: The code is deliberately allows the size parameter
2260 to not be a multiple of the alignment. This is a moronic requirement
2261 in the standard but it is only a requirement on the caller, not the
2264 + timespec_get added
2266 + uchar.h support added
2268 + CMPLX, CMPLXF, CMPLXL added
2270 Implemented by Ulrich Drepper.
2272 * Support for the IA-64 has been moved to ports.
2274 * Remove support for anything but ELF binary format
2276 * Checking versions of poll, ppoll added.
2277 Implemented by Ulrich Drepper.
2279 * More generic and 64-bit performance optimizations to math functions.
2280 Implemented by Ulrich Drepper.
2282 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2283 headers and functions available at compile time as they were before
2284 version 2.14. This option will be removed at some time in the future
2285 after the TI-RPC library becomes fully sufficient for the needs of
2286 existing applications.
2288 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2289 Note that glibc is not expected to work with any Linux kernel version
2292 * New header <sys/auxv.h> and function getauxval allowing easy access to
2293 the AT_* key-value pairs passed from the Linux kernel. The header also
2294 defines the HWCAP_* bits associated with the AT_HWCAP key.
2296 * New locales: mag_IN
2298 * New configure option --enable-systemtap builds SystemTap static probes
2299 into libc for setjmp and longjmp and into libpthread for various operations.
2300 So far the setjmp/longjmp probes and some of the libpthread probes are
2301 provided only for i*86 and x86_64.
2302 Implemented by Roland McGrath and Rayson Ho.
2304 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2306 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2309 * More optimized functions for SPARC. Implemented by David S. Miller.
2311 * Improved support for cross-compilation, especially bootstrap builds
2312 without a previously built glibc.
2314 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2315 Chris Metcalf from Tilera.
2317 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2318 now supported for ARM processors.
2320 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2321 as the name of the dynamic linker, to distinguish it from the
2322 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2324 * Support for CRIS has been removed from ports.
2326 * A new class of installed header has been documented for low-level
2327 platform-specific functionality. PowerPC added the first instance with a
2328 function to provide time base register access. Contributed by Tulio
2329 Magno Quites Machado Filho.
2331 * ix86 configurations now install header files that are consistent with
2332 what x86-64 configurations install. These same header files can be used
2333 for -m32, -m64, or -mx32 builds.
2334 Contributed by H.J. Lu.
2336 * Math library bug fixes. A thorough audit of all open math library bugs was
2337 conducted by Joseph Myers. Significant progress was made on many math
2338 library bugs resulting in more accurate exceptions and function results.
2339 Many thanks to all those that contributed including Andreas Jaeger for his
2340 patch review and work on the x87 trigonometric instruction issues.
2342 * Timezone data is no longer installed. Timezone-related binaries and scripts
2343 will continue to be installed. Users should obtain their timezone data from
2344 their distribution provider or from the tzdata package at
2345 <ftp://munnari.oz.au/pub/>.
2349 * The following bugs are resolved with this release:
2351 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2352 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2353 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2354 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2355 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2356 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2357 13484, 13506, 13515, 13523, 13524, 13538, 13540
2359 * New program pldd to list loaded object of a process
2360 Implemented by Ulrich Drepper.
2362 * Add nss_db support back to glibc. No more dependency on Berkeley db
2363 and support for initgroups lookups.
2364 Implemented by Ulrich Drepper.
2366 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2367 Contributed by HJ Lu.
2369 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2370 Contributed by HJ Lu.
2372 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2373 on x86-32 and x86-64.
2374 Contributed by Liubov Dmitrieva.
2376 * Optimized strchr and strrchr for SSE on x86-32.
2377 Contributed by Liubov Dmitrieva.
2379 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2380 for x86-64 and x86-32.
2381 Contributed by Liubov Dmitrieva.
2383 * New interfaces: scandirat, scandirat64
2384 Implemented by Ulrich Drepper.
2386 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2387 Implemented by Ulrich Drepper.
2389 * nscd now also caches the netgroup database.
2390 Implemented by Ulrich Drepper.
2392 * Integrate libm with gcc's -ffinite-math-only option.
2393 Implemented by Ulrich Drepper.
2395 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2396 to math functions. Implemented by Ulrich Drepper.
2398 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2399 Implemented by Ulrich Drepper.
2401 * New Linux interfaces: process_vm_readv, process_vm_writev
2403 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2404 Implemented by Ulrich Drepper.
2406 * Optimized nearbyint and strcasecmp for PPC.
2407 Implemented by Adhemerval Zanella.
2409 * New locales: bho_IN, unm_US, es_CU, ta_LK
2413 * The following bugs are resolved with this release:
2415 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2416 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2417 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2418 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2419 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2420 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2421 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2422 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2423 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2424 12795, 12811, 12813, 12814, 12841
2426 * The RPC implementation in libc is obsoleted. Old programs keep working
2427 but new programs cannot be linked with the routines in libc anymore.
2428 Programs in need of RPC functionality must be linked against TI-RPC.
2429 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2431 Visible changes of this change include (obviously) the inability to link
2432 programs using RPC functions without referencing the TI-RPC library and the
2433 removal of the RPC headers from the glibc headers.
2434 Implemented by Ulrich Drepper.
2436 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2437 syncfs, setns, sendmmsg
2439 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2440 yue_HK, lij_IT, mhr_RU
2442 * New encodings: CP770, CP771, CP772, CP773, CP774
2444 * New program sotruss to trace calls through PLTs
2445 Implemented by Ulrich Drepper.
2447 * The malloc hook implementation is marked deprecated and will be removed
2448 from the default implementation in the next version. The design never
2449 worked ever since the introduction of threads. Even programs which do
2450 not create threads themselves can use multiple threads created internally.
2454 * The following bugs are resolved with this release:
2456 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2457 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2458 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2459 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2462 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2464 * POWER7 optimizations: memset, memcmp, strncmp
2466 * New optimized string functions for x86-64: strnlen (SSE2),
2467 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2468 Implemented by Ulrich Drepper.
2472 * The following bugs are resolved with this release:
2474 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2475 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2476 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2477 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2478 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2479 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2480 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2481 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2483 * New interfaces: pthread_getname_np, pthread_setname_np
2485 * New Linux interface: recvmmsg
2487 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2489 * The dynamic linker now recognizes supported ABI versions from the
2490 EI_ABIVERSION field in the ELF header.
2491 Implemented by Ulrich Drepper.
2493 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2494 will not be used to fill in password fields in the passwd.byname replies.
2495 Instead it is used to synthesize the shadow.byname table, should it be
2496 missing. This is a useful mode in some installations involving Solaris.
2497 Implemented by Ulrich Drepper.
2499 * New locales: kok_IN, sq_MK, cv_RU
2503 * The following bugs are resolved with this release:
2505 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2506 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2507 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2508 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2509 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2510 10789, 10817, 10823, 10840, 10847
2512 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2514 Implemented by Ulrich Drepper.
2516 * Checking version of longjmp added that fails if an uninitialized stack
2517 frame would be created. Implemented by Ulrich Drepper.
2519 * STT_GNU_IFUNC is now supported in static executables.
2520 Implemented by H.J. Lu.
2522 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2524 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2525 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2526 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2527 Contributed by H.J. Lu.
2529 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2530 Implemented by Ulrich Drepper.
2532 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2534 Contributed by H.J. Lu.
2536 * Support for fma instruction in AVX on x86-64.
2537 Implemented by H.J. Lu and Ulrich Drepper.
2539 * AVX support in x86-64 auditing support in ld.so.
2540 Implemented by H.J. Lu.
2542 * STB_GNU_UNIQUE support added.
2543 Implemented by Ulrich Drepper.
2545 * Implement second fallback mode for optimized DNS lookup for even more
2546 broken environments. If two requests from the same port are not
2547 handled correctly close the socket and open a new one before sending
2548 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2549 can be used to select this mode right away, instead of rediscovering the
2550 necessity is every process again.
2551 Implemented by Ulrich Drepper.
2553 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2554 Implemented by Adam Tkac.
2556 * Optimized iconv conversions for S390x.
2557 Implemented by Andreas Krebbel.
2559 * Using condvars with PI mutexes is now more efficient due to kernel
2560 support for requeueing to PI futexes. NPTL support added for x86-64.
2561 Implemented by Ulrich Drepper.
2563 * New locales: ps_AF, my_MM
2567 * The following bugs are resolved with this release:
2569 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2570 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2571 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2572 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2573 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2575 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2576 Implemented by Ulrich Drepper.
2578 * New Linux interfaces: accept4, fallocate, fallocate64.
2579 Implemented by Ulrich Drepper.
2581 * Correct declarations of string function when used in C++ code. This
2582 could lead to compile errors for invalid C++ code.
2584 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2587 * New POSIX 2008 interface: psiginfo
2588 Implemented by Ulrich Drepper.
2590 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2591 Implemented by Ulrich Drepper.
2593 * Support for selecting between multiple function definitions at runtime
2594 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2596 * The libcrypt library can now use the hash function implementations in
2597 NSS. Implemented by Ulrich Drepper.
2599 * The malloc implementation can be compiled to be less memory efficient
2600 but higher performing in multi-threaded programs.
2601 Implemented by Ulrich Drepper.
2603 * New locales: nan_TW@latin, ks_IN
2605 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2606 Implemented by Ulrich Drepper.
2608 * Extended printf hook support. It is possible to use user-defined types
2609 and extend existing format specifiers.
2610 Implemented by Ulrich Drepper.
2612 * Handling for group shadow files has been added.
2613 Implemented by Ulrich Drepper.
2615 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2616 servers (the case, e.g., for some people using the built-in DNS
2617 server in ADSL modems/routers). There is a once-per-process timeout
2618 in case of a broken server. To avoid it, users can run nscd or put
2619 'options single-request' in /etc/resolv.conf.
2620 Implemented by Ulrich Drepper.
2624 * The following bugs are resolved with this release:
2626 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2627 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2628 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2629 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2630 6968, 6974, 6980, 6995, 7008, 7009, 7029
2632 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2633 up at the same time. Implemented by Ulrich Drepper.
2635 * TLS descriptors for LD and GD on x86 and x86-64.
2636 Implemented by Alexandre Oliva.
2638 * getaddrinfo now handles DCCP and UDPlite.
2639 Implemented by Ulrich Drepper.
2641 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2642 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2643 Implemented by Ulrich Drepper.
2645 * New implementation of memmem, strstr, and strcasestr which is O(n).
2646 Implemented by Eric Blake.
2648 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2650 * Implement "e" option for popen to open file descriptor with the
2651 close-on-exec flag set. Implemented by Ulrich Drepper.
2653 * Implement "b" mode for fmemopen. In this mode writes writes don't
2654 implicitly add a NUL byte and seeks from the end of the buffer really
2655 use the buffer end, not the string length as the basis.
2656 Implemented by Ulrich Drepper.
2658 * Many functions, exported and internal, now atomically set the close-on-exec
2659 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2661 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2663 Implemented by Pravin Satpute.
2665 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2669 * The following bugs are resolved with this release:
2671 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2672 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2673 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2674 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2675 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2676 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2677 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2679 * New locales: bo_CN, bo_IN, shs_CA.
2681 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2683 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2684 Implemented by Pravin Satpute.
2686 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2688 * nscd caches now all timeouts for DNS entries
2689 Implemented by Ulrich Drepper.
2691 * nscd is more efficient and wakes up less often.
2692 Implemented by Ulrich Drepper.
2694 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2695 vdprintf, and obstack_vprintf.
2696 Implemented by Jakub Jelinek.
2698 * Faster memset for x86-64.
2699 Implemented by Harsha Jagasia and H.J. Lu.
2701 * Faster memcpy on x86.
2702 Implemented by Ulrich Drepper.
2704 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2705 Implemented by Ulrich Drepper.
2707 * Faster sqrt and sqrtf implemention for some PPC variants.
2708 Implemented by Steven Munroe.
2712 * The following bugs are resolved with this release:
2714 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2715 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2716 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2717 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2718 5071, 5103, 5104, 5112, 5113, 5184, 5186
2720 * More checking functions: fread, fread_unlocked, open*, mq_open.
2721 Implemented by Jakub Jelinek and Ulrich Drepper.
2723 * Extend fortification to C++. Implemented by Jakub Jelinek.
2725 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2726 by not recognizing 'a' as a modifier when those specs are requested.
2727 Implemented by Jakub Jelinek.
2729 * PPC optimizations to math and string functions.
2730 Implemented by Steven Munroe.
2732 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2733 options to be passed. Implemented by Ulrich Drepper.
2735 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2737 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2738 Implemented by Ulrich Drepper.
2740 * Handle private futexes in the NPTL implementation.
2741 Implemented by Jakub Jelinek and Ulrich Drepper.
2743 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2744 Implemented by Roland McGrath and Ulrich Drepper.
2746 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2748 * SHA-256 and SHA-512 based password encryption.
2749 Implemented by Ulrich Drepper.
2751 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2752 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2755 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2756 Implemented by Ulrich Drepper.
2760 * The following bugs are resolved with this release:
2762 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2763 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2764 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2765 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2766 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2767 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2768 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2771 * New Linux interfaces: epoll_pwait, sched_getcpu.
2773 * New generic interfaces: strerror_l.
2775 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2779 * The following bugs are resolved with this release:
2781 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2782 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2783 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2784 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2785 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2786 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2787 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2788 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2789 3225, 3251, 3252, 3253, 3273
2791 * For Linux, the sorting of addresses returned by getaddrinfo now also
2792 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2793 Implemented by Ulrich Drepper.
2795 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2796 Implemented by Ulrich Drepper.
2798 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2800 * New iconv module for MIK. Contributed by Alexander Shopov.
2802 * For sites with broken group and/or passwd database, the auto-propagate
2803 option of nscd can prevent creating ID lookup entries from the results
2804 of a name lookup and vice versa. This usually is no problem but some
2805 site might have problems with the default behavior.
2806 Implemented by Ulrich Drepper.
2808 * Iterating over entire database in NIS can be slow. With the
2809 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2810 to trade time for memory. The entire database will be read at once.
2811 Implemented by Ulrich Drepper.
2813 * The interfaces introduced in RFC 3542 have been implemented by
2816 * Support for the new ELF hash table format was added by Ulrich Drepper.
2818 * Support for priority inheritance mutexes added by Jakub Jelinek and
2821 * Support for priority protected mutexes added by Jakub Jelinek.
2823 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2824 or_IN, csb_PL, fy_NL, sr_ME.
2828 * More overflow detection functions.
2830 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2831 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2833 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2834 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2835 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2836 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2837 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2838 by Masahide Washizawa.
2840 * It is now possible to install audit libraries for the dynamic linker using
2841 LD_AUDIT. Implemented by Ulrich Drepper.
2843 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2844 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2845 For a libc and libpthread that works well on Linux 2.4 kernels, we
2846 recommend using the stable 2.3 branch.
2848 * The new function open_wmemstream acts like open_memstream,
2849 but uses a wchar_t wide character buffer.
2851 * The new function ppoll is an improved version of poll, similar to pselect.
2853 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2854 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2855 renameat, unlinkat, linkat, symlinkat, readlinkat.
2857 * New Linux kernel system calls: unshare,
2858 inotify_init, inotify_add_watch, inotify_rm_watch.
2860 * The euidaccess function is now also known by the alias eaccess,
2861 for compatibility with some other systems.
2863 * Timezone data updated to 2006b version.
2867 * The following bugs are resolved with this release:
2869 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2870 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2871 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2872 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2873 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2874 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2876 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2878 * As of this release, GCC 4 can be used to compile the C Library.
2880 * Timezone data updated to 2005m version.
2884 * The following bugs are resolved with this release:
2886 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2887 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2888 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2889 777, 787, 821, 822, 823, 825
2891 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2895 * Support for RFC 3678. Real implementations exist only for Linux so far.
2896 Implemented by Ulrich Drepper.
2898 * nscd can now cache entries persistently. Expiring entries are reloaded.
2899 For speedups the cache can be shared in memory with client processes.
2900 Implemented by Ulrich Drepper.
2902 * nscd can now perform SELinux checks.
2903 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2905 * getaddrinfo queries are now cached. Canonical name lookup is performed
2907 Implemented by Ulrich Drepper.
2909 * The nothrow function attribute is used when headers are used by gcc when
2910 compiling C code. This can avoid significant amounts of exception
2913 * The malloc functions perform more error checking and are stricter when
2914 it comes to reacting on errors. The default action is to terminate
2915 the process after showing an error message. Implemented by Ulrich Drepper.
2917 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2918 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2920 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2921 namespaces using the new function dlmopen(). This feature is of course,
2922 like most other dynamic loading functionality, not available in statically
2923 linked applications. Implemented by Ulrich Drepper.
2925 * Low-overhead boundary checking variants of string and some stdio functions
2926 were added. These are to be used in conjunction with a gcc patch by
2927 Jakub Jelinek which adds calls to these functions if possible.
2928 Implemented by Jakub Jelinek and Ulrich Drepper.
2930 * Old code for several operating systems and machine architectures that
2931 have not been in working condition in a long time have been removed from
2932 the main source tree maintained by the GNU C Library's maintainers.
2933 These files are now reside in the separate `ports' source module
2934 that is usable as an add-on when building the library.
2938 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2939 interrogate the dynamic linker, compatible with the Solaris interface.
2941 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2942 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2944 * getifaddrs now uses the netlink interface on Linux to get its information.
2945 Implemented by Thorsten Kukuk.
2947 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2948 Implemented by Ulrich Drepper.
2950 * support for non-executable stacks on x86 has been added. Changes mostly
2953 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2956 * getaddrinfo now performs destination address selection according to
2961 * Thread-safe interfaces for many functions that access locale data
2962 were added in version 2.3, but these features were omitted from NEWS.
2963 Many functions have variants with an `_l' suffix that take a `locale_t'
2964 object as a parameter rather than consulting the current locale.
2965 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2966 create and maintain `locale_t' objects. Additionally, the new function
2967 `uselocale' sets "the current locale" (as used by functions not so
2968 parameterized) set for an individual thread. These features were added
2969 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2971 * The functions getresuid, getresgid, setresuid, and setresgid, which
2972 have long been available on Linux, are now declared in <unistd.h>
2973 and are now also available on the Hurd.
2975 * ELF thread-local storage support (TLS) now works on x86-64.
2977 * The new dynamic string token $LIB is expanded in shared library names.
2978 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2980 * Aldy Hernandez contributed complete software floating point support for
2981 PowerPC machines with no FPU.
2983 * fexecve is implemented on Linux.
2985 * The `btowc' function should work at least twice as fast due to
2986 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2988 * With approriate thread add-ons cancelable functions are now implemented
2989 in libc.so as well. No need to call the function in libpthread. This
2990 change allowed to finally disable the incorrect and expensive handling
2991 of weak definition in ld.so.
2993 * Yet more PLT entries in libc.so have been removed. We finally arrived
2994 at the bare minimum. Startup times improved appropriately.
2996 * Support for the new Linux/x86 system call interface was added. The
2997 AT_SYSINFO auxiliary vector entry is recognized and handled.
3001 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3004 * iconv (the program and the interface) now accepts empty names (excluding
3005 options like //TRANSLIT) to mean "use charset of current locale".
3007 * localedef can now transliterate characters in strings which are not in
3008 the provided charmap. The information from the input locale is used.
3010 * Prelinking support was added for ELF targets. This requires additional
3011 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3013 * Read-only stdio streams now use mmap to speed up operation by eliminating
3014 copying and buffer underflows. To use add 'm' to the mode string of
3015 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3017 * The malloc functions were completely rewritten by Wolfram Gloger based
3018 on Doug Lea's malloc-2.7.0.c.
3020 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3021 implementation of regex.
3023 * Bruno Haible upgraded the iconv and locale implementation to support
3026 * Contents of the LC_* and LANG environment variables in the CEN style are
3027 not recognized anymore. It never was used. Change by Ulrich Drepper.
3029 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3030 thread-local storage (TLS) ABI on some platforms.
3031 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3033 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3034 EUC-JISX0213, and TSCII.
3036 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3037 BSD-compatible interface for getting all network interface addresses.
3038 Implementation for IPv4 by Roland McGrath.
3040 * Loading of locale data is faster due to the introduction of a locale
3041 archive. Implemented by Roland McGrath and Ulrich Drepper.
3043 * Startup times are significantly reduced by not using exported functions
3044 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3047 * Steven Munroe contributed a port to PowerPC64/Linux.
3051 * The Hurd now uses the GNU libio implementation of stdio rather than the
3052 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3054 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3055 with the same behavior as the Linux system call of the same name.
3059 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3060 128-bit long double format.
3062 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3063 IBM1161, and IBM1162 charsets.
3065 * Andreas Jaeger contributed a port to x86-64/Linux.
3067 * Peter Bruin contributed a port to PowerPC/Hurd.
3069 * libc functions using I/O streams now can handle wide-oriented streams
3072 * optimizations in the dynamic linker. Binaries created by recent binutils
3073 versions start up quicker due to reduced time spent on relocations.
3075 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3079 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3080 asinh, atanh, j0 for the 128-bit long double format.
3082 * Bruno Haible updated all the code handling Unicode in some form to
3083 support Unicode 3.1.
3085 * Speed of regex for single-byte locales is back to previous levels.
3086 Patch by Isamu Hasegawa.
3088 * Alpha, SPARC, and IA-64 now also using floating stacks.
3090 * Startup time of internationalized applications greatly improved through
3091 iconv cache. Use iconvconfig to generate the cache file.
3092 Contributed by Ulrich Drepper.
3094 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3095 severe performance problems. Patch by David Mosberger.
3097 * The Hurd port got a lot more functionality like AIO, various stdio
3098 extensions, etc. Mainly done by Roland McGrath.
3100 * mtrace can now lookup symbols in shared libraries.
3104 * Intel's IA-64 math library is largely integrated. It provides fast and
3105 accurate implementatations for most basic and standard math functions
3106 in float, double, and long double format.
3108 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3109 and asin for the 96-bit long double format and asin, log, tan for the
3110 128-bit long double format.
3112 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3113 The basic double functions exist today. Contributed by Abraham Ziv
3114 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3115 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3117 * An asynchronous name lookup library was added. The interface is designed
3118 after POSIX AIO. The proposal was circulated beforehand to get comments.
3119 No negative ones came in. Implemented by Ulrich Drepper.
3121 * Port to S390/64bit contributed by Martin Schwidefsky
3122 <schwidefsky@de.ibm.com>.
3124 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3125 of functions for Linux/IA-64.
3127 * The RPC code is now thread safe. Threads can now use the same service
3128 of different services at the same time. Patch by Eric Norum
3129 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3131 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3132 family of functions for Linux/S390.
3134 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3135 of functions for Linux/x86.
3137 * Port to Linux/CRIS contributed by Axis Communications.
3141 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3142 we can now check for namespace violations and missing declarations. The
3143 result is that almost all headers are now Unix-compliant (as defined in
3144 the upcoming XPG6). The negative side is that some programs might need
3145 corrections, too, if they depend on the incorrect form of the headers in
3146 previous versions which defined too many symbols and included too many
3149 * regex now handles multibyte character sets correctly.
3150 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3152 * iconv (the program) does now conform to the upcoming XPG6 and handles
3153 charmaps. Instead of the charset names the path of charmaps can be
3154 provided and the conversion happens based on this data.
3155 Contributed by Ulrich Drepper.
3157 * The locale program now provides detailed information about the installed
3162 only lists the names of the supported locales
3166 provides details such as country, language, and codeset name.
3167 Contributed by Ulrich Drepper.
3171 * The gencat program now parses the input file according to the charset
3172 selected by the LC_CTYPE category. This is important for stateful
3173 character sets. To make generating catalogs easier there is a way
3174 to overwrite the charset selected by the locale: before the first
3175 message or $ quote line the catalog can contain a line like
3177 $ codeset=ISO-8859-2
3179 to select the charset (ISO-8859-2 in this case).
3181 Implemented by Shinya Hanataka and Ulrich Drepper.
3183 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3184 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3185 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3187 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3188 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3190 * The LANGUAGE environment variable is now ignored unless the locale is
3191 changed from the default "C" locale.
3193 * The usual bug fixes.
3197 * Greg McGary added runtime support for bounds checking using gcc's
3198 new -fbounded-pointers option. ix86 target is complete. PowerPC
3201 * Thorsten Kukuk added secure mode support to nscd.
3203 * The Berkeley db libraries have been removed.
3205 Related, the nss_db module is now in a separate package since it
3206 obviously requires a database library being available.
3208 * Wide character I/O streams implemented by Ulrich Drepper.
3210 * Functions from the extended socket API added by Ulrich Drepper.
3212 * Functions feenableexcept and fedisableexcept to control the
3213 behaviour of individual exceptions have been added by Andreas Jaeger.
3215 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3217 * The resolver code has been updated from bind 8.2.3-T5B which supports
3218 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3221 This change could in some situations effect backward compatibility. Since
3222 now `_res' is a thread-local instead of a global variable, modifying it
3223 in one thread does not have any effect in other threads.
3225 The resolver library was also extended to allow IPv6 as the transport
3226 protocol for the requests. This work was done by Stig Venaas.
3228 * Compatibility code for K&R C compilers has been removed from the
3229 header files. A ISO C compiler is needed to use the library
3230 (conforming to either C89 or C99 standard).
3232 * Complete rewrite of the localedef program to support multibyte character
3233 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3234 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3235 Implemented by Ulrich Drepper.
3237 Bruno Haible significantly improved the generation and use of the data
3238 structures for the wide character tables.
3240 * Plural handling in gettext implemented by Ulrich Drepper.
3242 * The utmp daemon has been removed.
3244 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3246 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3249 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3251 * POSIX barriers implemented by Kaz Kylheku.
3253 * POSIX spawn function family implemented by Ulrich Drepper.
3255 * POSIX spinlocks are now available.
3257 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3259 * the configure option --enable-kernel=X.Y.Z allows to strip out
3260 compatibility for kernel versions before X.Y.Z. This is currently only
3261 implemented for Linux.
3263 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3264 field sin6_scope_id. This means that all programs using IPv6 should be
3265 recompiled. Don't expect binary compatibility with previous glibc
3268 * various conversion modules for IBM character sets contributed by
3271 * IA-64 port by Jes Sorensen and HJ Lu.
3285 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3287 * New xdr functions are added; some rpc functions are now 64bit clean.
3289 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3291 * Fixed known incompatibilities with glibc 2.0.
3293 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3295 * Optimized a number of functions (especially the ELF dynamic loader).
3297 * Update timezone data files.
3299 * lots of charmaps corrections
3301 * some new locale definitions and charmaps
3306 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3307 conform to POSIX member type requirements. Symbol versions have been
3308 adjusted as needed within the library, and for direct use by applications,
3309 but there is potential for problems if third-party libraries use
3310 struct timeval as part of their interface. This does not present
3311 a problem for X and other "essential" system libraries.
3313 * An additional locale model to support C++ Standard Library locale
3314 model and probably more was implemented by Ulrich Drepper.
3316 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3319 * Miles Bader provided the `argp' function family to support hierachical
3320 command line argument parsing, layered on top of getopt.
3322 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3324 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3327 * scanf recognizes the %a and %A format for scanning floating point numbers.
3329 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3330 information and interfaces for the available integer types.
3332 * about 130 new math functions were added to implement the ISO C9x math
3335 * the new header <complex.h> contains definitions of the complex math
3336 functions from ISO C 9X.
3338 * the new header <tgmath.h> defines generic macros to use complex or
3339 real valued functions.
3341 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3343 * Andreas Jaeger provided a test suite for the math library.
3345 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3347 * Ulrich Drepper added character set conversion functions (iconv).
3349 * Optimized string functions have been added.
3351 * The localedata addon is now part of glibc.
3353 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3355 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3356 daemon for NSS (nscd).
3358 Missing a better place here are some numbers on improvements. Under
3359 Linux 2.1.125 un-tar-ing the kernel sources takes
3363 using local files 12.19s 6.88s 22.91s
3365 using NIS 13.92s 8.91s 26.34s
3367 using NIS & nscd 10.37s 7.34s 25.30s
3369 using NIS+ 27.57s 30.37s 640.46s
3371 using NIS+ & nscd 10.25s 7.83s 26.51s
3373 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3375 Keep in mind that non-namelookup related operations dominate above times.
3376 It was just a common complain that using NIS+ unpacking the kernel is
3379 [1] The old nscd implementation is not available anymore since it was
3380 distributed with glibc up to version 2.0.98 and thus is now replaced.
3382 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3384 * Mark Kettenis provided a Hesiod NSS module.
3386 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3387 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3389 * rcmd can now handle netgroups (Dick Streefland).
3391 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3394 * Support for the IPv6 protocol has been added to the socket API, as per the
3395 latest draft standards.
3397 * Support for Linux 2.2 has been added.
3399 * Interface changes relative to the latest 2.0.x release:
3400 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3401 addseverity NEW: Unix98
3402 alphasort64 NEW: LFS
3403 argp_err_exit_status NEW: argp, GNU ext
3404 argp_error NEW: argp, GNU ext
3405 argp_failure NEW: argp, GNU ext
3406 argp_help NEW: argp, GNU ext
3407 argp_parse NEW: argp, GNU ext
3408 argp_program_bug_address NEW: argp, GNU ext
3409 argp_program_version NEW: argp, GNU ext
3410 argp_program_version_hook NEW: argp, GNU ext
3411 argp_state_help NEW: argp, GNU ext
3412 argp_usage NEW: argp, GNU ext
3413 authdes_create NEW: Secure RPC
3414 authdes_getucred NEW: Secure RPC
3415 authdes_pk_create NEW: Secure RPC
3416 backtrace NEW: GNU ext.
3417 backtrace_symbols NEW: GNU ext.
3418 backtrace_symbols_fd NEW: GNU ext.
3420 cacosf NEW: ISO C 9x
3421 cacosh NEW: ISO C 9x
3422 cacoshf NEW: ISO C 9x
3423 cacoshl NEW: ISO C 9x
3424 cacosl NEW: ISO C 9x
3431 casinf NEW: ISO C 9x
3432 casinh NEW: ISO C 9x
3433 casinhf NEW: ISO C 9x
3434 casinhl NEW: ISO C 9x
3435 casinl NEW: ISO C 9x
3437 catanf NEW: ISO C 9x
3438 catanh NEW: ISO C 9x
3439 catanhf NEW: ISO C 9x
3440 catanhl NEW: ISO C 9x
3441 catanl NEW: ISO C 9x
3442 cbc_crypt NEW: Secure RPC
3446 ccoshf NEW: ISO C 9x
3447 ccoshl NEW: ISO C 9x
3453 cimagf NEW: ISO C 9x
3454 cimagl NEW: ISO C 9x
3455 clearerr_locked REMOVED
3456 clntunix_create NEW: sunrpc ext
3458 clog10 NEW: ISO C 9x
3459 clog10f NEW: ISO C 9x
3460 clog10l NEW: ISO C 9x
3470 cprojf NEW: ISO C 9x
3471 cprojl NEW: ISO C 9x
3473 crealf NEW: ISO C 9x
3474 creall NEW: ISO C 9x
3479 csinhf NEW: ISO C 9x
3480 csinhl NEW: ISO C 9x
3483 csqrtf NEW: ISO C 9x
3484 csqrtl NEW: ISO C 9x
3488 ctanhf NEW: ISO C 9x
3489 ctanhl NEW: ISO C 9x
3491 des_setparity NEW: Secure RPC
3492 ecb_crypt NEW: Secure RPC
3493 endutxent NEW: Unix98
3495 exp10f NEW: ISO C 9x
3496 exp10l NEW: ISO C 9x
3500 fattach NEW: STREAMS
3501 fdetach NEW: STREAMS
3505 feclearexcept NEW: ISO C 9x
3506 fegetenv NEW: ISO C 9x
3507 fegetexceptflag NEW: ISO C 9x
3508 fegetround NEW: ISO C 9x
3509 feholdexcept NEW: ISO C 9x
3511 feraiseexcept NEW: ISO C 9x
3512 ferror_locked REMOVED
3513 fesetenv NEW: ISO C 9x
3514 fesetexceptflag NEW: ISO C 9x
3515 fesetround NEW: ISO C 9x
3516 fetestexcept NEW: ISO C 9x
3517 feupdateenv NEW: ISO C 9x
3518 fflush_locked REMOVED
3522 fgets_unlocked NEW: GNU ext.
3523 fileno_locked REMOVED
3535 fputc_locked REMOVED
3536 fputs_unlocked NEW: GNU ext.
3537 fread_unlocked NEW: GNU ext.
3542 fstatvfs NEW: Unix98
3546 ftruncate64 NEW: LFS
3548 fwrite_unlocked NEW: GNU ext.
3549 gai_strerror NEW: IPv6
3553 getchar_locked REMOVED
3555 getdate_err NEW: Unix98
3556 getdate_r NEW: GNU ext.
3558 getnameinfo NEW: IPv6
3559 getnetname NEW: Secure RPC
3560 getpmsg NEW: STREAMS
3561 getpt NEW: Unix98 PTY
3562 getrlimit64 NEW: LFS
3563 getutxent NEW: Unix98
3564 getutxid NEW: Unix98
3565 getutxline NEW: Unix98
3566 glob64 NEW: GNU ext.
3567 globfree64 NEW: GNU ext.
3568 gnu_get_libc_release NEW: GNU ext.
3569 gnu_get_libc_version NEW: GNU ext.
3570 grantpt NEW: Unix98 PTY
3571 host2netname NEW: Secure RPC
3573 iconv_close NEW: iconv
3574 iconv_open NEW: iconv
3575 if_freenameindex NEW: IPv6
3576 if_indextoname NEW: IPv6
3577 if_nameindex NEW: IPv6
3578 if_nametoindex NEW: IPv6
3579 in6addr_any NEW: IPv6
3580 in6addr_loopback NEW: IPv6
3581 inet6_isipv4mapped NEW: IPv6
3582 isastream NEW: STREAMS
3583 iswblank NEW: GNU ext.
3584 key_decryptsession NEW: Secure RPC
3585 key_decryptsession_pk NEW: Secure RPC
3586 key_encryptsession NEW: Secure RPC
3587 key_encryptsession_pk NEW: Secure RPC
3588 key_gendes NEW: Secure RPC
3589 key_get_conv NEW: Secure RPC
3590 key_secretkey_is_set NEW: Secure RPC
3591 key_setnet NEW: Secure RPC
3592 key_setsecret NEW: Secure RPC
3593 llrint NEW: ISO C 9x
3594 llrintf NEW: ISO C 9x
3595 llrintl NEW: ISO C 9x
3596 llround NEW: ISO C 9x
3597 llroundf NEW: ISO C 9x
3598 llroundl NEW: ISO C 9x
3603 lrintf NEW: ISO C 9x
3604 lrintl NEW: ISO C 9x
3605 lround NEW: ISO C 9x
3606 lroundf NEW: ISO C 9x
3607 lroundl NEW: ISO C 9x
3609 makecontext NEW: Unix98
3610 mempcpy NEW: GNU ext.
3613 modify_ldt NEW: kernel
3617 nearbyint NEW: ISO C 9x
3618 nearbyintf NEW: ISO C 9x
3619 nearbyintl NEW: ISO C 9x
3620 netname2host NEW: Secure RPC
3621 netname2user NEW: Secure RPC
3622 nexttoward NEW: ISO C 9x
3623 nexttowardf NEW: ISO C 9x
3624 nexttowardl NEW: ISO C 9x
3628 passwd2des NEW: Secure RPC
3630 pow10f NEW: GNU ext.
3631 pow10l NEW: GNU ext.
3634 printf_size NEW: GNU ext.
3635 printf_size_info NEW: GNU ext.
3636 profil_counter REMOVED
3637 pthread_mutexattr_getkind_np REPLACED
3638 pthread_mutexattr_setkind_np REPLACED
3639 ptsname NEW: Unix98 PTY
3640 ptsname_r NEW: Unix98 PTY
3642 putchar_locked REMOVED
3643 putgrent NEW: GNU ext.
3645 putpmsg NEW: STREAMS
3646 pututxline NEW: Unix98
3650 readdir64_r NEW: LFS
3651 remquo NEW: ISO C 9x
3652 remquof NEW: ISO C 9x
3653 remquol NEW: ISO C 9x
3655 roundf NEW: ISO C 9x
3656 roundl NEW: ISO C 9x
3658 scalbln NEW: ISO C 9x
3659 scalblnf NEW: ISO C 9x
3660 scalblnl NEW: ISO C 9x
3662 sendfile NEW: kernel
3663 setcontext NEW: Unix98
3664 setrlimit64 NEW: LFS
3665 setutxent NEW: Unix98
3667 sigignore NEW: Unix98
3668 sigqueue NEW: POSIX.1b
3669 sigrelse NEW: Unix98
3670 sigset NEW: POSIX.1b
3671 sigtimedwait NEW: POSIX.1b
3672 sigwaitinfo NEW: POSIX.1b
3673 sincos NEW: GNU ext.
3674 sincosf NEW: GNU ext.
3675 sincosl NEW: GNU ext.
3679 strcasestr NEW: GNU ext.
3680 strtoimax NEW: ISO C 9x
3681 strtoumax NEW: ISO C 9x
3682 strverscmp NEW: GNU ext.
3683 svcauthdes_stats NEW: Secure RPC
3684 svcunix_create NEW: sunrpc etx
3685 svcunixfd_create NEW: sunrpc ext
3686 swapcontext NEW: Unix98
3687 tcgetsid NEW: Unix98 PTY
3688 tdestroy NEW: GNU ext.
3689 tgamma NEW: ISO C 9x
3690 tgammaf NEW: ISO C 9x
3691 tgammal NEW: ISO C 9x
3695 truncf NEW: ISO C 9x
3696 truncl NEW: ISO C 9x
3698 unlockpt NEW: Unix98 PTY
3699 updwtmpx NEW: Unix98
3700 user2netname NEW: Secure RPC
3701 utmpxname NEW: Unix98
3702 versionsort NEW: GNU ext.
3703 versionsort64 NEW: GNU ext.
3705 wcscasecmp NEW: GNU ext.
3706 wcsncasecmp NEW: GNU ext.
3707 wcsnlen NEW: GNU ext.
3708 wcstoimax NEW: ISO C 9x
3709 wcstoll NEW: ISO C 9x
3710 wcstoull NEW: ISO C 9x
3711 wcstoumax NEW: ISO C 9x
3713 wordexp NEW: POSIX.2
3714 wordfree NEW: POSIX.2
3715 write_profiling REMOVED
3716 xdecrypt NEW: Secure RPC
3717 xdr_authdes_cred NEW: Secure RPC
3718 xdr_authdes_verf NEW: Secure RPC
3719 xdr_cryptkeyarg NEW: Secure RPC
3720 xdr_cryptkeyarg2 NEW: Secure RPC
3721 xdr_cryptkeyres NEW: Secure RPC
3722 xdr_getcredres NEW: Secure RPC
3723 xdr_key_netstarg NEW: Secure RPC
3724 xdr_key_netstres NEW: Secure RPC
3725 xdr_keybuf NEW: Secure RPC
3726 xdr_keystatus NEW: Secure RPC
3727 xdr_netnamestr NEW: Secure RPC
3728 xdr_sizeof NEW: Secure RPC
3729 xdr_unixcred NEW: sunrpc ext
3730 xencrypt NEW: Secure RPC
3731 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3742 * inet_ntoa is thread-safe
3744 * updwtmp is moved from libutil to libc
3746 * rewrite of cbrt function
3748 * update of timezone data
3762 * add atoll function
3764 * fix complex problems in Berkeley DB code
3766 * fix math functions
3770 * fixed lots of header problems (especially Linux/GNU specific)
3772 * dynamic loader preserves all registers
3774 * Roland McGrath provided support for handling of auxiliary objects in
3775 the ELF dynamic loader.
3777 * support for parallel builds is improved
3781 * GNU extensions are no longer declared by default. To enable them you
3782 must define the macro `_GNU_SOURCE' in your program or compile with
3785 * The library has changed from using GNU ld symbol aliases to using weak
3786 symbols where available. The ELF object file format supports weak
3787 symbols; GNU ld also supports weak symbols in the a.out format. (There
3788 is also now support for other GNU ld extensions in ELF. Use the
3789 `--with-elf' option to configure to indicate you have ELF, and
3790 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3791 of many files which contained only symbol aliases, reducing the size of
3792 the source and the compiled library; many other files were renamed to
3793 less cryptic names previously occupied by the symbol alias files.
3794 There is a new header file <elf.h> for programs which operate on
3795 files in the ELF format.
3797 * Converted to Autoconf version 2, so `configure' has more options.
3798 Run `configure --help' to see the details.
3800 * The library can now be configured to build profiling, highly-optimized
3801 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3802 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3803 `configure' enable building these extra libraries. The shared library is
3804 built by default when using both ELF and GNU ld. When shared libraries
3805 are enabled, the new library `-ldl' is available for arbitrary run-time
3806 loading of shared objects; its interface is defined in <dlfcn.h>. The
3807 new header file <link.h> gives access to the internals of the run-time
3808 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3809 application of same name on other systems and it provides information
3810 about dynamically linked binaries.
3812 * The C library now provides the run-time support code for profiling
3813 executables compiled with `-pg'. Programs can control the profiling code
3814 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3815 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3816 the support for this file format was contributed by David Mosberger-Tang.
3818 * The math code has been replaced with a math library based on fdlibm from
3819 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3820 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3821 functions. The math functions now reside in a separate library, so
3822 programs using them will need to use `-lm' their linking commands.
3824 * John C. Bowman contributed optimized ix87 assembler inline functions.
3826 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3827 mechanism similar to that found in Solaris 2. This is now used for the
3828 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3829 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3830 file controls what services are used for each individual database. This
3831 works by loading shared libraries with names specified in `nsswitch.conf',
3832 so service modules can be changed or added at any time without even
3833 relinking any program. Currently there are the file, db, and NIS based
3834 NSS services available.
3836 * The new functions `strtoq' and `strtouq' parse integer values from
3837 strings, like `strtol' and `strtoul', but they return `long long int' and
3838 `unsigned long long int' values, respectively (64-bit quantities).
3840 * The new functions `strtof' and `strtold' parse floating-point values from
3841 strings, like `strtod', but they return `float' and `long double' values,
3842 respectively (on some machines `double' and `long double' are the same).
3844 * Ulrich Drepper has contributed new implementations of the floating-point
3845 printing and reading code used in the `printf' family of functions and
3846 `strtod', `strtof', and `strtold'. These new functions are perfectly
3847 accurate, and much faster than the old ones.
3849 * The implementation of the POSIX locale model was completely rewritten by
3850 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3851 to compile the POSIX locale definition.
3853 * The former dummy implementations of the strcoll and strxfrm function are
3854 now replaced by fully functional code contributed by Ulrich Drepper. The
3855 collation information comes from the POSIX locale definitions.
3857 * The new header <langinfo.h> defines an interface for accessing
3858 various locale-dependent data (using the locale chosen with `setlocale').
3860 * Ulrich Drepper has contributed a new suite of functions for operation on
3861 wide-character and multibyte-character strings, in <wchar.h>;
3862 and classification and case conversion of wide characters, in <wctype.h>.
3863 These new functions are conforming to the ISO C, Amendement 1 specification.
3865 * There is now a second implementation of the standard I/O library available.
3866 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3867 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3869 * You can now use positional parameter specifications in format strings
3870 for the `printf' and `scanf' families of functions. For example,
3871 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3872 ``Number 6, Mr Jones''. This is mainly useful when providing different
3873 format strings for different languages, whose grammars may dictate
3874 different orderings of the values being printed. To support this
3875 feature, the interface for `register_printf_handler' has changed; see
3876 the header file <printf.h> for details.
3878 * The `printf' and `scanf' families of functions now understand a new
3879 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3880 to group numbers as indicated by the locale; for `scanf' and friends, this
3881 says to accept as valid only a number with all the proper grouping
3882 separators in the right places. In the default "C" locale, numbers are
3883 not grouped; but locales for specific countries will define the usual
3884 conventions (i.e. separate thousands with `,' in the US locale).
3886 * The pgrp functions have been regularized, slightly incompatibly but much
3887 less confusingly. The core functions are now `getpgid' and `setpgid',
3888 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3889 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3890 provided for compatibility. There is no longer an incompatible `getpgrp'
3891 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3893 * The new header file <fts.h> and suite of functions simplify programs that
3894 operate on directory trees. This code comes from 4.4 BSD.
3896 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3897 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3898 NSS scheme used in glibc.
3900 * There is a new malloc debugging hook `__memalign_hook'.
3902 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3903 `unsigned int' in <sys/types.h>. These are for compatibility only and
3904 their use is discouraged.
3906 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3907 done differently, so that it works even without GNU ld.
3909 * New function `euidaccess' checks allowed access to a file like `access',
3910 but using the effective IDs instead of the real IDs.
3912 * The time zone data files have been updated for the latest and greatest
3913 local time conventions of the countries of the world.
3915 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3918 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3919 for formatting floating-point numbers. They are provided only for
3920 compatibility; new programs should use `sprintf' instead. There are
3921 also equivalent function for the `long double' floating-point type and
3922 all functions also exist in a reentrant form.
3924 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3925 functions for maintaining the login-record files (primarily of use to
3926 system programs such as `login'), and convenient functions for
3927 allocating and initializing a pseudo-terminal (pty) device.
3929 * Ulrich Drepper has contributed new support for System V style
3930 shared memory and IPC on systems that support it.
3932 * Ulrich Drepper has contributed several miscellaneous new functions found
3933 in System V: The `hsearch' family of functions provide an effective
3934 implementation of hash tables; `a64l' and `l64a' provide a very simple
3935 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3938 * Ulrich Drepper has contributed new reentrant counterparts for the
3939 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3941 * Ulrich Drepper has contributed new, highly-optimized versions of several
3942 string functions for the i486/Pentium family of processors.
3944 * Ulrich Drepper has updated the Linux-specific code, based largely
3945 on work done in Hongjiu Lu's version of GNU libc for Linux.
3946 The GNU library now supports Linux versions 2.0.10 and later,
3947 using the ELF object file format (i[3456]86-*-linux).
3949 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3951 * David Mosberger-Tang and Richard Henderson have ported the C library
3952 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3953 dynamic linking support for ELF/Alpha.
3955 * Richard Henderson contributed several Alpha optimized assembler function
3956 for arithmetic and string handling.
3958 * Ulrich Drepper has contributed a new set of message catalog functions to
3959 support multiple languages using the <libintl.h> interface, for use with
3960 his new package GNU gettext. Translation volunteers have contributed
3961 catalogs of the library's messages in Spanish, German, and Korean.
3963 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3964 program and the `catgets' function for reading the catalog files it
3965 creates. (The <libintl.h> interface is preferred; we include the
3966 <nl_types.h> interface using `catgets' only for source compatibility with
3967 programs already written to use it.)
3969 * New header file <values.h> gives SVID-compatible names for <limits.h>
3972 * Various new macros, declarations, and small header files for compatibility
3975 * New function `group_member' is a convenient way to check if a process has
3976 a given effective group ID.
3978 * When using GCC 2.7 and later, the socket functions are now declared in a
3979 special way so that passing an argument of type `struct sockaddr_in *',
3980 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3981 `struct sockaddr *' type, does not generate a type-clash warning.
3983 * New function `error' declared in header file <error.h> is a convenient
3984 function for printing error messages and optionally exiting; this is the
3985 canonical function used in GNU programs. The new functions `err', `warn',
3986 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3987 doing the same thing.
3989 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3990 POSIX.2 `glob' function to do ~ and {...} expansion.
3992 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3993 `clearenv' which is used in POSIX.9 is also available.
3995 * New function `getsid' returns session ID number on systems that support it.
3997 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3998 files <db.h> and <mpool.h> provide a rich set of functions for several
3999 types of simple databases stored in memory and in files, and <ndbm.h> is
4000 an old `ndbm'-compatible interface using the `db' functions. Link with
4001 `-ldb' to get these functions.
4003 * New macro `strdupa' copies a string like `strdup', but uses local stack
4004 space from `alloca' instead of dynamic heap space from `malloc'.
4006 * New function `strnlen' is like `strlen' but searches only a given maximum
4007 number of characters for the null terminator. `stpncpy', `strndup' and
4008 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4011 * New function `statfs' in header <sys/statfs.h>.
4013 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4014 provide convenient functions for operating on blocks of null-terminated
4017 * A new suite of functions in <utmp.h> handle all the details of reading
4018 and writing the utmp file.
4020 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4023 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4024 completely POSIX compliant and also implemented the extended functionality
4025 to handle alternate digit representation and alternate era date formats.
4027 * Ulrich Drepper provided an implementation of the `strptime' function
4028 defined in XPG4.2 which transforms a string into a `struct tm' value.
4030 * Paul Eggert provided the tzselect shell script as part of the timezone
4031 code. The shell script makes it easy to select the correct timezone
4034 * The implementation of the malloc family of functions is completely replaced
4035 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4036 The implementation uses the mmap function (if available) and it is
4037 optimized for the use in multi threaded programs.
4039 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4040 functions. This new functionality is usable by specifying a special
4041 salt string and it is compatible with implementation on *BSD systems.
4043 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4044 `getsubopt' to handle second level command line options, `bsd_signal'
4045 to access BSD style `signal' functionality, the obsolete `regexp' style
4048 * the `lchown' function is available on system which support this
4051 * The implementation of the shadow password handling function was contributed
4054 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4056 * POSIX.1g support was added. The <sys/select.h> header is available,
4057 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4058 implementation of `getaddrinfo'.
4062 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4064 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4065 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4067 * The new function `daemon' from 4.4 BSD is useful for server programs that
4068 want to put themselves in the background.
4070 * Joel Sherrill has contributed support for several standalone boards that
4071 run without an operating system.
4073 * `printf', `scanf' and friends now accept a `q' type modifier for long
4074 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4076 * All of the code taken from BSD (notably most of the math and networking
4077 routines) has been updated from the BSD 4.4-Lite release.
4079 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4081 * The new functions `getdomainname' and `setdomainname' fetch or change the
4082 YP/NIS domain name. These are system calls which exist on systems which
4085 * The time zone data files have been updated for the latest international
4088 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4089 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4093 * The C library now includes support for Sun RPC, from Sun's free
4094 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4095 are included. (There is still no support for YP.)
4097 * Tom Quinn has contributed a port of the C library to SGI machines running
4098 Irix 4 (mips-sgi-irix4).
4100 * The new `lockf' function is a simplified interface to the locking
4101 facilities of `fcntl', included for compatibility.
4103 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4105 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4107 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4110 * The error message printed by `assert' for a failed assertion now includes
4111 the name of the program (if using GNU ld) and the name of the calling
4112 function (with versions of GCC that support this).
4114 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4116 * The library now includes the <sys/mman.h> header file and memory
4117 management functions `mmap', `munmap', `mprotect', `msync', and
4118 `madvise', on systems that support those facilities.
4120 * The interface for `mcheck' has changed slightly: the function called to
4121 abort the program when an allocation inconsistency is detected now takes
4122 an argument that indicates the type of failure. The new function
4123 `mprobe' lets you request a consistency check for a particular block at
4124 any time (checks are normally done only when you call `free' or `realloc'
4127 * It is now possible to easily cross-compile the C library, building on one
4128 system a library to run on another machine and/or operating system. All
4129 you need to do is set the variable `HOST_CC' in `configparms' to the
4130 native compiler for programs to run on the machine you are building on (a
4131 few generator programs are used on Unix systems); set `CC' to the
4134 * The new function `fexecve' (only implemented on the GNU system) executes
4135 a program file given a file descriptor already open on the file.
4139 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4140 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4142 * You can set the variable `libprefix' in `configparms' to specify a prefix
4143 to be prepended to installed library files; this makes it easy to install
4144 the GNU C library to be linked as `-lgnuc' or whatever.
4146 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4147 copies a limited number of characters from a string, and returns the
4148 address of the last character written.
4150 * You no longer need to check for whether the installed `stddef.h' is
4151 compatible with the GNU C library. configure now checks for you.
4153 * You can now define a per-stream `fileno' function to convert the
4154 stream's cookie into an integral file descriptor.
4156 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4157 allocates zero bytes of storage, and returns a unique pointer which
4158 you can pass to `realloc' or `free'. The behavior is undefined if
4159 you dereference this pointer.
4161 * The C library now runs on Sony NEWS m68k machines running either
4162 NewsOS 3 or NewsOS 4.
4164 * The new `syscall' function is a system-dependent primitive function
4165 for invoking system calls. It has the canonical behavior on Unix
4166 systems, including unreliable return values for some calls (such as
4167 `pipe', `fork' and `getppid').
4169 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4170 to `EAGAIN', which is the preferred name. On systems whose kernels
4171 use two distinct codes, the C library now translates EWOULDBLOCK to
4172 EAGAIN in every system call function.
4176 * The GNU C Library Reference Manual is now distributed with the library.
4177 `make dvi' will produce a DVI file of the printed manual.
4178 `make info' will produce Info files that you can read on line using C-h i
4179 in Emacs or the `info' program.
4180 Please send comments on the manual to bug-glibc-manual@gnu.org.
4182 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4184 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4186 * Jason Merrill has contributed a port to the Sequent Symmetry running
4187 Dynix version 3 (i386-sequent-dynix).
4189 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4190 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4192 * New function `memory_warnings' lets you arrange to get warnings when
4193 malloc is running out of memory to allocate, like Emacs gives you.
4195 * The C library now contains the relocating allocator used in Emacs 19 for
4196 its editing buffers. This allocator (ralloc) minimizes allocation
4197 overhead and fragmentation by moving allocated regions around whenever it
4198 needs to. You always refer to a ralloc'd region with a "handle" (a
4199 pointer to a pointer--an object of type `void **').
4201 * There is a new `printf' format: `%m' gives you the string corresponding
4202 to the error code in `errno'.
4204 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4205 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4206 pass, the `a' modifier says to fill in a `char **' you pass with a
4209 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4210 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4211 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4213 * `mkstemp' is a traditional Unix function to atomically create and open a
4214 uniquely-named temporary file.
4218 * The standard location for the file that says what the local timezone is
4219 has changed again. It is now `/usr/local/etc/localtime' (or more
4220 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4222 * The distribution no longer contains any files with names longer than 14
4225 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4226 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4228 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4232 * The configuration process has changed quite a bit. The `configure'
4233 script is now used just like the configuration scripts for other GNU
4234 packages. The `sysdeps' directory hierarchy is much rearranged.
4235 The file `INSTALL' explains the new scheme in detail.
4237 * The header files no longer need to be processed into ANSI C and
4238 traditional C versions. There is just one set of files to install, and
4239 it will work with ANSI or old C compilers (including `gcc -traditional').
4241 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4242 MIPS DECStation running Ultrix 4.
4244 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4245 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4246 made itself into a shared library.
4248 * Yet further improved support for the i386, running 4.3 BSD-like systems
4249 (such as Mach 3 with the Unix single-server), or System V.
4251 * New function `strncasecmp' to do case-insensitive string comparison
4252 with limited length.
4254 * New function `strsep' is a reentrant alternative to `strtok'.
4256 * New functions `scandir' and `alphasort' for searching directories.
4258 * New function `setenv' is a better interface to `putenv'.
4260 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4261 function for traversing a directory tree.
4263 * The GNU obstack package is now also part of the C library.
4264 The new function `open_obstack_stream' creates a stdio stream that
4265 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4266 formatted output directly to an obstack.
4268 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4269 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4271 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4273 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4274 things to your strings.
4276 * There are some new test programs: test-fseek, testmb, and testrand.
4278 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4279 These ports are not finished, but are a good starting place for really
4280 supporting those systems.
4282 * `/etc/localtime' is now the standard location for the file that says what
4283 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4284 This follows the general principle that `/etc' is the place for all local
4285 configuration files.
4287 * The C library header files now use `extern "C"' when used by the C++
4288 compiler, so the C library should now work with C++ code.
4290 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4291 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4294 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4295 function which uses the `merge sort' algorithm, and is said to be
4296 significantly faster than the old GNU `qsort' function. Merge sort is
4297 now the standard `qsort' function. The new algorithm can require a lot
4298 of temporary storage; so, the old sorting function is called when the
4299 required storage is not available.
4301 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4302 provides the Unix `crypt' function, plus some other entry points.
4304 * The code and header files taken from 4.4 BSD have been updated with the
4305 latest files released from Berkeley.
4307 ----------------------------------------------------------------------
4308 Copyright information:
4310 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4312 Permission is granted to anyone to make or distribute verbatim copies
4313 of this document as received, in any medium, provided that the
4314 copyright notice and this permission notice are preserved,
4315 thus giving the recipient permission to redistribute in turn.
4317 Permission is granted to distribute modified versions
4318 of this document, or of portions of it,
4319 under the above conditions, provided also that they
4320 carry prominent notices stating who last changed them.
4323 version-control: never