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 <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin and tan
13 with FMA, contributed by Arjan van de Ven and H.J. Lu from Intel.
15 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
17 * Optimized generic expf, exp2f, logf, log2f and powf.
19 * In order to support faster and safer process termination the malloc API
20 family of functions will no longer print a failure address and stack
21 backtrace after detecting heap corruption. The goal is to minimize the
22 amount of work done after corruption is detected and to avoid potential
23 security issues in continued process execution. Reducing shutdown time
24 leads to lower overall process restart latency, so there is benefit both
25 from a security and performance perspective.
27 * The abort function terminates the process immediately, without flushing
28 stdio streams. Previous glibc versions used to flush streams, resulting
29 in deadlocks and further data corruption. This change also affects
30 process aborts as the result of assertion failures.
32 * On platforms where long double has the IEEE binary128 format (aarch64,
33 alpha, mips64, s390 and sparc), the math library now implements _Float128
34 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
35 are the same interfaces added in version 2.26 for some platforms where
36 this format is supported but is not the format of long double.
38 Deprecated and removed features, and other changes affecting compatibility:
40 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
41 defined by <sys/ptrace.h>.
43 * libm no longer supports SVID error handling (calling a user-provided
44 matherr function on error) or the _LIB_VERSION variable to control error
45 handling. (SVID error handling and the _LIB_VERSION variable still work
46 for binaries linked against older versions of the GNU C Library.) The
47 libieee.a library is no longer provided. math.h no longer defines struct
48 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
49 TLOSS, PLOSS and HUGE.
51 * The libm functions pow10, pow10f and pow10l are no longer supported for
52 new programs. Programs should use the standard names exp10, exp10f and
53 exp10l for these functions instead.
55 * The mcontext_t type is no longer the same as struct sigcontext. On
56 platforms where it was previously the same, this changes the C++ name
57 mangling for interfaces involving this type.
59 * The add-ons mechanism for building additional packages at the same time as
60 glibc has been removed. The --enable-add-ons configure option is now
63 Changes to build and runtime requirements:
65 [Add changes to build and runtime requirements here]
67 Security related changes:
69 CVE-2009-5064: The ldd script would sometimes run the program under
70 examination directly, without preventing code execution through the
71 dynamic linker. (The glibc project disputes that this is a security
72 vulnerability; only trusted binaries must be examined using the ldd
75 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
76 suffered from a one-byte overflow during ~ operator processing (either
77 on the stack or the heap, depending on the length of the user name).
78 Reported by Tim Rühsen.
80 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
81 would sometimes fail to free memory allocated during ~ operator
82 processing, leading to a memory leak and, potentially, to a denial
85 The glob function, when invoked with GLOB_TILDE and without
86 GLOB_NOESCAPE, could write past the end of a buffer while
87 unescaping user names. Reported by Tim Rühsen.
89 The following bugs are resolved with this release:
91 [The release manager will add the list generated by
92 scripts/list-fixed-bugs.py just before the release.]
99 * A per-thread cache has been added to malloc. Access to the cache requires
100 no locks and therefore significantly accelerates the fast path to allocate
101 and free small amounts of memory. Refilling an empty cache requires locking
102 the underlying arena. Performance measurements show significant gains in a
103 wide variety of user workloads. Workloads were captured using a special
104 instrumented malloc and analyzed with a malloc simulator. Contributed by
105 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
107 * Unicode 10.0.0 Support: Character encoding, character type info, and
108 transliteration tables are all updated to Unicode 10.0.0, using
109 generator scripts contributed by Mike FABIAN (Red Hat).
110 These updates cause user visible changes, especially the changes in
111 wcwidth for many emoji characters cause problems when emoji sequences
112 are rendered with pango, see for example:
113 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
115 * Collation of Hungarian has been overhauled and is now consistent with "The
116 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
119 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
121 - The GNU C Library will now detect when /etc/resolv.conf has been
122 modified and reload the changed configuration. The new resolver option
123 “no-reload” (RES_NORELOAD) disables this behavior.
125 - The GNU C Library now supports an arbitrary number of search domains
126 (configured using the “search” directive in /etc/resolv.conf);
127 previously, there was a hard limit of six domains. For backward
128 compatibility, applications that directly modify the ‘_res’ global
129 object are still limited to six search domains.
131 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
132 Library will now randomly pick a name server from the configuration as a
133 starting point. (Previously, the second name server was always used.)
135 * The tunables feature is now enabled by default. This allows users to tweak
136 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
138 * New function reallocarray, which resizes an allocated block (like realloc)
139 to the product of two sizes, with a guaranteed clean failure upon integer
140 overflow in the multiplication. Originally from OpenBSD, contributed by
141 Dennis Wölfing and Rüdiger Sonderfeld.
143 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
144 These are extended versions of preadv and pwritev, respectively, taking an
145 additional flags argument. The set of supported flags depends on the
146 running kernel; full support currently requires kernel 4.7 or later.
148 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
149 create a new session ID for the spawned process. This feature is
150 scheduled to be added to the next major revision of POSIX; for the time
151 being, it is available under _GNU_SOURCE.
153 * errno.h is now safe to use from C-preprocessed assembly language on all
154 supported operating systems. In this context, it will only define the
155 Exxxx constants, as preprocessor macros expanding to integer literals.
157 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
158 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
159 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
160 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
162 To compile programs that use this feature, the compiler must support
163 128-bit floating point with the type name _Float128 (as defined by TS
164 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
165 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
166 must be defined to make the new interfaces visible.
168 The new functions and macros correspond to those present for other
169 floating-point types (except for a few obsolescent interfaces not
170 supported for the new type), with F128 or f128 suffixes; for example,
171 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
172 printf or scanf formats for the new type; the strfromf128 and strtof128
173 interfaces should be used instead.
175 Deprecated and removed features, and other changes affecting compatibility:
177 * The synchronization that pthread_spin_unlock performs has been changed to
178 now be equivalent to a C11 atomic store with release memory order to the
179 spin lock's memory location. Previously, several (but not all)
180 architectures used stronger synchronization (e.g., containing what is
181 often called a full barrier). This change can improve performance, but
182 may affect odd fringe uses of spin locks that depend on the previous
183 behavior (e.g., using spin locks as atomic variables to try to implement
184 Dekker's mutual exclusion algorithm).
186 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
189 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
190 will only be built and installed when the GNU C Library is configured with
191 --enable-obsolete-rpc. This allows alternative RPC implementations, such
192 as TIRPC or rpcsvc-proto, to be used.
194 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
195 libnss_compat, are deprecated, and will not be built or installed by
198 The NIS(+) support library, libnsl, is also deprecated. By default, a
199 compatibility shared library will be built and installed, but not headers
200 or development libraries. Only a few NIS-related programs require this
201 library. (In particular, the GNU C Library has never required programs
202 that use 'gethostbyname' to be linked with libnsl.)
204 Replacement implementations based on TIRPC, which additionally support
205 IPv6, are available from <https://github.com/thkukuk/>. The configure
206 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
207 name service modules, to be built and installed.
209 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
210 support is enabled, the configured recursive resolver must support EDNS.
211 (Responding to EDNS-enabled queries with responses which are not
212 EDNS-enabled is fine, but FORMERR responses are not.)
214 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
215 servers have not supported this opcode for a long time.
217 * The _res_opcodes variable has been removed from libresolv. It had been
218 exported by accident.
220 * <string.h> no longer includes inline versions of any string functions,
221 as this kind of optimization is better done by the compiler. The macros
222 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
224 * The nonstandard header <xlocale.h> has been removed. Most programs should
225 use <locale.h> instead. If you have a specific need for the definition of
226 locale_t with no other declarations, please contact
227 libc-alpha@sourceware.org and explain.
229 * The obsolete header <sys/ultrasound.h> has been removed.
231 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
233 * The obsolete function cfree has been removed. Applications should use
236 * The stack_t type no longer has the name struct sigaltstack. This changes
237 the C++ name mangling for interfaces involving this type.
239 * The ucontext_t type no longer has the name struct ucontext. This changes
240 the C++ name mangling for interfaces involving this type.
242 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
243 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
244 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
245 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
246 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
247 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
248 fpregset_t type no longer has the name struct fpu. This changes the C++
249 name mangling for interfaces involving those types.
251 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
252 synced with the kernel:
254 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
255 are not supported on this architecture and have been removed.
257 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
258 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
259 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
261 Programs that assume the GET/SETREGS ptrace requests are universally
262 available will now fail to build, instead of malfunctioning at runtime.
264 Changes to build and runtime requirements:
266 * Linux kernel 3.2 or later is required at runtime, on all architectures
267 supported by that kernel. (This is a change from version 2.25 only for
270 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
272 * On most architectures, GCC 4.9 or later is required to build the GNU C
273 Library. On powerpc64le, GCC 6.2 or later is required.
275 Older GCC versions and non-GNU compilers are still supported when
276 compiling programs that use the GNU C Library. (We do not know exactly
277 how old, and some GNU extensions to C may be _de facto_ required. If you
278 are interested in helping us make this statement less vague, please
279 contact libc-alpha@sourceware.org.)
281 Security related changes:
283 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
284 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
286 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
287 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
289 * Avoid printing a backtrace from the __stack_chk_fail function since it is
290 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
293 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
294 fixed (CVE-2017-12133).
296 The following bugs are resolved with this release:
298 [984] network: Respond to changed resolv.conf in gethostbyname
299 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
300 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
302 [12189] libc: __stack_chk_fail should not attempt a backtrace
304 [14096] time: Race condition on timezone/tst-timezone.out
305 [14172] localedata: az_IR: new locale
306 [14995] build: glibc fails to build if gold is the default linker, even if
308 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
309 [16637] network: inet_pton function is accepting IPv6 with bad format
310 [16640] string: string/strtok.c: undefined behaviour inconsistent between
311 x86 and other generic code
312 [16875] localedata: ko_KR: fix lang_name
313 [17225] localedata: ar_SY: localized month names for May and June are
315 [17297] localedata: da_DK: wrong date_fmt string
316 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
317 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
318 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
319 [19066] localedata: ar_SA abbreviated day and month names are in English
320 [19569] network: resolv: Support an arbitrary number of search domains
321 [19570] network: Implement random DNS server selection in the stub
323 [19838] locale: localedef fails on PA-RISC
324 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
325 order of 0D36 and 0D37
326 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
328 [20098] libc: FAIL: debug/backtrace-tst on hppa
329 [20257] network: sunrpc: clntudp_call does not enforce timeout when
331 [20275] localedata: locale day/abday/mon/abmon should not have trailing
333 [20313] localedata: Update locale data to Unicode 9.0
334 [20424] manual: Document how to provide a malloc replacement
335 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
337 [20686] locale: Add el_GR@euro to SUPPORTED.
338 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
339 failures consistently
340 [21015] dynamic-link: Document and fix --enable-bind-now
341 [21016] nptl: pthread_cond support is broken on hppa
342 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
343 frame-pointer on i386
344 [21049] libc: segfault in longjmp_chk() due to clobbered processor
346 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
347 [21088] libc: Build fails with --enable-static-nss
348 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
349 generic c code is used
350 [21109] libc: Tunables broken on big-endian
351 [21112] math: powf has large ulp errors with base close to 1 and exponent
353 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
355 [21120] malloc: glibc malloc is incompatible with GCC 7
356 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
358 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
359 (-0.0) when linking with -lieee
360 [21171] math: log10, log2 and lgamma return incorrect results
361 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
362 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
363 [21207] localedata: ce_RU: update weekdays from CLDR
364 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
365 [21217] localedata: Update months from CLDR-31
366 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
367 [21243] libc: support_delete_temp_file should issue warning for failed
369 [21244] libc: support resolv_test_start() socket fd close should be
371 [21253] libc: localedef randomly segfaults when using -fstack-check due to
372 new posix_spawn implementation
373 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
374 leads to lower CPU frequency
375 [21259] libc: [alpha] termios.h missing IXANY for POSIX
376 [21261] libc: [sparc64] bits/setjmp.h namespace
377 [21267] network: [mips] bits/socket.h IOC* namespace
378 [21268] libc: [alpha] termios.h NL2, NL3 namespace
379 [21270] libc: mmap64 silently truncates large offset values
380 [21275] libc: posix_spawn always crashes on ia64 now
381 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
382 [21280] math: [powerpc] logbl for POWER7 return incorrect results
383 [21289] libc: Incorrect declaration for 32-bit platforms with
384 _FILE_OFFSET_BITS=64 causes build error
385 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
386 not support gethostbyname4_r
387 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
389 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
391 [21340] libc: Support POSIX_SPAWN_SETSID
392 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
393 [21359] network: ns_name_pack needs additional byte in destination buffer
394 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
395 against fragmentation attacks (CVE-2017-12132)
396 [21369] network: resolv: Remove EDNS fallback
397 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
399 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
400 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
401 [21393] stdio: Missing dup3 error check in freopen, freopen64
402 [21396] libc: Use AVX2 memcpy/memset on Skylake server
403 [21399] localedata: Bad description for U00EC in
404 localedata/charmaps/CP1254
405 [21411] malloc: realloc documentation error
406 [21426] network: sys/socket.h uio.h namespace
407 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
408 [21445] libc: signal.h bsd_signal namespace
409 [21455] network: Network headers stdint.h namespace
410 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
412 [21475] network: resolv: Overlong search path is truncated mid-label
413 [21511] libc: sigstack namespace
414 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
415 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
416 bits/syscall.h: No such file or directory
417 [21517] libc: struct sigaltstack namespace
418 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
419 [21533] localedata: Update locale data to Unicode 10.0
421 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
422 junk at end of line, first unrecognized character is `@'
423 [21538] libc: SIG_HOLD missing for XPG4
424 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
425 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
426 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
427 [21543] libc: sigevent namespace
428 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
429 size for general purpose registers in mcontext_t structure
430 [21550] libc: sigwait namespace
431 [21552] libc: XPG4 bsd_signal namespace
432 [21554] libc: sigpause namespace
433 [21560] libc: sys/wait.h signal.h namespace
434 [21561] libc: waitid namespace
435 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
436 '/usr/include/stdlib.h': Permission denied
437 [21575] libc: sys/wait.h missing struct rusage definition
438 [21584] libc: sigaltstack etc namespace
439 [21597] libc: siginterrupt namespace
440 [21607] math: hppa: FAIL: math/test-tgmath
441 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
443 [21622] libc: [tile] missing SA_* for POSIX.1:2008
444 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
445 stack and heap (CVE-2017-1000366)
446 [21625] libc: wait3 namespace
447 [21654] nss: Incorrect pointer alignment in NSS group merge result
449 [21657] network: Parse interface zone id for node-local multicast
450 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
452 [21666] libc: .symver is used on common symbol
453 [21668] network: resolv: res_init cross-thread broadcast introduces race
455 [21687] math: tgmath.h totalorder, totalordermag return type
456 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
458 [21696] libc: Incorrect assumption of of __cpu_mask in
459 posix/sched_cpucount.c
460 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
461 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
463 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
464 leading to relocation crash
465 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
466 `resolv_conf_matches (resp, conf)' failed.
467 [21710] localedata: Added Samoan language locale for Samoa
468 [21711] localedata: Pashto yesstr/nostr locale are missing
469 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
470 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
471 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
472 [21723] localedata: yesstr/nostr missing for Chinese language locale
473 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
475 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
477 [21728] localedata: New Locale for Tongan language
478 [21729] localedata: incorrect LC_NAME fields for hi_IN
479 [21733] localedata: yesstr and nostr are missing for zh_HK
480 [21734] localedata: Missing yesstr and nostr are for kw_GB
481 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
482 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
483 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
484 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
486 [21744] libc: Tests failing on --enable-tunables --enable-stack-
488 [21749] localedata: Wrong abbreviated day name (“abday”) for
490 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
491 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
492 [21759] localedata: missing yesstr and nostr for Tigrinya
493 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
494 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
495 [21767] localedata: Missing Bislama locales
496 [21768] localedata: Missing yesstr and nostr for aa_ET
497 [21770] localedata: Missing Field in li_NL
498 [21778] nptl: Robust mutex may deadlock
499 [21779] libc: MicroBlaze segfaults when loading libpthread
500 [21783] localedata: Fix int_select international_call_prefixes
501 [21784] localedata: Inconsistency in country_isbn
502 [21788] localedata: Missing Country Postal Abbreviations
503 [21794] localedata: Added-country_isbn-for-Italy
504 [21795] localedata: Add/Fix country_isbn for France
505 [21796] localedata: Added country_isbn for Republic of Korea
506 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
507 [21799] localedata: Added int_select international_call_prefixes
508 [21801] localedata: Added int_select international_call_prefixes
509 [21804] nptl: Double semicolon in thread-shared-types.h
510 [21807] localedata: LC_ADDRESS fix for pap_CW
511 [21808] localedata: Fix LC_ADDRESS for pap_AW
512 [21821] localedata: Added country_name in mai_IN
513 [21822] localedata: Fix LC_TIME for mai_IN
514 [21823] localedata: missing yesstr, nostr for sa_IN
515 [21825] localedata: Fix name_mrs for mag_IN
516 [21828] localedata: 2.26 changelog should mention user visible changes
518 [21835] localedata: Added Maithili language locale for Nepal
519 [21838] localedata: Removed redundant data for the_NP
520 [21839] localedata: Fix LC_MONETARY for ta_LK
521 [21844] localedata: Fix Latin characters and Months Sequence.
522 [21848] localedata: Fix mai_NP Title Name
527 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
528 24731-2:2010, is supported to enable declarations of functions from that
529 TR. Note that not all functions from that TR are supported by the GNU C
532 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
533 18661-1:2014, is supported to enable declarations of functions and macros
534 from that TS. Note that not all features from that TS are supported by
537 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
538 18661-4:2015, is supported to enable declarations of functions and macros
539 from that TS. Note that most features from that TS are not supported by
542 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
543 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
544 Since the GNU C Library defaults to a much newer revision of POSIX, this
545 will only affect programs that specifically request an old conformance
546 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
547 see a change in the visible declarations, but a program compiled with
548 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
551 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
552 defined by all multithreaded code, but glibc has not required this for
555 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
556 means that in a future release, the macros “major”, “minor”, and “makedev”
557 will only be available from <sys/sysmacros.h>.
559 These macros are not part of POSIX nor XSI, and their names frequently
560 collide with user code; see for instance glibc bug 19239 and Red Hat bug
561 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
562 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
565 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
566 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
567 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
569 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
570 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
571 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
572 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
573 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
574 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
575 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
576 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
577 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
578 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
579 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
582 * New <math.h> features are added from TS 18661-1:2014:
584 - Signaling NaN macros: SNANF, SNAN, SNANL.
586 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
587 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
588 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
590 - llogb functions: the llogb, llogbf and llogbl functions, and the
591 FP_LLOGB0 and FP_LLOGBNAN macros.
593 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
596 - Comparison macros: iseqsig.
598 - Classification macros: iscanonical, issubnormal, iszero.
600 - Total order functions: totalorder, totalorderf, totalorderl,
601 totalordermag, totalordermagf, totalordermagl.
603 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
605 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
606 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
608 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
609 are added to libc. They convert a floating-point number into string.
611 * Most of glibc can now be built with the stack smashing protector enabled.
612 It is recommended to build glibc with --enable-stack-protector=strong.
613 Implemented by Nick Alcock (Oracle).
615 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
616 intended to be used instead of memset() to erase sensitive data after use;
617 the compiler will not optimize out calls to explicit_bzero even if they
618 are "unnecessary" (in the sense that no _correct_ program can observe the
619 effects of the memory clear).
621 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
622 to float instead of double. This does not affect the ABI of any libraries
623 that are part of the GNU C Library, but may affect the ABI of other
624 libraries that use this type in their interfaces.
626 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
627 float_t and double_t types are now defined to long double instead of float
628 and double. These options are not the default, and this does not affect
629 the ABI of any libraries that are part of the GNU C Library, but it may
630 affect the ABI of other libraries that use this type in their interfaces,
631 if they are compiled or used with those options.
633 * The getentropy and getrandom functions, and the <sys/random.h> header file
636 * The buffer size for byte-oriented stdio streams is now limited to 8192
637 bytes by default. Previously, on Linux, the default buffer size on most
638 file systems was 4096 bytes (and thus remains unchanged), except on
639 network file systems, where the buffer size was unpredictable and could be
640 as large as several megabytes.
642 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
643 for the Linux quota interface which predates kernel version 2.4.22 has
646 * The malloc_get_state and malloc_set_state functions have been removed.
647 Already-existing binaries that dynamically link to these functions will
648 get a hidden implementation in which malloc_get_state is a stub. As far
649 as we know, these functions are used only by GNU Emacs and this change
650 will not adversely affect already-built Emacs executables. Any undumped
651 Emacs executables, which normally exist only during an Emacs build, should
652 be rebuilt by re-running “./configure; make” in the Emacs build tree.
654 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
655 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
656 “no-ip6-dotint” had already been the default, and support for the
657 “ip6-dotint” option was removed from the Internet in 2006.
659 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
660 flag from <resolv.h> have been removed. The option relied on a
661 backwards-incompatible DNS extension which was never deployed on the
664 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
665 RES_BLAST defined in the <resolv.h> header file have been deprecated.
666 They were already unimplemented.
668 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
669 _res.flags are deprecated. The flag was standardized in RFC 2133, but
670 removed again from the IETF name lookup interface specification in RFC
671 2553. Applications should use getaddrinfo instead.
673 * DNSSEC-related declarations and definitions have been removed from the
674 <arpa/nameser.h> header file, and libresolv will no longer attempt to
675 decode the data part of DNSSEC record types. Previous versions of glibc
676 only implemented minimal support for the previous version of DNSSEC, which
677 is incompatible with the currently deployed version.
679 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
680 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
681 <arpa/nameser.h> header file because the distinction between RR types and
682 meta-RR types is not officially standardized, subject to revision, and
683 thus not suitable for encoding in a macro.
685 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
686 and rhook members of the res_state type in <resolv.h> have been removed.
687 The glibc stub resolver did not support these hooks, but the header file
688 did not reflect that.
690 * For multi-arch support it is recommended to use a GCC which has
691 been built with support for GNU indirect functions. This ensures
692 that correct debugging information is generated for functions
693 selected by IFUNC resolvers. This support can either be enabled by
694 configuring GCC with '--enable-gnu-indirect-function', or by
695 enabling it by default by setting 'default_gnu_indirect_function'
696 variable for a particular architecture in the GCC source file
699 * GDB pretty printers have been added for mutex and condition variable
700 structures in POSIX Threads. When installed and loaded in gdb these pretty
701 printers show various pthread variables in human-readable form when read
702 using the 'print' or 'display' commands in gdb.
704 * Tunables feature added to allow tweaking of the runtime for an application
705 program. This feature can be enabled with the '--enable-tunables' configure
706 flag. The GNU C Library manual has details on usage and README.tunables has
707 instructions on adding new tunables to the library.
709 * A new version of condition variables functions have been implemented in
710 the NPTL implementation of POSIX Threads to provide stronger ordering
713 * A new version of pthread_rwlock functions have been implemented to use a more
714 scalable algorithm primarily through not using a critical section anymore to
717 Security related changes:
719 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
720 have been created with makecontext could fail to terminate due to a
721 missing .cantunwind annotation. This has been observed to lead to a hang
722 (denial of service) in some Go applications compiled with gccgo. Reported
723 by Andreas Schwab. (CVE-2016-6323)
725 * The DNS stub resolver functions would crash due to a NULL pointer
726 dereference when processing a query with a valid DNS question type which
727 was used internally in the implementation. The stub resolver now uses a
728 question type which is outside the range of valid question type values.
731 The following bugs are resolved with this release:
733 [4099] stdio: Overly agressive caching by stream i/o functions.
734 [7065] build: Support building glibc with -fstack-protector or -fstack-
736 [9842] localedata: en_CA: incorrect date format
737 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
738 before it started waiting
739 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
740 [15765] nptl: sem_open is wrongly a cancellation point
741 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
742 [16458] libc: endian.h and netinet/in.h byte order macros return values of
744 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
745 library linked with pthread
746 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
748 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
749 [17252] libc: getrandom and getentropy syscall
750 [17730] malloc: thread-local storage is sometimes improperly free()'d
751 after being __libc_memalign()'d
752 [18241] stdio: failed fseek on memstream does not set errno and can fail
754 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
756 [18463] nptl: pthread_cond_broadcast issue when surrounded by
757 PTHREAD_PRIO_INHERIT mutex on ARM
758 [18784] network: res_query and related function crash for special record
759 type queries (CVE-2015-5180)
760 [19380] math: strtod does not raise "inexact"
761 [19387] string: Integer overflow in memchr
762 [19390] string: Integer overflow in strncat
763 [19398] build: linknamespace tests fail with massively parallel build
764 [19402] nptl: Deadlock with robust shared mutex and asynchronous
766 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
767 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
768 [19514] libc: [PATCH] Fix spelling errors in spelling
769 "implement"/"implementation" in several places
770 [19582] network: Deprecate RES_USE_INET6
771 [19673] manual: clog10 docs appear to be erroneous
772 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
773 causes a segmentation fault
774 [19826] libc: invalid pointer returned from __tls_get_addr with static
776 [20016] network: resolv: Remove hooks support from the API
777 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
779 [20033] math: [x86_64] vectorized math function don't call the __finite
781 [20116] nptl: use after free in pthread_create
782 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
783 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
785 [20311] nptl: please install proc_service.h
786 [20366] libc: Compilation errors in installed headers in strict-compliance
788 [20370] malloc: malloc: Arena free list management is still racy
789 (incorrect fix in bug 19243)
790 [20386] libc: assert (X = 0) does not result in GCC warning
791 [20432] malloc: malloc: Minimize interface required for interposition
792 [20435] libc: Missing unwind info in __startcontext causes infinite loop
793 in _Unwind_Backtrace (CVE-2016-6323)
794 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
795 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
797 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
798 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
799 [20477] network: resolv: incorrect double-checked locking related to
801 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
803 [20495] math: x86_64 performance degradation due to AVX/SSE transition
805 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
806 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
807 cause transition penalty
808 [20517] math: sparcv9 missing fdiml compat symbol
809 [20524] manual: strverscmp is inconsistent
810 [20525] libc: <sys/quota.h> should be based on kernel headers
811 [20539] math: GCC 7 -static -lm fails to link at -O3
812 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
814 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
816 [20591] network: Remove obsolete DNSSEC support
817 [20592] network: DNS resource record type classification macros in
818 <arpa/nameser.h> are incorrect
819 [20593] network: Update DNS RR type definitions
820 [20611] network: getaddrinfo accepts invalid numeric scope IDs
821 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
823 [20629] network: libresolv: Remove support for bitstring labels
825 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
826 assertion: "maxidx >= 2"
827 [20660] math: [arm] Use VSQRT
828 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
829 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
830 [20689] libc: Test for FMA should also check for AVX.
831 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
832 [20715] math: iszero macro breaks existing code
833 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
834 [20728] libc: powerpc: Missing TOC stub in clone
835 [20729] build: glibc-2.24 fails to build for i486 with -Os
836 [20750] build: Build fails with default PIE enabled gcc-6.x
837 [20768] math: [alpha] sqrt fegetenv namespace
838 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
840 [20787] math: float_t is defined as float incorrectly on x86_64 even if
841 __FLT_EVAL_METHOD__ is 2
842 [20790] network: rpcgen buffer overrun in get_prog_declaration
843 [20822] nptl: race condition in __lll_unlock_elision on powerpc
844 [20829] libc: crypt snprintf namespace
845 [20847] libc: tst-vfork3 failure
846 [20855] math: Default bits/mathdef.h has inappropriate float_t
847 [20859] math: [sh4] FP_ILOGB0 invalid
848 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
849 U+20AC), not same as GBK
850 [20915] dynamic-link: global-dynamic TLS broken on aarch64
851 [20916] math: pow handling of sNaN arguments
852 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
853 [20919] math: Bad pow (qNaN, 0) result with -lieee
854 [20940] math: hypot sNaN handling
855 [20947] math: fmax, fmin sNaN handling
856 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
857 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
859 [20971] string: powerpc64/power7 memchr overflows internal pointer check
860 [20973] nptl: robust mutexes: Lost wake-ups
861 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
862 [20978] nis: strlen on null pointer in nss_nisplus
863 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
865 [21014] string: i686 memchr overflows internal pointer check
866 [21019] libc: [mips] n32 lseek incorrect on overflow
867 [21022] libc: [microblaze] __backtrace get_frame_size namespace
868 [21026] libc: [MIPS] readahead syscall is broken on n64
869 [21028] math: Fallback fesetexceptflag should always succeed
870 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
871 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
873 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
874 [21061] librt: [microblaze] librt lost clock_* exports
875 [21073] libc: tunables: insecure environment variables passed to
876 subprocesses with AT_SECURE
877 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
881 * The minimum Linux kernel version that this version of the GNU C Library
882 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
883 kernel version 2.6.32 or later suffices (on architectures that already
884 required kernel versions more recent than 3.2, those requirements remain
885 unchanged). Linux 3.2 or later kernel headers are required on all
888 * The pap_AN locale has been deleted. This has been deprecated for a long
889 time. It has been replaced by pap_AW & pap_CW, both of which have long
890 been included in previous releases.
892 * The readdir_r and readdir64_r functions have been deprecated. It is
893 recommended to use readdir and readdir64 instead.
895 * The type “union wait” has been removed. It was deprecated in the early
896 1990s and never part of POSIX. Application code should use the int type
897 instead of “union wait”.
899 * A new NSS action is added to facilitate large distributed system
900 administration. The action, MERGE, allows remote user stores like LDAP
901 to be merged into local user stores like /etc/groups in order to provide
902 easy to use, updated, and managed sets of merged credentials. The new
903 action can be used by configuring it in /etc/nsswitch.conf:
904 group: files [SUCCESS=merge] nis
905 Implemented by Stephen Gallagher (Red Hat).
907 * The deprecated __malloc_initialize_hook variable has been removed from the
910 * The long unused localedef --old-style option has been removed. It hasn't
911 done anything in over 16 years. Scripts using this option can safely
914 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
915 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
916 return the next representable value in the direction of positive infinity
917 and the nextdown functions return the next representable value in the
918 direction of negative infinity. These are currently enabled as GNU
921 Security related changes:
923 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
924 could result in a stack overflow when getnetbyname was called with an
925 overly long name. (CVE-2016-3075)
927 * Previously, getaddrinfo copied large amounts of address data to the stack,
928 even after the fix for CVE-2013-4458 has been applied, potentially
929 resulting in a stack overflow. getaddrinfo now uses a heap allocation
930 instead. Reported by Michael Petlan. (CVE-2016-3706)
932 * The glob function suffered from a stack-based buffer overflow when it was
933 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
934 Reported by Alexander Cherepanov. (CVE-2016-1234)
936 * The Sun RPC UDP client could exhaust all available stack space when
937 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
938 alloca plugin for GCC. (CVE-2016-4429)
940 * The IPv6 name server management code in libresolv could result in a memory
941 leak for each thread which is created, performs a failing naming lookup,
942 and exits. Over time, this could result in a denial of service due to
943 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
945 The following bugs are resolved with this release:
947 [1170] localedata: ne_NP: update Nepali locale definition file
948 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
950 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
951 [6796] math: fdim() does not set errno on overflow
952 [10354] libc: posix_spawn should use vfork() in more cases than presently
953 [11213] localedata: localedata: add copyright disclaimer to locale files
954 [12143] localedata: chr_US: new Cherokee locale
955 [12450] localedata: sgs_LT: new locale
956 [12676] localedata: ln_CD: new locale
957 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
959 [13304] math: fma, fmaf, fmal produce wrong results
960 [14259] build: --localedir arg to configure is ignored
961 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
962 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
963 [14934] localedata: es_CL: wrong first weekday chilean locale
964 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
966 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
968 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
969 [15368] nptl: raise() is not async-signal-safe
970 [15479] math: ceil, floor, round and trunc raise inexact exception
971 [15578] localedata: kk_KZ: various updates
972 [16003] localedata: pap_AN: punt old locale
973 [16137] localedata: iw_IL: punt old locale
974 [16190] localedata: eo: new esperanto locale
975 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
976 [16742] malloc: race condition: pthread_atfork() called before first
977 malloc() results in unexpected locking behaviour/deadlocks
978 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
980 [16983] localedata: postal_fmt does not allow %l and %n modifiers
981 [17565] localedata: pt_PT: wrong (work-)week start
982 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
983 [17950] build: Build fails with -msse
984 [18205] localedata: be_BY*: wrong first_weekday and first_workday
985 [18433] libc: posix_spawn does not return correctly upon failure to
987 [18453] localedata: charmaps/IBM875: incorrect codes
988 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
990 [18896] localedata: he_IL: improvements for currency
991 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
993 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
995 [19056] libc: Deprecate readdir_r
996 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
998 [19198] localedata: nl_NL: small improvements for Dutch locales
999 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1000 nameservers (CVE-2016-5417)
1001 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1003 [19400] locale: Language missing in "iso-639.def", trivial fix in
1005 [19431] malloc: Deadlock between fflush, getdelim, and fork
1006 [19505] libc: Incorrect file descriptor validity checks in
1007 posix_spawn_file_actions_add{open,close,dup2}
1008 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1009 when using RTLD_NEXT
1010 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1011 `intl/{gettextP,loadinfo}.h'
1012 [19534] libc: execle, execlp may use malloc
1013 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1014 [19573] network: res_nclose and __res_maybe_init disagree about name
1015 server initialization, breaking Hesiod
1016 [19575] localedata: Status of GB18030 tables
1017 [19581] localedata: sr_* date_fmt string contains additional newline
1018 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1020 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1021 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1022 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1023 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1025 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1027 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1029 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1030 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1031 [19633] locale: strfmon_l applies global locale to number formatting
1032 [19642] network: Memory leak in getnameinfo
1033 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1034 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1036 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1037 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1039 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1040 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1042 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1044 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1046 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1047 pointers and lengths in error-case.
1048 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1049 report errors on UTF-16 surrogates values.
1050 [19755] nscd: nscd assertion failure in gc
1051 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1052 [19759] libc: mempcpy shouldn't be inlined
1053 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1054 [19765] libc: s390 needs an optimized mempcpy
1055 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1056 NAME_MAX limit assumption (CVE-2016-1234)
1057 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1058 [19787] network: Missing and incorrect truncation checks in getnameinfo
1059 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1061 [19791] network: Assertion failure in res_query.c with un-connectable name
1063 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1064 [19822] math: libm.so install clobbers old version
1065 [19825] network: resolv: send_vc can return uninitialized data in second
1066 response to getaddrinfo
1067 [19830] network: nss_dns: should check RDATA length against buffer length
1068 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1069 confronted with A/AAAA records of invalid size
1070 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1071 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1072 [19853] stdio: Printing IBM long double in decimal with high precision is
1074 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1075 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1076 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1078 [19865] network: Assertion failure or memory leak in
1079 _nss_dns_getcanonname_r
1080 [19868] network: nss_dns: netent code does not skip over non-PTR records
1081 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1083 [19881] string: Improve x86-64 memset
1084 [19907] string: Incorrect memcpy tests
1085 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1087 [19925] libc: termios.h XCASE namespace
1088 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1089 [19929] libc: limits.h NL_NMAX namespace
1090 [19931] stdio: Memory leak in vfprintf
1091 [19957] libc: clone(CLONE_VM) access invalid parent memory
1092 [19963] localedata: en_IL: New locale
1093 [19989] stdio: stdio.h cuserid namespace
1094 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1096 [19996] locale: langinfo.h nl_langinfo_l namespace
1097 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1098 [20010] network: getaddrinfo: Stack overflow in hostent translation
1100 [20012] stdio: libio: fmemopen append mode failure
1101 [20014] stdio: stdio.h namespace for pre-threads POSIX
1102 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1103 [20023] libc: fcntl.h timespec namespace
1104 [20024] math: [x86_64] vectorized sincos trashes the stack
1105 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1106 [20041] time: sys/time.h timespec namespace
1107 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1108 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1109 [20051] libc: ttyslot in wrong header under wrong conditions
1110 [20054] libc: gethostname not declared for XPG4
1111 [20055] libc: termios.h missing tcgetsid for XPG4
1112 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1114 [20073] libc: sys/stat.h fchmod namespace
1115 [20074] libc: stdlib.h rand_r namespace
1116 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1117 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1119 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1120 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1122 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1123 [20119] libc: Wrong mask for processors level type from CPUID
1124 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1125 AS not supporting AVX512
1126 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1128 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1129 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1131 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1132 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1133 [20178] libc: posix_spawn{p} should not call exit
1134 [20191] stdio: libio: vtables hardening
1135 [20195] string: FMA4 detection requires CPUID execution with register
1137 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1138 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1140 [20212] math: acos (sNaN) returns sNaN
1141 [20213] math: asin (sNaN) returns sNaN
1142 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1143 [20218] math: [i386] asinhl (sNaN) returns sNaN
1144 [20219] math: [i386] atanhl (sNaN) returns sNaN
1145 [20222] stdio: fopencookie: Mangle function pointers
1146 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1147 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1148 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1150 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1151 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1152 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1153 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1154 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1155 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1156 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1157 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1158 [20240] math: modf (sNaN) returns sNaN
1159 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1160 [20250] math: frexp (sNaN) returns sNaN
1161 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1162 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1164 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1165 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1166 subscript is below array bounds [-Werror=array-bounds]
1167 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1168 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1170 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1171 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1173 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1174 "invalid" exceptions
1175 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1176 [20316] localedata: id_ID: Februari instead of Pebruari
1177 [20327] string: POWER8 strcasecmp returns incorrect result
1178 [20347] math: Failure: Test: j0_downward (0xap+0)
1179 [20348] libc: FAIL: misc/tst-preadvwritev64
1180 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1181 [20350] libc: There is no test for p{read,write}64
1182 [20357] math: Incorrect cos result for 1.5174239687223976
1183 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1187 * Unicode 8.0.0 Support: Character encoding, character type info, and
1188 transliteration tables are all updated to Unicode 8.0.0, using new
1189 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1190 These updates cause user visible changes, such as the fixes for bugs
1191 89, 16061, and 18568.
1193 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1194 kernel-internal CPU set size. This means that requests that change the
1195 CPU affinity which failed before (for example, an all-ones CPU mask) will
1196 now succeed. Applications that need to determine the effective CPU
1197 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1198 setting it because the kernel can adjust it (and the previous size check
1199 would not detect this in the majority of cases).
1201 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1202 the following new symbols are used: fts64_children, fts64_close,
1203 fts64_open, fts64_read and fts64_set.
1205 * getaddrinfo now detects certain invalid responses on an internal netlink
1206 socket. If such responses are received, an affected process will
1207 terminate with an error message of "Unexpected error <number> on netlink
1208 descriptor <number>" or "Unexpected netlink response of size <number> on
1209 descriptor <number>". The most likely cause for these errors is a
1210 multi-threaded application which erroneously closes and reuses the netlink
1211 file descriptor while it is used by getaddrinfo.
1213 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1214 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1215 unnecessary serialization of memory allocation requests across threads.
1216 The defect is now corrected. Users should see a substantial increase in
1217 the concurent throughput of allocation requests for applications which
1218 trigger this bug. Affected applications typically create and
1219 destroy threads frequently. (Bug 19048 was reported and analyzed by
1222 * There is now a --disable-timezone-tools configure option for disabling the
1223 building and installing of the timezone related utilities (zic, zdump, and
1224 tzselect). This is useful for people who build the timezone data and code
1225 independent of the GNU C Library.
1227 * The obsolete header <regexp.h> has been removed. Programs that require
1228 this header must be updated to use <regex.h> instead.
1230 * The obsolete functions bdflush, create_module, get_kernel_syms,
1231 query_module and uselib are no longer available to newly linked binaries;
1232 the header <sys/kdaemon.h> has been removed. These functions and header
1233 were specific to systems using the Linux kernel and could not usefully be
1234 used with the GNU C Library on systems with version 2.6 or later of the
1237 * Optimized string, wcsmbs and memory functions for IBM z13.
1238 Implemented by Stefan Liebler.
1240 * Newly linked programs that define a variable called signgam will no longer
1241 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1242 require signgam to be set by those functions must ensure that they use the
1243 variable provided by the GNU C Library and declared in <math.h>, without
1244 defining their own copy.
1246 * The minimum GCC version that can be used to build this version of the GNU
1247 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1248 still be used to compile programs using the GNU C Library.
1250 Security related changes:
1252 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1253 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1255 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1256 the pointer guard feature. It is always enabled. Previously,
1257 LD_POINTER_GUARD could be used to disable security hardening in binaries
1258 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1261 * An integer overflow in hcreate and hcreate_r could lead to an
1262 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1264 * The catopen function no longer has unbounded stack usage. Reported by
1265 Max. (CVE-2015-8779)
1267 * The nan, nanf and nanl functions no longer have unbounded stack usage
1268 depending on the length of the string passed as an argument to the
1269 functions. Reported by Joseph Myers. (CVE-2014-9761)
1271 * A stack-based buffer overflow was found in libresolv when invoked from
1272 libnss_dns, allowing specially crafted DNS responses to seize control
1273 of execution flow in the DNS client. The buffer overflow occurs in
1274 the functions send_dg (send datagram) and send_vc (send TCP) for the
1275 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1276 family. The use of AF_UNSPEC triggers the low-level resolver code to
1277 send out two parallel queries for A and AAAA. A mismanagement of the
1278 buffers used for those queries could result in the response of a query
1279 writing beyond the alloca allocated buffer created by
1280 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1281 the overflow. Thanks to the Google Security Team and Red Hat for
1282 reporting the security impact of this issue, and Robert Holiday of
1283 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1285 The following bugs are resolved with this release:
1287 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1288 [887] math: Math library function "logb" and "nextafter" inconsistent
1289 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1290 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1291 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1292 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1294 [4404] localedata: German translation of "Alarm clock" is misleading
1295 [6799] math: nextafter() and nexttoward() doen't set errno on
1296 overflow/underflow errors
1297 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1299 [10432] nis: _nss_nis_setnetgrent assertion failure
1300 [11460] libc: fts has no LFS support
1301 [12926] network: getaddrinfo()/make_request() may spin forever
1302 [13065] nptl: Race condition in pthread barriers
1303 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1304 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1306 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1307 [14912] libc: Rename non-installed bits/*.h headers
1308 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1309 [15367] math: Let gcc use __builtin_isinf
1310 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1311 [15421] math: lgamma wrongly sets signgam for ISO C
1312 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1313 FE_INVALID with argument out of range
1314 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1316 [15786] dynamic-link: ifunc resolver functions can smash function
1318 [15918] math: Unnecessary check for equality in hypotf()
1319 [16061] localedata: Review / update transliteration data
1320 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1322 [16141] time: strptime %z offset restriction
1323 [16171] math: drem should be alias of remainder
1324 [16296] math: fegetround is pure?
1325 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1326 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1328 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1329 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1331 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1332 "Exception "Inexact" set" on ppc32
1333 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1334 [16517] math: Missing underflow exception from tanf/tan/tanl
1335 [16519] math: Missing underflow exception from sinhf
1336 [16520] math: Missing underflow exception from tanhf
1337 [16521] math: Missing underflow exception from exp2
1338 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1340 [16734] stdio: fopen calls mmap to allocate its buffer
1341 [16961] math: nan function incorrect handling of bad sequences
1342 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1343 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1344 [16985] locale: localedef: confusing error message when opening output
1346 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1347 [17197] locale: Redundant shift character in iconv conversion output at
1349 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1350 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1351 [17250] dynamic-link: static linking breaks nss loading
1352 (getaddrinfo/getpwnam/etc...)
1353 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1355 [17441] math: isnan() should use __builtin_isnan() in GCC
1356 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1357 (related to lock elision)
1358 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1359 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1360 [17887] time: strptime should be able to parse "+01:00" style timezones
1361 [17905] libc: catopen() Multiple unbounded stack allocations
1363 [18084] libc: backtrace (..., 0) dumps core on x86
1364 [18086] libc: nice() sets errno to 0 on success
1365 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1366 is too large (CVE-2015-8778)
1367 [18251] dynamic-link: SONAME missing when audit modules provides path
1368 [18265] libc: add attributes for wchar string and memory functions
1369 [18370] math: csqrt missing underflows
1370 [18421] libc: [hppa] read-only segment has dynamic relocations
1371 [18472] libc: Obsolete syscall wrappers should be compat symbols
1372 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1373 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1374 [18525] localedata: Remove locale timezone information
1375 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1376 [18568] localedata: Update locale data to Unicode 8.0
1377 [18589] locale: sort-test.sh fails at random
1378 [18595] math: ctan, ctanh missing underflows
1379 [18604] libc: assert macro-expands its argument
1380 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1381 contains a vector instruction exception.
1382 [18611] math: j1, jn missing errno setting on underflow
1383 [18618] localedata: sync Chechen locale definitions with other *_RU
1385 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1386 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1387 [18665] network: In send_dg, the recvfrom function is NOT always using the
1388 buffer size of a newly created buffer (CVE-2015-7547)
1389 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1391 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1393 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1394 [18699] math: tilegx cproj() for various complex infinities does not yield
1396 [18724] libc: Harden put*ent functions against data injection
1397 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1398 [18755] build: build errors with -DNDEBUG
1399 [18757] stdio: fmemopen fails to set errno on failure
1400 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1402 [18781] libc: openat64 lacks O_LARGEFILE
1403 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1404 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1405 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1406 [18790] math: [ldbl-128ibm] tanhl inaccurate
1407 [18795] libc: stpncpy fortification misses buffer lengths that are
1408 statically too large
1409 [18796] build: build fails for --disable-mathvec
1410 [18803] math: hypot missing underflows
1411 [18820] stdio: fmemopen may leak memory on failure
1412 [18823] math: csqrt spurious underflows
1413 [18824] math: fma spurious underflows
1414 [18825] math: pow missing underflows
1415 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1416 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1417 [18870] build: sem_open.c fails to compile with missing symbol
1419 [18872] stdio: Fix memory leak in printf_positional
1420 [18873] libc: posix_fallocate overflow check ineffective
1421 [18875] math: Excess precision leads incorrect libm
1422 [18877] libc: arm: mmap offset regression
1423 [18887] libc: memory corruption when using getmntent on blank lines
1424 [18918] localedata: hu_HU: change time to HH:MM:SS format
1425 [18921] libc: Regression: extraneous stat() and fstat() performed by
1427 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1428 binaries (CVE-2015-8777)
1429 [18951] math: tgamma missing underflows
1430 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1432 [18953] localedata: lt_LT: change currency symbol to the euro
1433 [18956] math: powf inaccuracy
1434 [18961] math: [i386] exp missing underflows
1435 [18966] math: [i386] exp10 missing underflows
1436 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1437 [18969] build: multiple string test failures due to missing locale
1439 [18970] libc: Reference of pthread_setcancelstate in libc.a
1440 [18977] math: float / long double Bessel functions not in XSI POSIX
1441 [18980] math: i386 libm functions return with excess range and precision
1442 [18981] math: i386 scalb*, ldexp return with excess range and precision
1443 [18982] stdio: va_list and vprintf
1444 [18985] time: Passing out of range data to strftime() causes a segfault
1446 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1447 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1448 [19012] locale: iconv_open leaks memory on error path
1449 [19016] math: clog, clog10 inaccuracy
1450 [19018] nptl: Mangle function pointers in tls_dtor_list
1451 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1452 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1453 [19048] malloc: malloc: arena free list can become cyclic, increasing
1455 [19049] math: [powerpc] erfc incorrect zero sign
1456 [19050] math: [powerpc] log* incorrect zero sign
1457 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1458 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1459 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1460 [19074] network: Data race in _res_hconf_reorder_addrs
1461 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1462 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1463 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1465 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1467 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1468 [19086] manual: posix_fallocate64 documented argument order is wrong.
1469 [19088] math: lround, llround missing exceptions close to overflow
1471 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1472 [19095] math: dbl-64 lrint incorrect for 64-bit long
1473 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1474 [19124] dynamic-link: ld.so failed to build with older assmebler
1475 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1476 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1477 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1478 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1479 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1480 pthread_setaffinity_np
1481 [19156] math: [ldbl-128] j0l spurious underflows
1482 [19164] nptl: tst-getcpu fails with many possible CPUs
1483 [19168] math: math/test-ildoubl and math/test-ldouble failure
1484 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1485 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1487 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1488 (FE_NOMASK_ENV) do not clear SSE exceptions
1489 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1490 _int_new_arena/reused_arena
1491 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1492 [19201] math: dbl-64 remainder incorrect sign of zero result
1493 [19205] math: bits/math-finite.h conditions do not match math.h and
1495 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1496 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1497 for C99-based standards
1498 [19212] libc: features.h not -Wundef clean
1499 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1501 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1502 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1503 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1505 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1506 "inexact" exceptions
1507 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1509 [19242] libc: strtol incorrect in Turkish locales
1510 [19243] malloc: reused_arena can pick an arena on the free list, leading
1511 to an assertion failure and reference count corruption
1512 [19253] time: tzset() ineffective when temporary TZ did not include DST
1514 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1515 [19270] math: [hppa] Shared libm missing __isnanl
1516 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1517 MAP_HUGETLB and MAP_STACK defines
1518 [19313] nptl: Wrong __cpu_mask for x32
1519 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1520 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1521 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1522 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1523 [19363] time: x32: times() return value wrongly truncates/sign extends
1525 [19367] dynamic-link: Improve branch prediction on Silvermont
1526 [19369] network: Default domain name not reset by res_ninit when "search"
1527 / "domain" entry is removed from resolv.conf
1528 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1530 [19385] localedata: bg_BG: time separator should be colon, not comma
1531 [19408] libc: linux personality syscall wrapper may erroneously return an
1532 error on 32-bit architectures
1533 [19415] libc: dladdr returns wrong names on hppa
1534 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1535 IBM905, IBM907, and IBM909
1536 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1537 [19443] build: build failures with -DDEBUG
1538 [19451] build: Make check fails on test-double-vlen2
1539 [19462] libc: Glibc failed to build with -Os
1540 [19465] math: Wrong code with -Os
1541 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1543 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1545 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1546 [19486] math: S390: Math tests fail with "Exception Inexact set".
1547 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1548 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1550 [19590] math: Fail to build shared objects that use libmvec.so functions.
1554 * The following bugs are resolved with this release:
1556 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1557 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1558 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1559 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1560 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1561 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1562 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1563 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1564 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1565 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1566 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1567 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1568 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1569 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1570 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1571 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1572 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1573 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1574 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1575 18657, 18676, 18694, 18696, 18887.
1577 * Cache information can be queried via sysconf() function on s390 e.g. with
1578 _SC_LEVEL1_ICACHE_SIZE as argument.
1580 * A buffer overflow in gethostbyname_r and related functions performing DNS
1581 requests has been fixed. If the NSS functions were called with a
1582 misaligned buffer, the buffer length change due to pointer alignment was
1583 not taken into account. This could result in application crashes or,
1584 potentially arbitrary code execution, using crafted, but syntactically
1585 valid DNS responses. (CVE-2015-1781)
1587 * The time zone file parser has been made more robust against crafted time
1588 zone files, avoiding heap buffer overflows related to the processing of
1589 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1590 large time zone data files. Overly long time zone specifiers in the TZ
1591 variable no longer result in stack overflows and crashes.
1593 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1594 for LD and GD on x86 and x86-64, has been implemented. You will need
1595 binutils-2.24 or later to enable this optimization.
1597 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1598 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1599 Hat). These updates cause user visible changes, such as the fix for bug
1602 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1603 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1604 condition in some applications.
1606 * Added vector math library named libmvec with the following vectorized x86_64
1607 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1609 The library can be disabled with --disable-mathvec. Use of the functions is
1610 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1611 Shared library libmvec.so is linked in as needed when using -lm (no need to
1612 specify -lmvec explicitly for not static builds).
1613 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1615 * A new fmemopen implementation has been added with the goal of POSIX
1616 compliance. The new implementation fixes the following long-standing
1617 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1618 old implementation is still present for use be by existing binaries.
1620 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1621 releases. It has been fixed to match 2.19 and older, but binaries built
1622 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1624 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1625 Contributed by Roland McGrath (Google).
1627 * The header <regexp.h> is deprecated, and will be removed in a future
1628 release. Use of this header will trigger a deprecation warning.
1629 Application developers should update their code to use <regex.h> instead.
1631 This header was formerly part of SUS, but was deprecated in 1994 and
1632 removed from the standard in 2001. Also, the glibc implementation
1633 leaks memory. See BZ#18681 for more details.
1637 * The following bugs are resolved with this release:
1639 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1640 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1641 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1642 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1643 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1644 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1645 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1646 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1647 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1648 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1651 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1652 too little memory for the to-be-scanned arguments and overflow the
1653 allocated buffer. The implementation now correctly computes the required
1654 buffer size when using malloc, and switches to malloc from alloca as
1657 * A new semaphore algorithm has been implemented in generic C code for all
1658 machines. Previous custom assembly implementations of semaphore were
1659 difficult to reason about or ensure that they were safe. The new version
1660 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1661 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1662 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1664 * Port to Altera Nios II has been contributed by Mentor Graphics.
1666 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1667 implementations for powerpc64/powerpc64le.
1668 Implemented by Adhemerval Zanella (IBM).
1670 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1671 and powerpc64le. This may improve lock scaling of existing programs on
1672 HTM capable systems. The lock elision code is only enabled with
1673 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1674 powerpc will issue a transaction abort on every syscall to avoid side
1675 effects being visible outside transactions.
1677 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1678 AArch64. Contributed by ARM Ltd.
1680 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1682 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1683 under certain input conditions resulting in the execution of a shell for
1684 command substitution when the application did not request it. The
1685 implementation now checks WRDE_NOCMD immediately before executing the
1686 shell and returns the error WRDE_CMDSUB as expected.
1688 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1689 processing format strings with a large number of format specifiers.
1691 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1692 infinite loop if the DNS response contained a PTR record of an unexpected
1695 * The minimum GCC version that can be used to build this version of the GNU
1696 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1697 still be used to compile programs using the GNU C Library.
1699 * The GNU C Library is now built with -Werror by default. This can be
1700 disabled by configuring with --disable-werror.
1702 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1704 * The obsolete sigvec function has been removed. This was the original
1705 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1706 programs have been using instead for about 25 years. Of course, ABI
1707 compatibility for old binaries using sigvec remains intact.
1709 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1710 with newer versions of bison.
1712 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1713 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1714 registers overlay two consecutive single-precision registers. MIPS32R2
1715 introduced a new FPU mode (FR=1) where double-precision registers extend the
1716 corresponding single-precision registers which is incompatible with the
1717 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1718 require the use of FR=1 making a transition necessary. New o32 ABI
1719 extensions enable users to migrate over time from the original o32 ABI
1720 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1721 tracks the ABI of any loaded object and verifies that new objects are
1722 compatible. Mode transitions will also be requested as required and
1723 unsupportable objects will be rejected. The ABI checks include both soft and
1724 hard float ABIs for o32, n32 and n64.
1726 GCC 5 with GNU binutils 2.25 onwards:
1727 It is strongly recommended that all o32 system libraries are built using the
1728 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1729 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1730 compiler using --with-fp-32=xx to set this by default.
1734 * The following bugs are resolved with this release:
1736 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1737 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1738 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1739 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1740 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1741 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1742 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1743 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1744 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1745 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1746 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1747 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1748 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1749 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1750 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1752 * Reverted change of ABI data structures for s390 and s390x:
1753 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1754 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1755 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1756 are preserved pointing straight to the same implementation as the old ones.
1757 Given that, new callers will simply provide a too-big buffer to these
1758 functions. Any applications/libraries out there that embed jmp_buf or
1759 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1760 against 2.19 headers will have to rebuilt again. This is necessary in any
1761 case to revert the breakage in their ABI caused by the glibc change.
1763 * Support for file description locks is added to systems running the
1764 Linux kernel. The standard file locking interfaces are extended to
1765 operate on file descriptions, not file descriptors, via the use of
1766 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1767 are associated with an open file instead of a process.
1769 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1771 * The minimum Linux kernel version that this version of the GNU C Library
1772 can be used with is 2.6.32.
1774 * Running the testsuite no longer terminates as soon as a test fails.
1775 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1776 with PASS or FAIL lines for individual tests. A summary of the results is
1777 printed, including a list of failing lists, and "make check" exits with
1778 error status if there were any unexpected failures. "make check
1779 stop-on-test-failure=y" may be used to keep the old behavior.
1781 * The am33 port, which had not worked for several years, has been removed
1784 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1785 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1786 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1787 conflicted with POSIX (support for which was removed in 2.19), the
1788 interfaces those macros enabled remain available when compiling with
1789 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1790 test macros defined.
1792 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1794 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1795 This may improve lock scaling of existing programs on TX capable systems.
1796 The lock elision code is only built with --enable-lock-elision=yes and
1797 then requires a GCC version supporting the TX builtins. With lock elision
1798 default mutexes are elided via __builtin_tbegin, if the cpu supports
1799 transactions. By default lock elision is not enabled and the elision code
1802 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1803 copy the path argument. This allowed programs to cause posix_spawn to
1804 deference a dangling pointer, or use an unexpected pathname argument if
1805 the string was modified after the posix_spawn_file_actions_addopen
1808 * All supported architectures now use the main glibc sysdeps directory
1809 instead of some being in a separate "ports" directory (which was
1810 distributed separately before glibc 2.17).
1812 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1813 On configurations that support it (all Linux configurations), it's now
1814 used regardless of the --enable-add-ons switch to configure. It is no
1815 longer possible to build such configurations without pthreads support.
1817 * Locale names, including those obtained from environment variables (LANG
1818 and the LC_* variables), are more tightly checked for proper syntax.
1819 setlocale will now fail (with EINVAL) for locale names that are overly
1820 long, contain slashes without starting with a slash, or contain ".." path
1821 components. (CVE-2014-0475) Previously, some valid locale names were
1822 silently replaced with the "C" locale when running in AT_SECURE mode
1823 (e.g., in a SUID program). This is no longer necessary because of the
1826 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1827 with application code using Intel MPX instructions. (With all previous
1828 versions, the MPX register state could be clobbered when making calls
1829 into or out of a shared library.) Note that while the new dynamic
1830 linker is compatible with all known x86 hardware whether or not it
1831 supports Intel MPX, some x86 instruction-set emulators might fail to
1832 handle the new instruction encodings. This is known to affect Valgrind
1833 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1834 release), and might affect other tools that do instruction emulation.
1836 * Support for loadable gconv transliteration modules has been removed.
1837 The support for transliteration modules has been non-functional for
1838 over a decade, and the removal is prompted by security defects. The
1839 normal gconv conversion modules are still supported. Transliteration
1840 with //TRANSLIT is still possible, and the //IGNORE specifier
1841 continues to be supported. (CVE-2014-5119)
1843 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1844 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1845 resulting a denial-of-service security vulnerability in applications which
1846 use functions related to iconv. (CVE-2014-6040)
1850 * The following bugs are resolved with this release:
1852 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1853 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1854 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1855 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1856 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1857 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1858 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1859 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1860 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1861 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1862 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1863 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1864 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1865 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1866 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1867 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1868 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1869 16431, 16453, 16474, 16506, 16510, 16529
1871 * Slovenian translations for glibc messages have been contributed by the
1872 Translation Project's Slovenian team of translators.
1874 * The public headers no longer use __unused nor __block. This change is to
1875 support compiling programs that are derived from BSD sources and use
1876 __unused internally, and to support compiling with Clang's -fblock
1877 extension which uses __block.
1879 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1880 large collation sequences to optimize multiple passes. This cache
1881 computation may overflow for large collation sequences and may cause a
1882 stack or buffer overflow. This is now fixed to use a slower algorithm
1883 which does not use a cache if there is an integer overflow.
1885 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1886 rules for large collation sequences to optimize multiple passes and falls
1887 back to alloca if malloc fails, resulting in a possible stack overflow.
1888 The implementation now falls back to an uncached collation sequence lookup
1891 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1892 initialized for static applications resulting in the security feature
1893 being disabled. The pointer guard is now correctly initialized to a
1894 random value for static applications. Existing static applications need
1895 to be recompiled to take advantage of the fix (bug 15754).
1897 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1898 to the d_name member of struct dirent, or omit the terminating NUL
1899 character. (Bugzilla #14699).
1901 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1902 aligned_alloc functions could allocate too few bytes or corrupt the
1903 heap when passed very large allocation size values (Bugzilla #15855,
1906 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1907 for AF_INET6 has been fixed (Bugzilla #16072).
1909 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1910 pap_CW, quz_PE, the_NP.
1912 * Substantially revised locales: gd_GB, ht_HT
1914 * The LC_ADDRESS field was updated to support country_car for almost all
1917 * ISO 1427 definitions were updated.
1919 * ISO 3166 definitions were updated.
1921 * The localedef utility now supports --big-endian and --little-endian
1922 command-line options to generate locales for a different system from that
1923 for which the C library was built.
1925 * Binary locale files now only depend on the endianness of the system for
1926 which they are generated and not on other properties of that system. As a
1927 consequence, binary files generated with new localedef may be incompatible
1928 with old versions of the GNU C Library, and binary files generated with
1929 old localedef may be incompatible with this version of the GNU C Library,
1930 in the following circumstances:
1932 + Locale files may be incompatible on m68k systems.
1934 + Locale archive files (but not separate files for individual locales) may
1935 be incompatible on systems where plain "char" is signed.
1937 * The configure option --disable-versioning has been removed. Builds with
1938 --disable-versioning had not worked for several years.
1940 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1941 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1943 * SystemTap probes for malloc have been introduced.
1945 * SystemTap probes for slow multiple precision fallback paths of
1946 transcendental functions have been introduced.
1948 * Support for powerpc64le has been added.
1950 * The soft-float powerpc port now supports e500 processors.
1952 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1954 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1955 set of header declarations that are enabled by default, even when other
1956 feature test macros or compiler options such as -std=c99 would otherwise
1957 disable some of those declarations.
1959 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1960 conflict with POSIX. The libbsd-compat library (which was a dummy library
1961 that did nothing) has also been removed.
1963 * Preliminary documentation about Multi-Thread, Async-Signal and
1964 Async-Cancel Safety has been added.
1966 * Change of ABI data structures for s390 and s390x:
1967 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1968 allow for future hardware extensions. All library functions that accept or
1969 return these structures were versioned in order to provide backwards
1970 compatibility with existing code. However, not all cases can be handled
1971 with interface versioning. If an application embeds either structure into
1972 another structure and that structure is passed to another compilation unit
1973 which expects the newer larger-sized object then any access to the new
1974 fields will result in undefined behaviour. Similarly any access to fields
1975 that were moved by the enlarging of the structures will cause undefined
1976 behaviour. To fix the undefined behaviour all objects that are part of the
1977 final in-memory image must agree on the size of structures, and this may
1978 require recompilation.
1982 * The following bugs are resolved with this release:
1984 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1985 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1986 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1987 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1988 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1989 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1990 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1991 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1992 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1993 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1994 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1995 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1996 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1999 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2000 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2001 Distributions can re-enable building and using pt_chown via the new configure
2002 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2003 considerable security risks and should only be used if the distribution
2004 understands and accepts the risks.
2006 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2009 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2010 fixed (Bugzilla #15330).
2012 * Add support for calling C++11 thread_local object destructors on thread
2013 and program exit. This needs compiler support for offloading C++11
2014 destructor calls to glibc.
2016 * Improved worst case performance of libm functions with double inputs and
2019 * Support for priority inherited mutexes in pthread condition variables on
2020 non-x86 architectures.
2022 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2024 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2026 * Optimized string functions for ARM. Implemented by Will Newton and
2029 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2031 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2034 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2035 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2037 * Added a benchmark framework to track performance of functions in glibc.
2039 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2040 It is based on draft TS 18661 and currently enabled as a GNU extension.
2042 * On Linux, the clock function now uses the clock_gettime system call
2043 for improved precision, rather than old times system call.
2045 * Added support for version-3 tz format files. This is needed when using
2046 the tz database release 2013e or later, and affects a few unusual cases --
2047 currently only TZ='America/Godthab' for time stamps after 2037.
2049 * Added new API functions pthread_getattr_default_np and
2050 pthread_setattr_default_np to get and set the default pthread
2051 attributes of a process.
2053 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2054 This may improve lock scaling of existing programs on TSX capable systems.
2055 When the --enable-lock-elision=yes parameter is specified at configure
2056 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2059 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2060 availability). Implemented by Ryan S. Arnold.
2062 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2066 * The following bugs are resolved with this release:
2068 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2069 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2070 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2071 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2072 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2073 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2074 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2075 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2076 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2077 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2078 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2079 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2080 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2081 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2082 14879, 14889, 14893, 14898, 14914.
2084 * Optimization of memcpy for MIPS.
2086 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2087 EMFILE has been fixed (Bugzilla #14889).
2089 * The add-on ports collection is now distributed in the "ports" subdirectory
2090 of the main GNU C Library distribution, rather than separately.
2092 * Port to ARM AArch64 contributed by Linaro.
2094 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2095 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2097 Implemented by Andreas Krebbel.
2099 * The new function secure_getenv allows secure access to the environment,
2100 returning NULL if running in a SUID/SGID process. This function replaces
2101 the internal function __secure_getenv.
2103 * SystemTap static probes have been added into the dynamic linker.
2104 Implemented by Gary Benson.
2106 * Optimizations of string functions strstr, strcasestr and memmem.
2107 Implemented by Maxim Kuvyrkov.
2109 * The minimum Linux kernel version that this version of the GNU C Library
2110 can be used with is 2.6.16.
2112 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2113 powerpc POWER7. Implemented by Will Schmidt.
2115 * New configure option --disable-nscd builds the C library such that it
2116 never attempts to contact the Name Service Caching Daemon (nscd).
2117 New configure option --disable-build-nscd avoids building nscd itself;
2118 this is the default if --disable-nscd is used.
2120 * Improved support for cross-compilation, including cross-testing and
2121 bootstrap builds without a previously built glibc.
2123 * Several testsuite tests are now able to test multiple IFUNC variants of an
2124 interface, rather than just testing the one that would be chooen by
2127 * New configure options --with-bugurl and --with-pkgversion, for
2128 distributors to use to embed their bug-reporting and package version
2129 information in --help and --version output.
2131 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2132 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2133 allows creation of chroots without the procfs mounted on /proc.
2135 * The `crypt' function now fails if passed salt bytes that violate the
2136 specification for those values. On Linux, the `crypt' function will
2137 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2138 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2139 when the mode is enabled.
2141 * The `clock_*' suite of functions (declared in <time.h>) is now available
2142 directly in the main C library. Previously it was necessary to link with
2143 -lrt to use these functions. This change has the effect that a
2144 single-threaded program that uses a function such as `clock_gettime' (and
2145 is not linked with -lrt) will no longer implicitly load the pthreads
2146 library at runtime and so will not suffer the overheads associated with
2147 multi-thread support in other code such as the C++ runtime library.
2149 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2154 * The following bugs are resolved with this release:
2156 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2157 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2158 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2159 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2160 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2161 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2162 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2163 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2164 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2165 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2166 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2167 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2168 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2169 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2170 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2171 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2172 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2173 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2174 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2175 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2176 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2177 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2180 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2181 configuring glibc with:
2182 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2183 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2184 Implemented by H.J. Lu.
2188 + define static_assert
2190 + do not declare gets
2192 + declare at_quick_exit and quick_exit also for ISO C11
2194 + aligned_alloc. NB: The code is deliberately allows the size parameter
2195 to not be a multiple of the alignment. This is a moronic requirement
2196 in the standard but it is only a requirement on the caller, not the
2199 + timespec_get added
2201 + uchar.h support added
2203 + CMPLX, CMPLXF, CMPLXL added
2205 Implemented by Ulrich Drepper.
2207 * Support for the IA-64 has been moved to ports.
2209 * Remove support for anything but ELF binary format
2211 * Checking versions of poll, ppoll added.
2212 Implemented by Ulrich Drepper.
2214 * More generic and 64-bit performance optimizations to math functions.
2215 Implemented by Ulrich Drepper.
2217 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2218 headers and functions available at compile time as they were before
2219 version 2.14. This option will be removed at some time in the future
2220 after the TI-RPC library becomes fully sufficient for the needs of
2221 existing applications.
2223 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2224 Note that glibc is not expected to work with any Linux kernel version
2227 * New header <sys/auxv.h> and function getauxval allowing easy access to
2228 the AT_* key-value pairs passed from the Linux kernel. The header also
2229 defines the HWCAP_* bits associated with the AT_HWCAP key.
2231 * New locales: mag_IN
2233 * New configure option --enable-systemtap builds SystemTap static probes
2234 into libc for setjmp and longjmp and into libpthread for various operations.
2235 So far the setjmp/longjmp probes and some of the libpthread probes are
2236 provided only for i*86 and x86_64.
2237 Implemented by Roland McGrath and Rayson Ho.
2239 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2241 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2244 * More optimized functions for SPARC. Implemented by David S. Miller.
2246 * Improved support for cross-compilation, especially bootstrap builds
2247 without a previously built glibc.
2249 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2250 Chris Metcalf from Tilera.
2252 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2253 now supported for ARM processors.
2255 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2256 as the name of the dynamic linker, to distinguish it from the
2257 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2259 * Support for CRIS has been removed from ports.
2261 * A new class of installed header has been documented for low-level
2262 platform-specific functionality. PowerPC added the first instance with a
2263 function to provide time base register access. Contributed by Tulio
2264 Magno Quites Machado Filho.
2266 * ix86 configurations now install header files that are consistent with
2267 what x86-64 configurations install. These same header files can be used
2268 for -m32, -m64, or -mx32 builds.
2269 Contributed by H.J. Lu.
2271 * Math library bug fixes. A thorough audit of all open math library bugs was
2272 conducted by Joseph Myers. Significant progress was made on many math
2273 library bugs resulting in more accurate exceptions and function results.
2274 Many thanks to all those that contributed including Andreas Jaeger for his
2275 patch review and work on the x87 trigonometric instruction issues.
2277 * Timezone data is no longer installed. Timezone-related binaries and scripts
2278 will continue to be installed. Users should obtain their timezone data from
2279 their distribution provider or from the tzdata package at
2280 <ftp://munnari.oz.au/pub/>.
2284 * The following bugs are resolved with this release:
2286 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2287 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2288 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2289 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2290 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2291 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2292 13484, 13506, 13515, 13523, 13524, 13538, 13540
2294 * New program pldd to list loaded object of a process
2295 Implemented by Ulrich Drepper.
2297 * Add nss_db support back to glibc. No more dependency on Berkeley db
2298 and support for initgroups lookups.
2299 Implemented by Ulrich Drepper.
2301 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2302 Contributed by HJ Lu.
2304 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2305 Contributed by HJ Lu.
2307 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2308 on x86-32 and x86-64.
2309 Contributed by Liubov Dmitrieva.
2311 * Optimized strchr and strrchr for SSE on x86-32.
2312 Contributed by Liubov Dmitrieva.
2314 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2315 for x86-64 and x86-32.
2316 Contributed by Liubov Dmitrieva.
2318 * New interfaces: scandirat, scandirat64
2319 Implemented by Ulrich Drepper.
2321 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2322 Implemented by Ulrich Drepper.
2324 * nscd now also caches the netgroup database.
2325 Implemented by Ulrich Drepper.
2327 * Integrate libm with gcc's -ffinite-math-only option.
2328 Implemented by Ulrich Drepper.
2330 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2331 to math functions. Implemented by Ulrich Drepper.
2333 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2334 Implemented by Ulrich Drepper.
2336 * New Linux interfaces: process_vm_readv, process_vm_writev
2338 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2339 Implemented by Ulrich Drepper.
2341 * Optimized nearbyint and strcasecmp for PPC.
2342 Implemented by Adhemerval Zanella.
2344 * New locales: bho_IN, unm_US, es_CU, ta_LK
2348 * The following bugs are resolved with this release:
2350 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2351 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2352 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2353 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2354 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2355 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2356 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2357 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2358 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2359 12795, 12811, 12813, 12814, 12841
2361 * The RPC implementation in libc is obsoleted. Old programs keep working
2362 but new programs cannot be linked with the routines in libc anymore.
2363 Programs in need of RPC functionality must be linked against TI-RPC.
2364 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2366 Visible changes of this change include (obviously) the inability to link
2367 programs using RPC functions without referencing the TI-RPC library and the
2368 removal of the RPC headers from the glibc headers.
2369 Implemented by Ulrich Drepper.
2371 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2372 syncfs, setns, sendmmsg
2374 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2375 yue_HK, lij_IT, mhr_RU
2377 * New encodings: CP770, CP771, CP772, CP773, CP774
2379 * New program sotruss to trace calls through PLTs
2380 Implemented by Ulrich Drepper.
2382 * The malloc hook implementation is marked deprecated and will be removed
2383 from the default implementation in the next version. The design never
2384 worked ever since the introduction of threads. Even programs which do
2385 not create threads themselves can use multiple threads created internally.
2389 * The following bugs are resolved with this release:
2391 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2392 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2393 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2394 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2397 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2399 * POWER7 optimizations: memset, memcmp, strncmp
2401 * New optimized string functions for x86-64: strnlen (SSE2),
2402 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2403 Implemented by Ulrich Drepper.
2407 * The following bugs are resolved with this release:
2409 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2410 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2411 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2412 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2413 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2414 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2415 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2416 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2418 * New interfaces: pthread_getname_np, pthread_setname_np
2420 * New Linux interface: recvmmsg
2422 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2424 * The dynamic linker now recognizes supported ABI versions from the
2425 EI_ABIVERSION field in the ELF header.
2426 Implemented by Ulrich Drepper.
2428 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2429 will not be used to fill in password fields in the passwd.byname replies.
2430 Instead it is used to synthesize the shadow.byname table, should it be
2431 missing. This is a useful mode in some installations involving Solaris.
2432 Implemented by Ulrich Drepper.
2434 * New locales: kok_IN, sq_MK, cv_RU
2438 * The following bugs are resolved with this release:
2440 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2441 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2442 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2443 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2444 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2445 10789, 10817, 10823, 10840, 10847
2447 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2449 Implemented by Ulrich Drepper.
2451 * Checking version of longjmp added that fails if an uninitialized stack
2452 frame would be created. Implemented by Ulrich Drepper.
2454 * STT_GNU_IFUNC is now supported in static executables.
2455 Implemented by H.J. Lu.
2457 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2459 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2460 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2461 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2462 Contributed by H.J. Lu.
2464 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2465 Implemented by Ulrich Drepper.
2467 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2469 Contributed by H.J. Lu.
2471 * Support for fma instruction in AVX on x86-64.
2472 Implemented by H.J. Lu and Ulrich Drepper.
2474 * AVX support in x86-64 auditing support in ld.so.
2475 Implemented by H.J. Lu.
2477 * STB_GNU_UNIQUE support added.
2478 Implemented by Ulrich Drepper.
2480 * Implement second fallback mode for optimized DNS lookup for even more
2481 broken environments. If two requests from the same port are not
2482 handled correctly close the socket and open a new one before sending
2483 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2484 can be used to select this mode right away, instead of rediscovering the
2485 necessity is every process again.
2486 Implemented by Ulrich Drepper.
2488 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2489 Implemented by Adam Tkac.
2491 * Optimized iconv conversions for S390x.
2492 Implemented by Andreas Krebbel.
2494 * Using condvars with PI mutexes is now more efficient due to kernel
2495 support for requeueing to PI futexes. NPTL support added for x86-64.
2496 Implemented by Ulrich Drepper.
2498 * New locales: ps_AF, my_MM
2502 * The following bugs are resolved with this release:
2504 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2505 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2506 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2507 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2508 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2510 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2511 Implemented by Ulrich Drepper.
2513 * New Linux interfaces: accept4, fallocate, fallocate64.
2514 Implemented by Ulrich Drepper.
2516 * Correct declarations of string function when used in C++ code. This
2517 could lead to compile errors for invalid C++ code.
2519 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2522 * New POSIX 2008 interface: psiginfo
2523 Implemented by Ulrich Drepper.
2525 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2526 Implemented by Ulrich Drepper.
2528 * Support for selecting between multiple function definitions at runtime
2529 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2531 * The libcrypt library can now use the hash function implementations in
2532 NSS. Implemented by Ulrich Drepper.
2534 * The malloc implementation can be compiled to be less memory efficient
2535 but higher performing in multi-threaded programs.
2536 Implemented by Ulrich Drepper.
2538 * New locales: nan_TW@latin, ks_IN
2540 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2541 Implemented by Ulrich Drepper.
2543 * Extended printf hook support. It is possible to use user-defined types
2544 and extend existing format specifiers.
2545 Implemented by Ulrich Drepper.
2547 * Handling for group shadow files has been added.
2548 Implemented by Ulrich Drepper.
2550 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2551 servers (the case, e.g., for some people using the built-in DNS
2552 server in ADSL modems/routers). There is a once-per-process timeout
2553 in case of a broken server. To avoid it, users can run nscd or put
2554 'options single-request' in /etc/resolv.conf.
2555 Implemented by Ulrich Drepper.
2559 * The following bugs are resolved with this release:
2561 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2562 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2563 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2564 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2565 6968, 6974, 6980, 6995, 7008, 7009, 7029
2567 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2568 up at the same time. Implemented by Ulrich Drepper.
2570 * TLS descriptors for LD and GD on x86 and x86-64.
2571 Implemented by Alexandre Oliva.
2573 * getaddrinfo now handles DCCP and UDPlite.
2574 Implemented by Ulrich Drepper.
2576 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2577 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2578 Implemented by Ulrich Drepper.
2580 * New implementation of memmem, strstr, and strcasestr which is O(n).
2581 Implemented by Eric Blake.
2583 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2585 * Implement "e" option for popen to open file descriptor with the
2586 close-on-exec flag set. Implemented by Ulrich Drepper.
2588 * Implement "b" mode for fmemopen. In this mode writes writes don't
2589 implicitly add a NUL byte and seeks from the end of the buffer really
2590 use the buffer end, not the string length as the basis.
2591 Implemented by Ulrich Drepper.
2593 * Many functions, exported and internal, now atomically set the close-on-exec
2594 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2596 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2598 Implemented by Pravin Satpute.
2600 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2604 * The following bugs are resolved with this release:
2606 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2607 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2608 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2609 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2610 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2611 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2612 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2614 * New locales: bo_CN, bo_IN, shs_CA.
2616 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2618 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2619 Implemented by Pravin Satpute.
2621 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2623 * nscd caches now all timeouts for DNS entries
2624 Implemented by Ulrich Drepper.
2626 * nscd is more efficient and wakes up less often.
2627 Implemented by Ulrich Drepper.
2629 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2630 vdprintf, and obstack_vprintf.
2631 Implemented by Jakub Jelinek.
2633 * Faster memset for x86-64.
2634 Implemented by Harsha Jagasia and H.J. Lu.
2636 * Faster memcpy on x86.
2637 Implemented by Ulrich Drepper.
2639 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2640 Implemented by Ulrich Drepper.
2642 * Faster sqrt and sqrtf implemention for some PPC variants.
2643 Implemented by Steven Munroe.
2647 * The following bugs are resolved with this release:
2649 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2650 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2651 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2652 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2653 5071, 5103, 5104, 5112, 5113, 5184, 5186
2655 * More checking functions: fread, fread_unlocked, open*, mq_open.
2656 Implemented by Jakub Jelinek and Ulrich Drepper.
2658 * Extend fortification to C++. Implemented by Jakub Jelinek.
2660 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2661 by not recognizing 'a' as a modifier when those specs are requested.
2662 Implemented by Jakub Jelinek.
2664 * PPC optimizations to math and string functions.
2665 Implemented by Steven Munroe.
2667 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2668 options to be passed. Implemented by Ulrich Drepper.
2670 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2672 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2673 Implemented by Ulrich Drepper.
2675 * Handle private futexes in the NPTL implementation.
2676 Implemented by Jakub Jelinek and Ulrich Drepper.
2678 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2679 Implemented by Roland McGrath and Ulrich Drepper.
2681 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2683 * SHA-256 and SHA-512 based password encryption.
2684 Implemented by Ulrich Drepper.
2686 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2687 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2690 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2691 Implemented by Ulrich Drepper.
2695 * The following bugs are resolved with this release:
2697 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2698 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2699 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2700 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2701 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2702 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2703 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2706 * New Linux interfaces: epoll_pwait, sched_getcpu.
2708 * New generic interfaces: strerror_l.
2710 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2714 * The following bugs are resolved with this release:
2716 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2717 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2718 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2719 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2720 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2721 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2722 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2723 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2724 3225, 3251, 3252, 3253, 3273
2726 * For Linux, the sorting of addresses returned by getaddrinfo now also
2727 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2728 Implemented by Ulrich Drepper.
2730 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2731 Implemented by Ulrich Drepper.
2733 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2735 * New iconv module for MIK. Contributed by Alexander Shopov.
2737 * For sites with broken group and/or passwd database, the auto-propagate
2738 option of nscd can prevent creating ID lookup entries from the results
2739 of a name lookup and vice versa. This usually is no problem but some
2740 site might have problems with the default behavior.
2741 Implemented by Ulrich Drepper.
2743 * Iterating over entire database in NIS can be slow. With the
2744 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2745 to trade time for memory. The entire database will be read at once.
2746 Implemented by Ulrich Drepper.
2748 * The interfaces introduced in RFC 3542 have been implemented by
2751 * Support for the new ELF hash table format was added by Ulrich Drepper.
2753 * Support for priority inheritance mutexes added by Jakub Jelinek and
2756 * Support for priority protected mutexes added by Jakub Jelinek.
2758 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2759 or_IN, csb_PL, fy_NL, sr_ME.
2763 * More overflow detection functions.
2765 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2766 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2768 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2769 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2770 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2771 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2772 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2773 by Masahide Washizawa.
2775 * It is now possible to install audit libraries for the dynamic linker using
2776 LD_AUDIT. Implemented by Ulrich Drepper.
2778 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2779 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2780 For a libc and libpthread that works well on Linux 2.4 kernels, we
2781 recommend using the stable 2.3 branch.
2783 * The new function open_wmemstream acts like open_memstream,
2784 but uses a wchar_t wide character buffer.
2786 * The new function ppoll is an improved version of poll, similar to pselect.
2788 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2789 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2790 renameat, unlinkat, linkat, symlinkat, readlinkat.
2792 * New Linux kernel system calls: unshare,
2793 inotify_init, inotify_add_watch, inotify_rm_watch.
2795 * The euidaccess function is now also known by the alias eaccess,
2796 for compatibility with some other systems.
2798 * Timezone data updated to 2006b version.
2802 * The following bugs are resolved with this release:
2804 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2805 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2806 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2807 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2808 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2809 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2811 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2813 * As of this release, GCC 4 can be used to compile the C Library.
2815 * Timezone data updated to 2005m version.
2819 * The following bugs are resolved with this release:
2821 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2822 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2823 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2824 777, 787, 821, 822, 823, 825
2826 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2830 * Support for RFC 3678. Real implementations exist only for Linux so far.
2831 Implemented by Ulrich Drepper.
2833 * nscd can now cache entries persistently. Expiring entries are reloaded.
2834 For speedups the cache can be shared in memory with client processes.
2835 Implemented by Ulrich Drepper.
2837 * nscd can now perform SELinux checks.
2838 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2840 * getaddrinfo queries are now cached. Canonical name lookup is performed
2842 Implemented by Ulrich Drepper.
2844 * The nothrow function attribute is used when headers are used by gcc when
2845 compiling C code. This can avoid significant amounts of exception
2848 * The malloc functions perform more error checking and are stricter when
2849 it comes to reacting on errors. The default action is to terminate
2850 the process after showing an error message. Implemented by Ulrich Drepper.
2852 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2853 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2855 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2856 namespaces using the new function dlmopen(). This feature is of course,
2857 like most other dynamic loading functionality, not available in statically
2858 linked applications. Implemented by Ulrich Drepper.
2860 * Low-overhead boundary checking variants of string and some stdio functions
2861 were added. These are to be used in conjunction with a gcc patch by
2862 Jakub Jelinek which adds calls to these functions if possible.
2863 Implemented by Jakub Jelinek and Ulrich Drepper.
2865 * Old code for several operating systems and machine architectures that
2866 have not been in working condition in a long time have been removed from
2867 the main source tree maintained by the GNU C Library's maintainers.
2868 These files are now reside in the separate `ports' source module
2869 that is usable as an add-on when building the library.
2873 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2874 interrogate the dynamic linker, compatible with the Solaris interface.
2876 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2877 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2879 * getifaddrs now uses the netlink interface on Linux to get its information.
2880 Implemented by Thorsten Kukuk.
2882 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2883 Implemented by Ulrich Drepper.
2885 * support for non-executable stacks on x86 has been added. Changes mostly
2888 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2891 * getaddrinfo now performs destination address selection according to
2896 * Thread-safe interfaces for many functions that access locale data
2897 were added in version 2.3, but these features were omitted from NEWS.
2898 Many functions have variants with an `_l' suffix that take a `locale_t'
2899 object as a parameter rather than consulting the current locale.
2900 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2901 create and maintain `locale_t' objects. Additionally, the new function
2902 `uselocale' sets "the current locale" (as used by functions not so
2903 parameterized) set for an individual thread. These features were added
2904 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2906 * The functions getresuid, getresgid, setresuid, and setresgid, which
2907 have long been available on Linux, are now declared in <unistd.h>
2908 and are now also available on the Hurd.
2910 * ELF thread-local storage support (TLS) now works on x86-64.
2912 * The new dynamic string token $LIB is expanded in shared library names.
2913 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2915 * Aldy Hernandez contributed complete software floating point support for
2916 PowerPC machines with no FPU.
2918 * fexecve is implemented on Linux.
2920 * The `btowc' function should work at least twice as fast due to
2921 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2923 * With approriate thread add-ons cancelable functions are now implemented
2924 in libc.so as well. No need to call the function in libpthread. This
2925 change allowed to finally disable the incorrect and expensive handling
2926 of weak definition in ld.so.
2928 * Yet more PLT entries in libc.so have been removed. We finally arrived
2929 at the bare minimum. Startup times improved appropriately.
2931 * Support for the new Linux/x86 system call interface was added. The
2932 AT_SYSINFO auxiliary vector entry is recognized and handled.
2936 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2939 * iconv (the program and the interface) now accepts empty names (excluding
2940 options like //TRANSLIT) to mean "use charset of current locale".
2942 * localedef can now transliterate characters in strings which are not in
2943 the provided charmap. The information from the input locale is used.
2945 * Prelinking support was added for ELF targets. This requires additional
2946 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2948 * Read-only stdio streams now use mmap to speed up operation by eliminating
2949 copying and buffer underflows. To use add 'm' to the mode string of
2950 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2952 * The malloc functions were completely rewritten by Wolfram Gloger based
2953 on Doug Lea's malloc-2.7.0.c.
2955 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2956 implementation of regex.
2958 * Bruno Haible upgraded the iconv and locale implementation to support
2961 * Contents of the LC_* and LANG environment variables in the CEN style are
2962 not recognized anymore. It never was used. Change by Ulrich Drepper.
2964 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2965 thread-local storage (TLS) ABI on some platforms.
2966 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2968 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2969 EUC-JISX0213, and TSCII.
2971 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2972 BSD-compatible interface for getting all network interface addresses.
2973 Implementation for IPv4 by Roland McGrath.
2975 * Loading of locale data is faster due to the introduction of a locale
2976 archive. Implemented by Roland McGrath and Ulrich Drepper.
2978 * Startup times are significantly reduced by not using exported functions
2979 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2982 * Steven Munroe contributed a port to PowerPC64/Linux.
2986 * The Hurd now uses the GNU libio implementation of stdio rather than the
2987 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2989 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2990 with the same behavior as the Linux system call of the same name.
2994 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2995 128-bit long double format.
2997 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2998 IBM1161, and IBM1162 charsets.
3000 * Andreas Jaeger contributed a port to x86-64/Linux.
3002 * Peter Bruin contributed a port to PowerPC/Hurd.
3004 * libc functions using I/O streams now can handle wide-oriented streams
3007 * optimizations in the dynamic linker. Binaries created by recent binutils
3008 versions start up quicker due to reduced time spent on relocations.
3010 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3014 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3015 asinh, atanh, j0 for the 128-bit long double format.
3017 * Bruno Haible updated all the code handling Unicode in some form to
3018 support Unicode 3.1.
3020 * Speed of regex for single-byte locales is back to previous levels.
3021 Patch by Isamu Hasegawa.
3023 * Alpha, SPARC, and IA-64 now also using floating stacks.
3025 * Startup time of internationalized applications greatly improved through
3026 iconv cache. Use iconvconfig to generate the cache file.
3027 Contributed by Ulrich Drepper.
3029 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3030 severe performance problems. Patch by David Mosberger.
3032 * The Hurd port got a lot more functionality like AIO, various stdio
3033 extensions, etc. Mainly done by Roland McGrath.
3035 * mtrace can now lookup symbols in shared libraries.
3039 * Intel's IA-64 math library is largely integrated. It provides fast and
3040 accurate implementatations for most basic and standard math functions
3041 in float, double, and long double format.
3043 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3044 and asin for the 96-bit long double format and asin, log, tan for the
3045 128-bit long double format.
3047 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3048 The basic double functions exist today. Contributed by Abraham Ziv
3049 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3050 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3052 * An asynchronous name lookup library was added. The interface is designed
3053 after POSIX AIO. The proposal was circulated beforehand to get comments.
3054 No negative ones came in. Implemented by Ulrich Drepper.
3056 * Port to S390/64bit contributed by Martin Schwidefsky
3057 <schwidefsky@de.ibm.com>.
3059 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3060 of functions for Linux/IA-64.
3062 * The RPC code is now thread safe. Threads can now use the same service
3063 of different services at the same time. Patch by Eric Norum
3064 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3066 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3067 family of functions for Linux/S390.
3069 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3070 of functions for Linux/x86.
3072 * Port to Linux/CRIS contributed by Axis Communications.
3076 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3077 we can now check for namespace violations and missing declarations. The
3078 result is that almost all headers are now Unix-compliant (as defined in
3079 the upcoming XPG6). The negative side is that some programs might need
3080 corrections, too, if they depend on the incorrect form of the headers in
3081 previous versions which defined too many symbols and included too many
3084 * regex now handles multibyte character sets correctly.
3085 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3087 * iconv (the program) does now conform to the upcoming XPG6 and handles
3088 charmaps. Instead of the charset names the path of charmaps can be
3089 provided and the conversion happens based on this data.
3090 Contributed by Ulrich Drepper.
3092 * The locale program now provides detailed information about the installed
3097 only lists the names of the supported locales
3101 provides details such as country, language, and codeset name.
3102 Contributed by Ulrich Drepper.
3106 * The gencat program now parses the input file according to the charset
3107 selected by the LC_CTYPE category. This is important for stateful
3108 character sets. To make generating catalogs easier there is a way
3109 to overwrite the charset selected by the locale: before the first
3110 message or $ quote line the catalog can contain a line like
3112 $ codeset=ISO-8859-2
3114 to select the charset (ISO-8859-2 in this case).
3116 Implemented by Shinya Hanataka and Ulrich Drepper.
3118 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3119 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3120 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3122 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3123 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3125 * The LANGUAGE environment variable is now ignored unless the locale is
3126 changed from the default "C" locale.
3128 * The usual bug fixes.
3132 * Greg McGary added runtime support for bounds checking using gcc's
3133 new -fbounded-pointers option. ix86 target is complete. PowerPC
3136 * Thorsten Kukuk added secure mode support to nscd.
3138 * The Berkeley db libraries have been removed.
3140 Related, the nss_db module is now in a separate package since it
3141 obviously requires a database library being available.
3143 * Wide character I/O streams implemented by Ulrich Drepper.
3145 * Functions from the extended socket API added by Ulrich Drepper.
3147 * Functions feenableexcept and fedisableexcept to control the
3148 behaviour of individual exceptions have been added by Andreas Jaeger.
3150 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3152 * The resolver code has been updated from bind 8.2.3-T5B which supports
3153 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3156 This change could in some situations effect backward compatibility. Since
3157 now `_res' is a thread-local instead of a global variable, modifying it
3158 in one thread does not have any effect in other threads.
3160 The resolver library was also extended to allow IPv6 as the transport
3161 protocol for the requests. This work was done by Stig Venaas.
3163 * Compatibility code for K&R C compilers has been removed from the
3164 header files. A ISO C compiler is needed to use the library
3165 (conforming to either C89 or C99 standard).
3167 * Complete rewrite of the localedef program to support multibyte character
3168 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3169 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3170 Implemented by Ulrich Drepper.
3172 Bruno Haible significantly improved the generation and use of the data
3173 structures for the wide character tables.
3175 * Plural handling in gettext implemented by Ulrich Drepper.
3177 * The utmp daemon has been removed.
3179 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3181 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3184 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3186 * POSIX barriers implemented by Kaz Kylheku.
3188 * POSIX spawn function family implemented by Ulrich Drepper.
3190 * POSIX spinlocks are now available.
3192 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3194 * the configure option --enable-kernel=X.Y.Z allows to strip out
3195 compatibility for kernel versions before X.Y.Z. This is currently only
3196 implemented for Linux.
3198 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3199 field sin6_scope_id. This means that all programs using IPv6 should be
3200 recompiled. Don't expect binary compatibility with previous glibc
3203 * various conversion modules for IBM character sets contributed by
3206 * IA-64 port by Jes Sorensen and HJ Lu.
3220 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3222 * New xdr functions are added; some rpc functions are now 64bit clean.
3224 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3226 * Fixed known incompatibilities with glibc 2.0.
3228 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3230 * Optimized a number of functions (especially the ELF dynamic loader).
3232 * Update timezone data files.
3234 * lots of charmaps corrections
3236 * some new locale definitions and charmaps
3241 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3242 conform to POSIX member type requirements. Symbol versions have been
3243 adjusted as needed within the library, and for direct use by applications,
3244 but there is potential for problems if third-party libraries use
3245 struct timeval as part of their interface. This does not present
3246 a problem for X and other "essential" system libraries.
3248 * An additional locale model to support C++ Standard Library locale
3249 model and probably more was implemented by Ulrich Drepper.
3251 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3254 * Miles Bader provided the `argp' function family to support hierachical
3255 command line argument parsing, layered on top of getopt.
3257 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3259 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3262 * scanf recognizes the %a and %A format for scanning floating point numbers.
3264 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3265 information and interfaces for the available integer types.
3267 * about 130 new math functions were added to implement the ISO C9x math
3270 * the new header <complex.h> contains definitions of the complex math
3271 functions from ISO C 9X.
3273 * the new header <tgmath.h> defines generic macros to use complex or
3274 real valued functions.
3276 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3278 * Andreas Jaeger provided a test suite for the math library.
3280 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3282 * Ulrich Drepper added character set conversion functions (iconv).
3284 * Optimized string functions have been added.
3286 * The localedata addon is now part of glibc.
3288 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3290 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3291 daemon for NSS (nscd).
3293 Missing a better place here are some numbers on improvements. Under
3294 Linux 2.1.125 un-tar-ing the kernel sources takes
3298 using local files 12.19s 6.88s 22.91s
3300 using NIS 13.92s 8.91s 26.34s
3302 using NIS & nscd 10.37s 7.34s 25.30s
3304 using NIS+ 27.57s 30.37s 640.46s
3306 using NIS+ & nscd 10.25s 7.83s 26.51s
3308 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3310 Keep in mind that non-namelookup related operations dominate above times.
3311 It was just a common complain that using NIS+ unpacking the kernel is
3314 [1] The old nscd implementation is not available anymore since it was
3315 distributed with glibc up to version 2.0.98 and thus is now replaced.
3317 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3319 * Mark Kettenis provided a Hesiod NSS module.
3321 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3322 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3324 * rcmd can now handle netgroups (Dick Streefland).
3326 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3329 * Support for the IPv6 protocol has been added to the socket API, as per the
3330 latest draft standards.
3332 * Support for Linux 2.2 has been added.
3334 * Interface changes relative to the latest 2.0.x release:
3335 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3336 addseverity NEW: Unix98
3337 alphasort64 NEW: LFS
3338 argp_err_exit_status NEW: argp, GNU ext
3339 argp_error NEW: argp, GNU ext
3340 argp_failure NEW: argp, GNU ext
3341 argp_help NEW: argp, GNU ext
3342 argp_parse NEW: argp, GNU ext
3343 argp_program_bug_address NEW: argp, GNU ext
3344 argp_program_version NEW: argp, GNU ext
3345 argp_program_version_hook NEW: argp, GNU ext
3346 argp_state_help NEW: argp, GNU ext
3347 argp_usage NEW: argp, GNU ext
3348 authdes_create NEW: Secure RPC
3349 authdes_getucred NEW: Secure RPC
3350 authdes_pk_create NEW: Secure RPC
3351 backtrace NEW: GNU ext.
3352 backtrace_symbols NEW: GNU ext.
3353 backtrace_symbols_fd NEW: GNU ext.
3355 cacosf NEW: ISO C 9x
3356 cacosh NEW: ISO C 9x
3357 cacoshf NEW: ISO C 9x
3358 cacoshl NEW: ISO C 9x
3359 cacosl NEW: ISO C 9x
3366 casinf NEW: ISO C 9x
3367 casinh NEW: ISO C 9x
3368 casinhf NEW: ISO C 9x
3369 casinhl NEW: ISO C 9x
3370 casinl NEW: ISO C 9x
3372 catanf NEW: ISO C 9x
3373 catanh NEW: ISO C 9x
3374 catanhf NEW: ISO C 9x
3375 catanhl NEW: ISO C 9x
3376 catanl NEW: ISO C 9x
3377 cbc_crypt NEW: Secure RPC
3381 ccoshf NEW: ISO C 9x
3382 ccoshl NEW: ISO C 9x
3388 cimagf NEW: ISO C 9x
3389 cimagl NEW: ISO C 9x
3390 clearerr_locked REMOVED
3391 clntunix_create NEW: sunrpc ext
3393 clog10 NEW: ISO C 9x
3394 clog10f NEW: ISO C 9x
3395 clog10l NEW: ISO C 9x
3405 cprojf NEW: ISO C 9x
3406 cprojl NEW: ISO C 9x
3408 crealf NEW: ISO C 9x
3409 creall NEW: ISO C 9x
3414 csinhf NEW: ISO C 9x
3415 csinhl NEW: ISO C 9x
3418 csqrtf NEW: ISO C 9x
3419 csqrtl NEW: ISO C 9x
3423 ctanhf NEW: ISO C 9x
3424 ctanhl NEW: ISO C 9x
3426 des_setparity NEW: Secure RPC
3427 ecb_crypt NEW: Secure RPC
3428 endutxent NEW: Unix98
3430 exp10f NEW: ISO C 9x
3431 exp10l NEW: ISO C 9x
3435 fattach NEW: STREAMS
3436 fdetach NEW: STREAMS
3440 feclearexcept NEW: ISO C 9x
3441 fegetenv NEW: ISO C 9x
3442 fegetexceptflag NEW: ISO C 9x
3443 fegetround NEW: ISO C 9x
3444 feholdexcept NEW: ISO C 9x
3446 feraiseexcept NEW: ISO C 9x
3447 ferror_locked REMOVED
3448 fesetenv NEW: ISO C 9x
3449 fesetexceptflag NEW: ISO C 9x
3450 fesetround NEW: ISO C 9x
3451 fetestexcept NEW: ISO C 9x
3452 feupdateenv NEW: ISO C 9x
3453 fflush_locked REMOVED
3457 fgets_unlocked NEW: GNU ext.
3458 fileno_locked REMOVED
3470 fputc_locked REMOVED
3471 fputs_unlocked NEW: GNU ext.
3472 fread_unlocked NEW: GNU ext.
3477 fstatvfs NEW: Unix98
3481 ftruncate64 NEW: LFS
3483 fwrite_unlocked NEW: GNU ext.
3484 gai_strerror NEW: IPv6
3488 getchar_locked REMOVED
3490 getdate_err NEW: Unix98
3491 getdate_r NEW: GNU ext.
3493 getnameinfo NEW: IPv6
3494 getnetname NEW: Secure RPC
3495 getpmsg NEW: STREAMS
3496 getpt NEW: Unix98 PTY
3497 getrlimit64 NEW: LFS
3498 getutxent NEW: Unix98
3499 getutxid NEW: Unix98
3500 getutxline NEW: Unix98
3501 glob64 NEW: GNU ext.
3502 globfree64 NEW: GNU ext.
3503 gnu_get_libc_release NEW: GNU ext.
3504 gnu_get_libc_version NEW: GNU ext.
3505 grantpt NEW: Unix98 PTY
3506 host2netname NEW: Secure RPC
3508 iconv_close NEW: iconv
3509 iconv_open NEW: iconv
3510 if_freenameindex NEW: IPv6
3511 if_indextoname NEW: IPv6
3512 if_nameindex NEW: IPv6
3513 if_nametoindex NEW: IPv6
3514 in6addr_any NEW: IPv6
3515 in6addr_loopback NEW: IPv6
3516 inet6_isipv4mapped NEW: IPv6
3517 isastream NEW: STREAMS
3518 iswblank NEW: GNU ext.
3519 key_decryptsession NEW: Secure RPC
3520 key_decryptsession_pk NEW: Secure RPC
3521 key_encryptsession NEW: Secure RPC
3522 key_encryptsession_pk NEW: Secure RPC
3523 key_gendes NEW: Secure RPC
3524 key_get_conv NEW: Secure RPC
3525 key_secretkey_is_set NEW: Secure RPC
3526 key_setnet NEW: Secure RPC
3527 key_setsecret NEW: Secure RPC
3528 llrint NEW: ISO C 9x
3529 llrintf NEW: ISO C 9x
3530 llrintl NEW: ISO C 9x
3531 llround NEW: ISO C 9x
3532 llroundf NEW: ISO C 9x
3533 llroundl NEW: ISO C 9x
3538 lrintf NEW: ISO C 9x
3539 lrintl NEW: ISO C 9x
3540 lround NEW: ISO C 9x
3541 lroundf NEW: ISO C 9x
3542 lroundl NEW: ISO C 9x
3544 makecontext NEW: Unix98
3545 mempcpy NEW: GNU ext.
3548 modify_ldt NEW: kernel
3552 nearbyint NEW: ISO C 9x
3553 nearbyintf NEW: ISO C 9x
3554 nearbyintl NEW: ISO C 9x
3555 netname2host NEW: Secure RPC
3556 netname2user NEW: Secure RPC
3557 nexttoward NEW: ISO C 9x
3558 nexttowardf NEW: ISO C 9x
3559 nexttowardl NEW: ISO C 9x
3563 passwd2des NEW: Secure RPC
3565 pow10f NEW: GNU ext.
3566 pow10l NEW: GNU ext.
3569 printf_size NEW: GNU ext.
3570 printf_size_info NEW: GNU ext.
3571 profil_counter REMOVED
3572 pthread_mutexattr_getkind_np REPLACED
3573 pthread_mutexattr_setkind_np REPLACED
3574 ptsname NEW: Unix98 PTY
3575 ptsname_r NEW: Unix98 PTY
3577 putchar_locked REMOVED
3578 putgrent NEW: GNU ext.
3580 putpmsg NEW: STREAMS
3581 pututxline NEW: Unix98
3585 readdir64_r NEW: LFS
3586 remquo NEW: ISO C 9x
3587 remquof NEW: ISO C 9x
3588 remquol NEW: ISO C 9x
3590 roundf NEW: ISO C 9x
3591 roundl NEW: ISO C 9x
3593 scalbln NEW: ISO C 9x
3594 scalblnf NEW: ISO C 9x
3595 scalblnl NEW: ISO C 9x
3597 sendfile NEW: kernel
3598 setcontext NEW: Unix98
3599 setrlimit64 NEW: LFS
3600 setutxent NEW: Unix98
3602 sigignore NEW: Unix98
3603 sigqueue NEW: POSIX.1b
3604 sigrelse NEW: Unix98
3605 sigset NEW: POSIX.1b
3606 sigtimedwait NEW: POSIX.1b
3607 sigwaitinfo NEW: POSIX.1b
3608 sincos NEW: GNU ext.
3609 sincosf NEW: GNU ext.
3610 sincosl NEW: GNU ext.
3614 strcasestr NEW: GNU ext.
3615 strtoimax NEW: ISO C 9x
3616 strtoumax NEW: ISO C 9x
3617 strverscmp NEW: GNU ext.
3618 svcauthdes_stats NEW: Secure RPC
3619 svcunix_create NEW: sunrpc etx
3620 svcunixfd_create NEW: sunrpc ext
3621 swapcontext NEW: Unix98
3622 tcgetsid NEW: Unix98 PTY
3623 tdestroy NEW: GNU ext.
3624 tgamma NEW: ISO C 9x
3625 tgammaf NEW: ISO C 9x
3626 tgammal NEW: ISO C 9x
3630 truncf NEW: ISO C 9x
3631 truncl NEW: ISO C 9x
3633 unlockpt NEW: Unix98 PTY
3634 updwtmpx NEW: Unix98
3635 user2netname NEW: Secure RPC
3636 utmpxname NEW: Unix98
3637 versionsort NEW: GNU ext.
3638 versionsort64 NEW: GNU ext.
3640 wcscasecmp NEW: GNU ext.
3641 wcsncasecmp NEW: GNU ext.
3642 wcsnlen NEW: GNU ext.
3643 wcstoimax NEW: ISO C 9x
3644 wcstoll NEW: ISO C 9x
3645 wcstoull NEW: ISO C 9x
3646 wcstoumax NEW: ISO C 9x
3648 wordexp NEW: POSIX.2
3649 wordfree NEW: POSIX.2
3650 write_profiling REMOVED
3651 xdecrypt NEW: Secure RPC
3652 xdr_authdes_cred NEW: Secure RPC
3653 xdr_authdes_verf NEW: Secure RPC
3654 xdr_cryptkeyarg NEW: Secure RPC
3655 xdr_cryptkeyarg2 NEW: Secure RPC
3656 xdr_cryptkeyres NEW: Secure RPC
3657 xdr_getcredres NEW: Secure RPC
3658 xdr_key_netstarg NEW: Secure RPC
3659 xdr_key_netstres NEW: Secure RPC
3660 xdr_keybuf NEW: Secure RPC
3661 xdr_keystatus NEW: Secure RPC
3662 xdr_netnamestr NEW: Secure RPC
3663 xdr_sizeof NEW: Secure RPC
3664 xdr_unixcred NEW: sunrpc ext
3665 xencrypt NEW: Secure RPC
3666 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3677 * inet_ntoa is thread-safe
3679 * updwtmp is moved from libutil to libc
3681 * rewrite of cbrt function
3683 * update of timezone data
3697 * add atoll function
3699 * fix complex problems in Berkeley DB code
3701 * fix math functions
3705 * fixed lots of header problems (especially Linux/GNU specific)
3707 * dynamic loader preserves all registers
3709 * Roland McGrath provided support for handling of auxiliary objects in
3710 the ELF dynamic loader.
3712 * support for parallel builds is improved
3716 * GNU extensions are no longer declared by default. To enable them you
3717 must define the macro `_GNU_SOURCE' in your program or compile with
3720 * The library has changed from using GNU ld symbol aliases to using weak
3721 symbols where available. The ELF object file format supports weak
3722 symbols; GNU ld also supports weak symbols in the a.out format. (There
3723 is also now support for other GNU ld extensions in ELF. Use the
3724 `--with-elf' option to configure to indicate you have ELF, and
3725 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3726 of many files which contained only symbol aliases, reducing the size of
3727 the source and the compiled library; many other files were renamed to
3728 less cryptic names previously occupied by the symbol alias files.
3729 There is a new header file <elf.h> for programs which operate on
3730 files in the ELF format.
3732 * Converted to Autoconf version 2, so `configure' has more options.
3733 Run `configure --help' to see the details.
3735 * The library can now be configured to build profiling, highly-optimized
3736 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3737 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3738 `configure' enable building these extra libraries. The shared library is
3739 built by default when using both ELF and GNU ld. When shared libraries
3740 are enabled, the new library `-ldl' is available for arbitrary run-time
3741 loading of shared objects; its interface is defined in <dlfcn.h>. The
3742 new header file <link.h> gives access to the internals of the run-time
3743 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3744 application of same name on other systems and it provides information
3745 about dynamically linked binaries.
3747 * The C library now provides the run-time support code for profiling
3748 executables compiled with `-pg'. Programs can control the profiling code
3749 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3750 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3751 the support for this file format was contributed by David Mosberger-Tang.
3753 * The math code has been replaced with a math library based on fdlibm from
3754 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3755 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3756 functions. The math functions now reside in a separate library, so
3757 programs using them will need to use `-lm' their linking commands.
3759 * John C. Bowman contributed optimized ix87 assembler inline functions.
3761 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3762 mechanism similar to that found in Solaris 2. This is now used for the
3763 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3764 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3765 file controls what services are used for each individual database. This
3766 works by loading shared libraries with names specified in `nsswitch.conf',
3767 so service modules can be changed or added at any time without even
3768 relinking any program. Currently there are the file, db, and NIS based
3769 NSS services available.
3771 * The new functions `strtoq' and `strtouq' parse integer values from
3772 strings, like `strtol' and `strtoul', but they return `long long int' and
3773 `unsigned long long int' values, respectively (64-bit quantities).
3775 * The new functions `strtof' and `strtold' parse floating-point values from
3776 strings, like `strtod', but they return `float' and `long double' values,
3777 respectively (on some machines `double' and `long double' are the same).
3779 * Ulrich Drepper has contributed new implementations of the floating-point
3780 printing and reading code used in the `printf' family of functions and
3781 `strtod', `strtof', and `strtold'. These new functions are perfectly
3782 accurate, and much faster than the old ones.
3784 * The implementation of the POSIX locale model was completely rewritten by
3785 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3786 to compile the POSIX locale definition.
3788 * The former dummy implementations of the strcoll and strxfrm function are
3789 now replaced by fully functional code contributed by Ulrich Drepper. The
3790 collation information comes from the POSIX locale definitions.
3792 * The new header <langinfo.h> defines an interface for accessing
3793 various locale-dependent data (using the locale chosen with `setlocale').
3795 * Ulrich Drepper has contributed a new suite of functions for operation on
3796 wide-character and multibyte-character strings, in <wchar.h>;
3797 and classification and case conversion of wide characters, in <wctype.h>.
3798 These new functions are conforming to the ISO C, Amendement 1 specification.
3800 * There is now a second implementation of the standard I/O library available.
3801 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3802 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3804 * You can now use positional parameter specifications in format strings
3805 for the `printf' and `scanf' families of functions. For example,
3806 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3807 ``Number 6, Mr Jones''. This is mainly useful when providing different
3808 format strings for different languages, whose grammars may dictate
3809 different orderings of the values being printed. To support this
3810 feature, the interface for `register_printf_handler' has changed; see
3811 the header file <printf.h> for details.
3813 * The `printf' and `scanf' families of functions now understand a new
3814 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3815 to group numbers as indicated by the locale; for `scanf' and friends, this
3816 says to accept as valid only a number with all the proper grouping
3817 separators in the right places. In the default "C" locale, numbers are
3818 not grouped; but locales for specific countries will define the usual
3819 conventions (i.e. separate thousands with `,' in the US locale).
3821 * The pgrp functions have been regularized, slightly incompatibly but much
3822 less confusingly. The core functions are now `getpgid' and `setpgid',
3823 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3824 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3825 provided for compatibility. There is no longer an incompatible `getpgrp'
3826 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3828 * The new header file <fts.h> and suite of functions simplify programs that
3829 operate on directory trees. This code comes from 4.4 BSD.
3831 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3832 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3833 NSS scheme used in glibc.
3835 * There is a new malloc debugging hook `__memalign_hook'.
3837 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3838 `unsigned int' in <sys/types.h>. These are for compatibility only and
3839 their use is discouraged.
3841 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3842 done differently, so that it works even without GNU ld.
3844 * New function `euidaccess' checks allowed access to a file like `access',
3845 but using the effective IDs instead of the real IDs.
3847 * The time zone data files have been updated for the latest and greatest
3848 local time conventions of the countries of the world.
3850 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3853 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3854 for formatting floating-point numbers. They are provided only for
3855 compatibility; new programs should use `sprintf' instead. There are
3856 also equivalent function for the `long double' floating-point type and
3857 all functions also exist in a reentrant form.
3859 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3860 functions for maintaining the login-record files (primarily of use to
3861 system programs such as `login'), and convenient functions for
3862 allocating and initializing a pseudo-terminal (pty) device.
3864 * Ulrich Drepper has contributed new support for System V style
3865 shared memory and IPC on systems that support it.
3867 * Ulrich Drepper has contributed several miscellaneous new functions found
3868 in System V: The `hsearch' family of functions provide an effective
3869 implementation of hash tables; `a64l' and `l64a' provide a very simple
3870 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3873 * Ulrich Drepper has contributed new reentrant counterparts for the
3874 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3876 * Ulrich Drepper has contributed new, highly-optimized versions of several
3877 string functions for the i486/Pentium family of processors.
3879 * Ulrich Drepper has updated the Linux-specific code, based largely
3880 on work done in Hongjiu Lu's version of GNU libc for Linux.
3881 The GNU library now supports Linux versions 2.0.10 and later,
3882 using the ELF object file format (i[3456]86-*-linux).
3884 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3886 * David Mosberger-Tang and Richard Henderson have ported the C library
3887 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3888 dynamic linking support for ELF/Alpha.
3890 * Richard Henderson contributed several Alpha optimized assembler function
3891 for arithmetic and string handling.
3893 * Ulrich Drepper has contributed a new set of message catalog functions to
3894 support multiple languages using the <libintl.h> interface, for use with
3895 his new package GNU gettext. Translation volunteers have contributed
3896 catalogs of the library's messages in Spanish, German, and Korean.
3898 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3899 program and the `catgets' function for reading the catalog files it
3900 creates. (The <libintl.h> interface is preferred; we include the
3901 <nl_types.h> interface using `catgets' only for source compatibility with
3902 programs already written to use it.)
3904 * New header file <values.h> gives SVID-compatible names for <limits.h>
3907 * Various new macros, declarations, and small header files for compatibility
3910 * New function `group_member' is a convenient way to check if a process has
3911 a given effective group ID.
3913 * When using GCC 2.7 and later, the socket functions are now declared in a
3914 special way so that passing an argument of type `struct sockaddr_in *',
3915 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3916 `struct sockaddr *' type, does not generate a type-clash warning.
3918 * New function `error' declared in header file <error.h> is a convenient
3919 function for printing error messages and optionally exiting; this is the
3920 canonical function used in GNU programs. The new functions `err', `warn',
3921 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3922 doing the same thing.
3924 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3925 POSIX.2 `glob' function to do ~ and {...} expansion.
3927 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3928 `clearenv' which is used in POSIX.9 is also available.
3930 * New function `getsid' returns session ID number on systems that support it.
3932 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3933 files <db.h> and <mpool.h> provide a rich set of functions for several
3934 types of simple databases stored in memory and in files, and <ndbm.h> is
3935 an old `ndbm'-compatible interface using the `db' functions. Link with
3936 `-ldb' to get these functions.
3938 * New macro `strdupa' copies a string like `strdup', but uses local stack
3939 space from `alloca' instead of dynamic heap space from `malloc'.
3941 * New function `strnlen' is like `strlen' but searches only a given maximum
3942 number of characters for the null terminator. `stpncpy', `strndup' and
3943 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3946 * New function `statfs' in header <sys/statfs.h>.
3948 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3949 provide convenient functions for operating on blocks of null-terminated
3952 * A new suite of functions in <utmp.h> handle all the details of reading
3953 and writing the utmp file.
3955 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3958 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3959 completely POSIX compliant and also implemented the extended functionality
3960 to handle alternate digit representation and alternate era date formats.
3962 * Ulrich Drepper provided an implementation of the `strptime' function
3963 defined in XPG4.2 which transforms a string into a `struct tm' value.
3965 * Paul Eggert provided the tzselect shell script as part of the timezone
3966 code. The shell script makes it easy to select the correct timezone
3969 * The implementation of the malloc family of functions is completely replaced
3970 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3971 The implementation uses the mmap function (if available) and it is
3972 optimized for the use in multi threaded programs.
3974 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3975 functions. This new functionality is usable by specifying a special
3976 salt string and it is compatible with implementation on *BSD systems.
3978 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3979 `getsubopt' to handle second level command line options, `bsd_signal'
3980 to access BSD style `signal' functionality, the obsolete `regexp' style
3983 * the `lchown' function is available on system which support this
3986 * The implementation of the shadow password handling function was contributed
3989 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3991 * POSIX.1g support was added. The <sys/select.h> header is available,
3992 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3993 implementation of `getaddrinfo'.
3997 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3999 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4000 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4002 * The new function `daemon' from 4.4 BSD is useful for server programs that
4003 want to put themselves in the background.
4005 * Joel Sherrill has contributed support for several standalone boards that
4006 run without an operating system.
4008 * `printf', `scanf' and friends now accept a `q' type modifier for long
4009 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4011 * All of the code taken from BSD (notably most of the math and networking
4012 routines) has been updated from the BSD 4.4-Lite release.
4014 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4016 * The new functions `getdomainname' and `setdomainname' fetch or change the
4017 YP/NIS domain name. These are system calls which exist on systems which
4020 * The time zone data files have been updated for the latest international
4023 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4024 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4028 * The C library now includes support for Sun RPC, from Sun's free
4029 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4030 are included. (There is still no support for YP.)
4032 * Tom Quinn has contributed a port of the C library to SGI machines running
4033 Irix 4 (mips-sgi-irix4).
4035 * The new `lockf' function is a simplified interface to the locking
4036 facilities of `fcntl', included for compatibility.
4038 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4040 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4042 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4045 * The error message printed by `assert' for a failed assertion now includes
4046 the name of the program (if using GNU ld) and the name of the calling
4047 function (with versions of GCC that support this).
4049 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4051 * The library now includes the <sys/mman.h> header file and memory
4052 management functions `mmap', `munmap', `mprotect', `msync', and
4053 `madvise', on systems that support those facilities.
4055 * The interface for `mcheck' has changed slightly: the function called to
4056 abort the program when an allocation inconsistency is detected now takes
4057 an argument that indicates the type of failure. The new function
4058 `mprobe' lets you request a consistency check for a particular block at
4059 any time (checks are normally done only when you call `free' or `realloc'
4062 * It is now possible to easily cross-compile the C library, building on one
4063 system a library to run on another machine and/or operating system. All
4064 you need to do is set the variable `HOST_CC' in `configparms' to the
4065 native compiler for programs to run on the machine you are building on (a
4066 few generator programs are used on Unix systems); set `CC' to the
4069 * The new function `fexecve' (only implemented on the GNU system) executes
4070 a program file given a file descriptor already open on the file.
4074 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4075 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4077 * You can set the variable `libprefix' in `configparms' to specify a prefix
4078 to be prepended to installed library files; this makes it easy to install
4079 the GNU C library to be linked as `-lgnuc' or whatever.
4081 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4082 copies a limited number of characters from a string, and returns the
4083 address of the last character written.
4085 * You no longer need to check for whether the installed `stddef.h' is
4086 compatible with the GNU C library. configure now checks for you.
4088 * You can now define a per-stream `fileno' function to convert the
4089 stream's cookie into an integral file descriptor.
4091 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4092 allocates zero bytes of storage, and returns a unique pointer which
4093 you can pass to `realloc' or `free'. The behavior is undefined if
4094 you dereference this pointer.
4096 * The C library now runs on Sony NEWS m68k machines running either
4097 NewsOS 3 or NewsOS 4.
4099 * The new `syscall' function is a system-dependent primitive function
4100 for invoking system calls. It has the canonical behavior on Unix
4101 systems, including unreliable return values for some calls (such as
4102 `pipe', `fork' and `getppid').
4104 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4105 to `EAGAIN', which is the preferred name. On systems whose kernels
4106 use two distinct codes, the C library now translates EWOULDBLOCK to
4107 EAGAIN in every system call function.
4111 * The GNU C Library Reference Manual is now distributed with the library.
4112 `make dvi' will produce a DVI file of the printed manual.
4113 `make info' will produce Info files that you can read on line using C-h i
4114 in Emacs or the `info' program.
4115 Please send comments on the manual to bug-glibc-manual@gnu.org.
4117 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4119 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4121 * Jason Merrill has contributed a port to the Sequent Symmetry running
4122 Dynix version 3 (i386-sequent-dynix).
4124 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4125 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4127 * New function `memory_warnings' lets you arrange to get warnings when
4128 malloc is running out of memory to allocate, like Emacs gives you.
4130 * The C library now contains the relocating allocator used in Emacs 19 for
4131 its editing buffers. This allocator (ralloc) minimizes allocation
4132 overhead and fragmentation by moving allocated regions around whenever it
4133 needs to. You always refer to a ralloc'd region with a "handle" (a
4134 pointer to a pointer--an object of type `void **').
4136 * There is a new `printf' format: `%m' gives you the string corresponding
4137 to the error code in `errno'.
4139 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4140 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4141 pass, the `a' modifier says to fill in a `char **' you pass with a
4144 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4145 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4146 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4148 * `mkstemp' is a traditional Unix function to atomically create and open a
4149 uniquely-named temporary file.
4153 * The standard location for the file that says what the local timezone is
4154 has changed again. It is now `/usr/local/etc/localtime' (or more
4155 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4157 * The distribution no longer contains any files with names longer than 14
4160 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4161 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4163 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4167 * The configuration process has changed quite a bit. The `configure'
4168 script is now used just like the configuration scripts for other GNU
4169 packages. The `sysdeps' directory hierarchy is much rearranged.
4170 The file `INSTALL' explains the new scheme in detail.
4172 * The header files no longer need to be processed into ANSI C and
4173 traditional C versions. There is just one set of files to install, and
4174 it will work with ANSI or old C compilers (including `gcc -traditional').
4176 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4177 MIPS DECStation running Ultrix 4.
4179 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4180 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4181 made itself into a shared library.
4183 * Yet further improved support for the i386, running 4.3 BSD-like systems
4184 (such as Mach 3 with the Unix single-server), or System V.
4186 * New function `strncasecmp' to do case-insensitive string comparison
4187 with limited length.
4189 * New function `strsep' is a reentrant alternative to `strtok'.
4191 * New functions `scandir' and `alphasort' for searching directories.
4193 * New function `setenv' is a better interface to `putenv'.
4195 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4196 function for traversing a directory tree.
4198 * The GNU obstack package is now also part of the C library.
4199 The new function `open_obstack_stream' creates a stdio stream that
4200 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4201 formatted output directly to an obstack.
4203 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4204 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4206 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4208 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4209 things to your strings.
4211 * There are some new test programs: test-fseek, testmb, and testrand.
4213 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4214 These ports are not finished, but are a good starting place for really
4215 supporting those systems.
4217 * `/etc/localtime' is now the standard location for the file that says what
4218 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4219 This follows the general principle that `/etc' is the place for all local
4220 configuration files.
4222 * The C library header files now use `extern "C"' when used by the C++
4223 compiler, so the C library should now work with C++ code.
4225 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4226 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4229 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4230 function which uses the `merge sort' algorithm, and is said to be
4231 significantly faster than the old GNU `qsort' function. Merge sort is
4232 now the standard `qsort' function. The new algorithm can require a lot
4233 of temporary storage; so, the old sorting function is called when the
4234 required storage is not available.
4236 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4237 provides the Unix `crypt' function, plus some other entry points.
4239 * The code and header files taken from 4.4 BSD have been updated with the
4240 latest files released from Berkeley.
4242 ----------------------------------------------------------------------
4243 Copyright information:
4245 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4247 Permission is granted to anyone to make or distribute verbatim copies
4248 of this document as received, in any medium, provided that the
4249 copyright notice and this permission notice are preserved,
4250 thus giving the recipient permission to redistribute in turn.
4252 Permission is granted to distribute modified versions
4253 of this document, or of portions of it,
4254 under the above conditions, provided also that they
4255 carry prominent notices stating who last changed them.
4258 version-control: never