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 Deprecated and removed features, and other changes affecting compatibility:
34 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
35 defined by <sys/ptrace.h>.
37 * libm no longer supports SVID error handling (calling a user-provided
38 matherr function on error) or the _LIB_VERSION variable to control error
39 handling. (SVID error handling and the _LIB_VERSION variable still work
40 for binaries linked against older versions of the GNU C Library.) The
41 libieee.a library is no longer provided. math.h no longer defines struct
42 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
43 TLOSS, PLOSS and HUGE.
45 * The libm functions pow10, pow10f and pow10l are no longer supported for
46 new programs. Programs should use the standard names exp10, exp10f and
47 exp10l for these functions instead.
49 * The mcontext_t type is no longer the same as struct sigcontext. On
50 platforms where it was previously the same, this changes the C++ name
51 mangling for interfaces involving this type.
53 Changes to build and runtime requirements:
55 [Add changes to build and runtime requirements here]
57 Security related changes:
59 CVE-2009-5064: The ldd script would sometimes run the program under
60 examination directly, without preventing code execution through the
61 dynamic linker. (The glibc project disputes that this is a security
62 vulnerability; only trusted binaries must be examined using the ldd
65 The following bugs are resolved with this release:
67 [The release manager will add the list generated by
68 scripts/list-fixed-bugs.py just before the release.]
75 * A per-thread cache has been added to malloc. Access to the cache requires
76 no locks and therefore significantly accelerates the fast path to allocate
77 and free small amounts of memory. Refilling an empty cache requires locking
78 the underlying arena. Performance measurements show significant gains in a
79 wide variety of user workloads. Workloads were captured using a special
80 instrumented malloc and analyzed with a malloc simulator. Contributed by
81 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
83 * Unicode 10.0.0 Support: Character encoding, character type info, and
84 transliteration tables are all updated to Unicode 10.0.0, using
85 generator scripts contributed by Mike FABIAN (Red Hat).
86 These updates cause user visible changes, especially the changes in
87 wcwidth for many emoji characters cause problems when emoji sequences
88 are rendered with pango, see for example:
89 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
91 * Collation of Hungarian has been overhauled and is now consistent with "The
92 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
95 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
97 - The GNU C Library will now detect when /etc/resolv.conf has been
98 modified and reload the changed configuration. The new resolver option
99 “no-reload” (RES_NORELOAD) disables this behavior.
101 - The GNU C Library now supports an arbitrary number of search domains
102 (configured using the “search” directive in /etc/resolv.conf);
103 previously, there was a hard limit of six domains. For backward
104 compatibility, applications that directly modify the ‘_res’ global
105 object are still limited to six search domains.
107 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
108 Library will now randomly pick a name server from the configuration as a
109 starting point. (Previously, the second name server was always used.)
111 * The tunables feature is now enabled by default. This allows users to tweak
112 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
114 * New function reallocarray, which resizes an allocated block (like realloc)
115 to the product of two sizes, with a guaranteed clean failure upon integer
116 overflow in the multiplication. Originally from OpenBSD, contributed by
117 Dennis Wölfing and Rüdiger Sonderfeld.
119 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
120 These are extended versions of preadv and pwritev, respectively, taking an
121 additional flags argument. The set of supported flags depends on the
122 running kernel; full support currently requires kernel 4.7 or later.
124 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
125 create a new session ID for the spawned process. This feature is
126 scheduled to be added to the next major revision of POSIX; for the time
127 being, it is available under _GNU_SOURCE.
129 * errno.h is now safe to use from C-preprocessed assembly language on all
130 supported operating systems. In this context, it will only define the
131 Exxxx constants, as preprocessor macros expanding to integer literals.
133 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
134 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
135 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
136 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
138 To compile programs that use this feature, the compiler must support
139 128-bit floating point with the type name _Float128 (as defined by TS
140 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
141 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
142 must be defined to make the new interfaces visible.
144 The new functions and macros correspond to those present for other
145 floating-point types (except for a few obsolescent interfaces not
146 supported for the new type), with F128 or f128 suffixes; for example,
147 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
148 printf or scanf formats for the new type; the strfromf128 and strtof128
149 interfaces should be used instead.
151 Deprecated and removed features, and other changes affecting compatibility:
153 * The synchronization that pthread_spin_unlock performs has been changed to
154 now be equivalent to a C11 atomic store with release memory order to the
155 spin lock's memory location. Previously, several (but not all)
156 architectures used stronger synchronization (e.g., containing what is
157 often called a full barrier). This change can improve performance, but
158 may affect odd fringe uses of spin locks that depend on the previous
159 behavior (e.g., using spin locks as atomic variables to try to implement
160 Dekker's mutual exclusion algorithm).
162 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
165 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
166 will only be built and installed when the GNU C Library is configured with
167 --enable-obsolete-rpc. This allows alternative RPC implementations, such
168 as TIRPC or rpcsvc-proto, to be used.
170 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
171 libnss_compat, are deprecated, and will not be built or installed by
174 The NIS(+) support library, libnsl, is also deprecated. By default, a
175 compatibility shared library will be built and installed, but not headers
176 or development libraries. Only a few NIS-related programs require this
177 library. (In particular, the GNU C Library has never required programs
178 that use 'gethostbyname' to be linked with libnsl.)
180 Replacement implementations based on TIRPC, which additionally support
181 IPv6, are available from <https://github.com/thkukuk/>. The configure
182 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
183 name service modules, to be built and installed.
185 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
186 support is enabled, the configured recursive resolver must support EDNS.
187 (Responding to EDNS-enabled queries with responses which are not
188 EDNS-enabled is fine, but FORMERR responses are not.)
190 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
191 servers have not supported this opcode for a long time.
193 * The _res_opcodes variable has been removed from libresolv. It had been
194 exported by accident.
196 * <string.h> no longer includes inline versions of any string functions,
197 as this kind of optimization is better done by the compiler. The macros
198 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
200 * The nonstandard header <xlocale.h> has been removed. Most programs should
201 use <locale.h> instead. If you have a specific need for the definition of
202 locale_t with no other declarations, please contact
203 libc-alpha@sourceware.org and explain.
205 * The obsolete header <sys/ultrasound.h> has been removed.
207 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
209 * The obsolete function cfree has been removed. Applications should use
212 * The stack_t type no longer has the name struct sigaltstack. This changes
213 the C++ name mangling for interfaces involving this type.
215 * The ucontext_t type no longer has the name struct ucontext. This changes
216 the C++ name mangling for interfaces involving this type.
218 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
219 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
220 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
221 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
222 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
223 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
224 fpregset_t type no longer has the name struct fpu. This changes the C++
225 name mangling for interfaces involving those types.
227 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
228 synced with the kernel:
230 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
231 are not supported on this architecture and have been removed.
233 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
234 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
235 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
237 Programs that assume the GET/SETREGS ptrace requests are universally
238 available will now fail to build, instead of malfunctioning at runtime.
240 Changes to build and runtime requirements:
242 * Linux kernel 3.2 or later is required at runtime, on all architectures
243 supported by that kernel. (This is a change from version 2.25 only for
246 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
248 * On most architectures, GCC 4.9 or later is required to build the GNU C
249 Library. On powerpc64le, GCC 6.2 or later is required.
251 Older GCC versions and non-GNU compilers are still supported when
252 compiling programs that use the GNU C Library. (We do not know exactly
253 how old, and some GNU extensions to C may be _de facto_ required. If you
254 are interested in helping us make this statement less vague, please
255 contact libc-alpha@sourceware.org.)
257 Security related changes:
259 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
260 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
262 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
263 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
265 * Avoid printing a backtrace from the __stack_chk_fail function since it is
266 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
269 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
270 fixed (CVE-2017-12133).
272 The following bugs are resolved with this release:
274 [984] network: Respond to changed resolv.conf in gethostbyname
275 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
276 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
278 [12189] libc: __stack_chk_fail should not attempt a backtrace
280 [14096] time: Race condition on timezone/tst-timezone.out
281 [14172] localedata: az_IR: new locale
282 [14995] build: glibc fails to build if gold is the default linker, even if
284 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
285 [16637] network: inet_pton function is accepting IPv6 with bad format
286 [16640] string: string/strtok.c: undefined behaviour inconsistent between
287 x86 and other generic code
288 [16875] localedata: ko_KR: fix lang_name
289 [17225] localedata: ar_SY: localized month names for May and June are
291 [17297] localedata: da_DK: wrong date_fmt string
292 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
293 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
294 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
295 [19066] localedata: ar_SA abbreviated day and month names are in English
296 [19569] network: resolv: Support an arbitrary number of search domains
297 [19570] network: Implement random DNS server selection in the stub
299 [19838] locale: localedef fails on PA-RISC
300 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
301 order of 0D36 and 0D37
302 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
304 [20098] libc: FAIL: debug/backtrace-tst on hppa
305 [20257] network: sunrpc: clntudp_call does not enforce timeout when
307 [20275] localedata: locale day/abday/mon/abmon should not have trailing
309 [20313] localedata: Update locale data to Unicode 9.0
310 [20424] manual: Document how to provide a malloc replacement
311 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
313 [20686] locale: Add el_GR@euro to SUPPORTED.
314 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
315 failures consistently
316 [21015] dynamic-link: Document and fix --enable-bind-now
317 [21016] nptl: pthread_cond support is broken on hppa
318 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
319 frame-pointer on i386
320 [21049] libc: segfault in longjmp_chk() due to clobbered processor
322 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
323 [21088] libc: Build fails with --enable-static-nss
324 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
325 generic c code is used
326 [21109] libc: Tunables broken on big-endian
327 [21112] math: powf has large ulp errors with base close to 1 and exponent
329 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
331 [21120] malloc: glibc malloc is incompatible with GCC 7
332 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
334 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
335 (-0.0) when linking with -lieee
336 [21171] math: log10, log2 and lgamma return incorrect results
337 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
338 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
339 [21207] localedata: ce_RU: update weekdays from CLDR
340 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
341 [21217] localedata: Update months from CLDR-31
342 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
343 [21243] libc: support_delete_temp_file should issue warning for failed
345 [21244] libc: support resolv_test_start() socket fd close should be
347 [21253] libc: localedef randomly segfaults when using -fstack-check due to
348 new posix_spawn implementation
349 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
350 leads to lower CPU frequency
351 [21259] libc: [alpha] termios.h missing IXANY for POSIX
352 [21261] libc: [sparc64] bits/setjmp.h namespace
353 [21267] network: [mips] bits/socket.h IOC* namespace
354 [21268] libc: [alpha] termios.h NL2, NL3 namespace
355 [21270] libc: mmap64 silently truncates large offset values
356 [21275] libc: posix_spawn always crashes on ia64 now
357 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
358 [21280] math: [powerpc] logbl for POWER7 return incorrect results
359 [21289] libc: Incorrect declaration for 32-bit platforms with
360 _FILE_OFFSET_BITS=64 causes build error
361 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
362 not support gethostbyname4_r
363 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
365 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
367 [21340] libc: Support POSIX_SPAWN_SETSID
368 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
369 [21359] network: ns_name_pack needs additional byte in destination buffer
370 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
371 against fragmentation attacks (CVE-2017-12132)
372 [21369] network: resolv: Remove EDNS fallback
373 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
375 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
376 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
377 [21393] stdio: Missing dup3 error check in freopen, freopen64
378 [21396] libc: Use AVX2 memcpy/memset on Skylake server
379 [21399] localedata: Bad description for U00EC in
380 localedata/charmaps/CP1254
381 [21411] malloc: realloc documentation error
382 [21426] network: sys/socket.h uio.h namespace
383 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
384 [21445] libc: signal.h bsd_signal namespace
385 [21455] network: Network headers stdint.h namespace
386 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
388 [21475] network: resolv: Overlong search path is truncated mid-label
389 [21511] libc: sigstack namespace
390 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
391 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
392 bits/syscall.h: No such file or directory
393 [21517] libc: struct sigaltstack namespace
394 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
395 [21533] localedata: Update locale data to Unicode 10.0
397 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
398 junk at end of line, first unrecognized character is `@'
399 [21538] libc: SIG_HOLD missing for XPG4
400 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
401 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
402 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
403 [21543] libc: sigevent namespace
404 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
405 size for general purpose registers in mcontext_t structure
406 [21550] libc: sigwait namespace
407 [21552] libc: XPG4 bsd_signal namespace
408 [21554] libc: sigpause namespace
409 [21560] libc: sys/wait.h signal.h namespace
410 [21561] libc: waitid namespace
411 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
412 '/usr/include/stdlib.h': Permission denied
413 [21575] libc: sys/wait.h missing struct rusage definition
414 [21584] libc: sigaltstack etc namespace
415 [21597] libc: siginterrupt namespace
416 [21607] math: hppa: FAIL: math/test-tgmath
417 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
419 [21622] libc: [tile] missing SA_* for POSIX.1:2008
420 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
421 stack and heap (CVE-2017-1000366)
422 [21625] libc: wait3 namespace
423 [21654] nss: Incorrect pointer alignment in NSS group merge result
425 [21657] network: Parse interface zone id for node-local multicast
426 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
428 [21666] libc: .symver is used on common symbol
429 [21668] network: resolv: res_init cross-thread broadcast introduces race
431 [21687] math: tgmath.h totalorder, totalordermag return type
432 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
434 [21696] libc: Incorrect assumption of of __cpu_mask in
435 posix/sched_cpucount.c
436 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
437 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
439 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
440 leading to relocation crash
441 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
442 `resolv_conf_matches (resp, conf)' failed.
443 [21710] localedata: Added Samoan language locale for Samoa
444 [21711] localedata: Pashto yesstr/nostr locale are missing
445 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
446 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
447 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
448 [21723] localedata: yesstr/nostr missing for Chinese language locale
449 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
451 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
453 [21728] localedata: New Locale for Tongan language
454 [21729] localedata: incorrect LC_NAME fields for hi_IN
455 [21733] localedata: yesstr and nostr are missing for zh_HK
456 [21734] localedata: Missing yesstr and nostr are for kw_GB
457 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
458 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
459 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
460 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
462 [21744] libc: Tests failing on --enable-tunables --enable-stack-
464 [21749] localedata: Wrong abbreviated day name (“abday”) for
466 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
467 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
468 [21759] localedata: missing yesstr and nostr for Tigrinya
469 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
470 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
471 [21767] localedata: Missing Bislama locales
472 [21768] localedata: Missing yesstr and nostr for aa_ET
473 [21770] localedata: Missing Field in li_NL
474 [21778] nptl: Robust mutex may deadlock
475 [21779] libc: MicroBlaze segfaults when loading libpthread
476 [21783] localedata: Fix int_select international_call_prefixes
477 [21784] localedata: Inconsistency in country_isbn
478 [21788] localedata: Missing Country Postal Abbreviations
479 [21794] localedata: Added-country_isbn-for-Italy
480 [21795] localedata: Add/Fix country_isbn for France
481 [21796] localedata: Added country_isbn for Republic of Korea
482 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
483 [21799] localedata: Added int_select international_call_prefixes
484 [21801] localedata: Added int_select international_call_prefixes
485 [21804] nptl: Double semicolon in thread-shared-types.h
486 [21807] localedata: LC_ADDRESS fix for pap_CW
487 [21808] localedata: Fix LC_ADDRESS for pap_AW
488 [21821] localedata: Added country_name in mai_IN
489 [21822] localedata: Fix LC_TIME for mai_IN
490 [21823] localedata: missing yesstr, nostr for sa_IN
491 [21825] localedata: Fix name_mrs for mag_IN
492 [21828] localedata: 2.26 changelog should mention user visible changes
494 [21835] localedata: Added Maithili language locale for Nepal
495 [21838] localedata: Removed redundant data for the_NP
496 [21839] localedata: Fix LC_MONETARY for ta_LK
497 [21844] localedata: Fix Latin characters and Months Sequence.
498 [21848] localedata: Fix mai_NP Title Name
503 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
504 24731-2:2010, is supported to enable declarations of functions from that
505 TR. Note that not all functions from that TR are supported by the GNU C
508 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
509 18661-1:2014, is supported to enable declarations of functions and macros
510 from that TS. Note that not all features from that TS are supported by
513 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
514 18661-4:2015, is supported to enable declarations of functions and macros
515 from that TS. Note that most features from that TS are not supported by
518 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
519 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
520 Since the GNU C Library defaults to a much newer revision of POSIX, this
521 will only affect programs that specifically request an old conformance
522 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
523 see a change in the visible declarations, but a program compiled with
524 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
527 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
528 defined by all multithreaded code, but glibc has not required this for
531 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
532 means that in a future release, the macros “major”, “minor”, and “makedev”
533 will only be available from <sys/sysmacros.h>.
535 These macros are not part of POSIX nor XSI, and their names frequently
536 collide with user code; see for instance glibc bug 19239 and Red Hat bug
537 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
538 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
541 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
542 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
543 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
545 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
546 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
547 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
548 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
549 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
550 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
551 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
552 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
553 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
554 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
555 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
558 * New <math.h> features are added from TS 18661-1:2014:
560 - Signaling NaN macros: SNANF, SNAN, SNANL.
562 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
563 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
564 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
566 - llogb functions: the llogb, llogbf and llogbl functions, and the
567 FP_LLOGB0 and FP_LLOGBNAN macros.
569 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
572 - Comparison macros: iseqsig.
574 - Classification macros: iscanonical, issubnormal, iszero.
576 - Total order functions: totalorder, totalorderf, totalorderl,
577 totalordermag, totalordermagf, totalordermagl.
579 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
581 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
582 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
584 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
585 are added to libc. They convert a floating-point number into string.
587 * Most of glibc can now be built with the stack smashing protector enabled.
588 It is recommended to build glibc with --enable-stack-protector=strong.
589 Implemented by Nick Alcock (Oracle).
591 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
592 intended to be used instead of memset() to erase sensitive data after use;
593 the compiler will not optimize out calls to explicit_bzero even if they
594 are "unnecessary" (in the sense that no _correct_ program can observe the
595 effects of the memory clear).
597 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
598 to float instead of double. This does not affect the ABI of any libraries
599 that are part of the GNU C Library, but may affect the ABI of other
600 libraries that use this type in their interfaces.
602 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
603 float_t and double_t types are now defined to long double instead of float
604 and double. These options are not the default, and this does not affect
605 the ABI of any libraries that are part of the GNU C Library, but it may
606 affect the ABI of other libraries that use this type in their interfaces,
607 if they are compiled or used with those options.
609 * The getentropy and getrandom functions, and the <sys/random.h> header file
612 * The buffer size for byte-oriented stdio streams is now limited to 8192
613 bytes by default. Previously, on Linux, the default buffer size on most
614 file systems was 4096 bytes (and thus remains unchanged), except on
615 network file systems, where the buffer size was unpredictable and could be
616 as large as several megabytes.
618 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
619 for the Linux quota interface which predates kernel version 2.4.22 has
622 * The malloc_get_state and malloc_set_state functions have been removed.
623 Already-existing binaries that dynamically link to these functions will
624 get a hidden implementation in which malloc_get_state is a stub. As far
625 as we know, these functions are used only by GNU Emacs and this change
626 will not adversely affect already-built Emacs executables. Any undumped
627 Emacs executables, which normally exist only during an Emacs build, should
628 be rebuilt by re-running “./configure; make” in the Emacs build tree.
630 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
631 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
632 “no-ip6-dotint” had already been the default, and support for the
633 “ip6-dotint” option was removed from the Internet in 2006.
635 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
636 flag from <resolv.h> have been removed. The option relied on a
637 backwards-incompatible DNS extension which was never deployed on the
640 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
641 RES_BLAST defined in the <resolv.h> header file have been deprecated.
642 They were already unimplemented.
644 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
645 _res.flags are deprecated. The flag was standardized in RFC 2133, but
646 removed again from the IETF name lookup interface specification in RFC
647 2553. Applications should use getaddrinfo instead.
649 * DNSSEC-related declarations and definitions have been removed from the
650 <arpa/nameser.h> header file, and libresolv will no longer attempt to
651 decode the data part of DNSSEC record types. Previous versions of glibc
652 only implemented minimal support for the previous version of DNSSEC, which
653 is incompatible with the currently deployed version.
655 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
656 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
657 <arpa/nameser.h> header file because the distinction between RR types and
658 meta-RR types is not officially standardized, subject to revision, and
659 thus not suitable for encoding in a macro.
661 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
662 and rhook members of the res_state type in <resolv.h> have been removed.
663 The glibc stub resolver did not support these hooks, but the header file
664 did not reflect that.
666 * For multi-arch support it is recommended to use a GCC which has
667 been built with support for GNU indirect functions. This ensures
668 that correct debugging information is generated for functions
669 selected by IFUNC resolvers. This support can either be enabled by
670 configuring GCC with '--enable-gnu-indirect-function', or by
671 enabling it by default by setting 'default_gnu_indirect_function'
672 variable for a particular architecture in the GCC source file
675 * GDB pretty printers have been added for mutex and condition variable
676 structures in POSIX Threads. When installed and loaded in gdb these pretty
677 printers show various pthread variables in human-readable form when read
678 using the 'print' or 'display' commands in gdb.
680 * Tunables feature added to allow tweaking of the runtime for an application
681 program. This feature can be enabled with the '--enable-tunables' configure
682 flag. The GNU C Library manual has details on usage and README.tunables has
683 instructions on adding new tunables to the library.
685 * A new version of condition variables functions have been implemented in
686 the NPTL implementation of POSIX Threads to provide stronger ordering
689 * A new version of pthread_rwlock functions have been implemented to use a more
690 scalable algorithm primarily through not using a critical section anymore to
693 Security related changes:
695 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
696 have been created with makecontext could fail to terminate due to a
697 missing .cantunwind annotation. This has been observed to lead to a hang
698 (denial of service) in some Go applications compiled with gccgo. Reported
699 by Andreas Schwab. (CVE-2016-6323)
701 * The DNS stub resolver functions would crash due to a NULL pointer
702 dereference when processing a query with a valid DNS question type which
703 was used internally in the implementation. The stub resolver now uses a
704 question type which is outside the range of valid question type values.
707 The following bugs are resolved with this release:
709 [4099] stdio: Overly agressive caching by stream i/o functions.
710 [7065] build: Support building glibc with -fstack-protector or -fstack-
712 [9842] localedata: en_CA: incorrect date format
713 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
714 before it started waiting
715 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
716 [15765] nptl: sem_open is wrongly a cancellation point
717 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
718 [16458] libc: endian.h and netinet/in.h byte order macros return values of
720 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
721 library linked with pthread
722 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
724 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
725 [17252] libc: getrandom and getentropy syscall
726 [17730] malloc: thread-local storage is sometimes improperly free()'d
727 after being __libc_memalign()'d
728 [18241] stdio: failed fseek on memstream does not set errno and can fail
730 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
732 [18463] nptl: pthread_cond_broadcast issue when surrounded by
733 PTHREAD_PRIO_INHERIT mutex on ARM
734 [18784] network: res_query and related function crash for special record
735 type queries (CVE-2015-5180)
736 [19380] math: strtod does not raise "inexact"
737 [19387] string: Integer overflow in memchr
738 [19390] string: Integer overflow in strncat
739 [19398] build: linknamespace tests fail with massively parallel build
740 [19402] nptl: Deadlock with robust shared mutex and asynchronous
742 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
743 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
744 [19514] libc: [PATCH] Fix spelling errors in spelling
745 "implement"/"implementation" in several places
746 [19582] network: Deprecate RES_USE_INET6
747 [19673] manual: clog10 docs appear to be erroneous
748 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
749 causes a segmentation fault
750 [19826] libc: invalid pointer returned from __tls_get_addr with static
752 [20016] network: resolv: Remove hooks support from the API
753 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
755 [20033] math: [x86_64] vectorized math function don't call the __finite
757 [20116] nptl: use after free in pthread_create
758 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
759 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
761 [20311] nptl: please install proc_service.h
762 [20366] libc: Compilation errors in installed headers in strict-compliance
764 [20370] malloc: malloc: Arena free list management is still racy
765 (incorrect fix in bug 19243)
766 [20386] libc: assert (X = 0) does not result in GCC warning
767 [20432] malloc: malloc: Minimize interface required for interposition
768 [20435] libc: Missing unwind info in __startcontext causes infinite loop
769 in _Unwind_Backtrace (CVE-2016-6323)
770 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
771 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
773 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
774 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
775 [20477] network: resolv: incorrect double-checked locking related to
777 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
779 [20495] math: x86_64 performance degradation due to AVX/SSE transition
781 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
782 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
783 cause transition penalty
784 [20517] math: sparcv9 missing fdiml compat symbol
785 [20524] manual: strverscmp is inconsistent
786 [20525] libc: <sys/quota.h> should be based on kernel headers
787 [20539] math: GCC 7 -static -lm fails to link at -O3
788 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
790 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
792 [20591] network: Remove obsolete DNSSEC support
793 [20592] network: DNS resource record type classification macros in
794 <arpa/nameser.h> are incorrect
795 [20593] network: Update DNS RR type definitions
796 [20611] network: getaddrinfo accepts invalid numeric scope IDs
797 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
799 [20629] network: libresolv: Remove support for bitstring labels
801 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
802 assertion: "maxidx >= 2"
803 [20660] math: [arm] Use VSQRT
804 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
805 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
806 [20689] libc: Test for FMA should also check for AVX.
807 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
808 [20715] math: iszero macro breaks existing code
809 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
810 [20728] libc: powerpc: Missing TOC stub in clone
811 [20729] build: glibc-2.24 fails to build for i486 with -Os
812 [20750] build: Build fails with default PIE enabled gcc-6.x
813 [20768] math: [alpha] sqrt fegetenv namespace
814 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
816 [20787] math: float_t is defined as float incorrectly on x86_64 even if
817 __FLT_EVAL_METHOD__ is 2
818 [20790] network: rpcgen buffer overrun in get_prog_declaration
819 [20822] nptl: race condition in __lll_unlock_elision on powerpc
820 [20829] libc: crypt snprintf namespace
821 [20847] libc: tst-vfork3 failure
822 [20855] math: Default bits/mathdef.h has inappropriate float_t
823 [20859] math: [sh4] FP_ILOGB0 invalid
824 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
825 U+20AC), not same as GBK
826 [20915] dynamic-link: global-dynamic TLS broken on aarch64
827 [20916] math: pow handling of sNaN arguments
828 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
829 [20919] math: Bad pow (qNaN, 0) result with -lieee
830 [20940] math: hypot sNaN handling
831 [20947] math: fmax, fmin sNaN handling
832 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
833 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
835 [20971] string: powerpc64/power7 memchr overflows internal pointer check
836 [20973] nptl: robust mutexes: Lost wake-ups
837 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
838 [20978] nis: strlen on null pointer in nss_nisplus
839 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
841 [21014] string: i686 memchr overflows internal pointer check
842 [21019] libc: [mips] n32 lseek incorrect on overflow
843 [21022] libc: [microblaze] __backtrace get_frame_size namespace
844 [21026] libc: [MIPS] readahead syscall is broken on n64
845 [21028] math: Fallback fesetexceptflag should always succeed
846 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
847 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
849 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
850 [21061] librt: [microblaze] librt lost clock_* exports
851 [21073] libc: tunables: insecure environment variables passed to
852 subprocesses with AT_SECURE
853 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
857 * The minimum Linux kernel version that this version of the GNU C Library
858 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
859 kernel version 2.6.32 or later suffices (on architectures that already
860 required kernel versions more recent than 3.2, those requirements remain
861 unchanged). Linux 3.2 or later kernel headers are required on all
864 * The pap_AN locale has been deleted. This has been deprecated for a long
865 time. It has been replaced by pap_AW & pap_CW, both of which have long
866 been included in previous releases.
868 * The readdir_r and readdir64_r functions have been deprecated. It is
869 recommended to use readdir and readdir64 instead.
871 * The type “union wait” has been removed. It was deprecated in the early
872 1990s and never part of POSIX. Application code should use the int type
873 instead of “union wait”.
875 * A new NSS action is added to facilitate large distributed system
876 administration. The action, MERGE, allows remote user stores like LDAP
877 to be merged into local user stores like /etc/groups in order to provide
878 easy to use, updated, and managed sets of merged credentials. The new
879 action can be used by configuring it in /etc/nsswitch.conf:
880 group: files [SUCCESS=merge] nis
881 Implemented by Stephen Gallagher (Red Hat).
883 * The deprecated __malloc_initialize_hook variable has been removed from the
886 * The long unused localedef --old-style option has been removed. It hasn't
887 done anything in over 16 years. Scripts using this option can safely
890 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
891 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
892 return the next representable value in the direction of positive infinity
893 and the nextdown functions return the next representable value in the
894 direction of negative infinity. These are currently enabled as GNU
897 Security related changes:
899 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
900 could result in a stack overflow when getnetbyname was called with an
901 overly long name. (CVE-2016-3075)
903 * Previously, getaddrinfo copied large amounts of address data to the stack,
904 even after the fix for CVE-2013-4458 has been applied, potentially
905 resulting in a stack overflow. getaddrinfo now uses a heap allocation
906 instead. Reported by Michael Petlan. (CVE-2016-3706)
908 * The glob function suffered from a stack-based buffer overflow when it was
909 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
910 Reported by Alexander Cherepanov. (CVE-2016-1234)
912 * The Sun RPC UDP client could exhaust all available stack space when
913 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
914 alloca plugin for GCC. (CVE-2016-4429)
916 * The IPv6 name server management code in libresolv could result in a memory
917 leak for each thread which is created, performs a failing naming lookup,
918 and exits. Over time, this could result in a denial of service due to
919 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
921 The following bugs are resolved with this release:
923 [1170] localedata: ne_NP: update Nepali locale definition file
924 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
926 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
927 [6796] math: fdim() does not set errno on overflow
928 [10354] libc: posix_spawn should use vfork() in more cases than presently
929 [11213] localedata: localedata: add copyright disclaimer to locale files
930 [12143] localedata: chr_US: new Cherokee locale
931 [12450] localedata: sgs_LT: new locale
932 [12676] localedata: ln_CD: new locale
933 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
935 [13304] math: fma, fmaf, fmal produce wrong results
936 [14259] build: --localedir arg to configure is ignored
937 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
938 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
939 [14934] localedata: es_CL: wrong first weekday chilean locale
940 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
942 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
944 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
945 [15368] nptl: raise() is not async-signal-safe
946 [15479] math: ceil, floor, round and trunc raise inexact exception
947 [15578] localedata: kk_KZ: various updates
948 [16003] localedata: pap_AN: punt old locale
949 [16137] localedata: iw_IL: punt old locale
950 [16190] localedata: eo: new esperanto locale
951 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
952 [16742] malloc: race condition: pthread_atfork() called before first
953 malloc() results in unexpected locking behaviour/deadlocks
954 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
956 [16983] localedata: postal_fmt does not allow %l and %n modifiers
957 [17565] localedata: pt_PT: wrong (work-)week start
958 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
959 [17950] build: Build fails with -msse
960 [18205] localedata: be_BY*: wrong first_weekday and first_workday
961 [18433] libc: posix_spawn does not return correctly upon failure to
963 [18453] localedata: charmaps/IBM875: incorrect codes
964 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
966 [18896] localedata: he_IL: improvements for currency
967 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
969 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
971 [19056] libc: Deprecate readdir_r
972 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
974 [19198] localedata: nl_NL: small improvements for Dutch locales
975 [19257] network: Per-thread memory leak in __res_vinit with IPv6
976 nameservers (CVE-2016-5417)
977 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
979 [19400] locale: Language missing in "iso-639.def", trivial fix in
981 [19431] malloc: Deadlock between fflush, getdelim, and fork
982 [19505] libc: Incorrect file descriptor validity checks in
983 posix_spawn_file_actions_add{open,close,dup2}
984 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
986 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
987 `intl/{gettextP,loadinfo}.h'
988 [19534] libc: execle, execlp may use malloc
989 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
990 [19573] network: res_nclose and __res_maybe_init disagree about name
991 server initialization, breaking Hesiod
992 [19575] localedata: Status of GB18030 tables
993 [19581] localedata: sr_* date_fmt string contains additional newline
994 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
996 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
997 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
998 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
999 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1001 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1003 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1005 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1006 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1007 [19633] locale: strfmon_l applies global locale to number formatting
1008 [19642] network: Memory leak in getnameinfo
1009 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1010 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1012 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1013 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1015 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1016 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1018 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1020 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1022 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1023 pointers and lengths in error-case.
1024 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1025 report errors on UTF-16 surrogates values.
1026 [19755] nscd: nscd assertion failure in gc
1027 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1028 [19759] libc: mempcpy shouldn't be inlined
1029 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1030 [19765] libc: s390 needs an optimized mempcpy
1031 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1032 NAME_MAX limit assumption (CVE-2016-1234)
1033 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1034 [19787] network: Missing and incorrect truncation checks in getnameinfo
1035 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1037 [19791] network: Assertion failure in res_query.c with un-connectable name
1039 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1040 [19822] math: libm.so install clobbers old version
1041 [19825] network: resolv: send_vc can return uninitialized data in second
1042 response to getaddrinfo
1043 [19830] network: nss_dns: should check RDATA length against buffer length
1044 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1045 confronted with A/AAAA records of invalid size
1046 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1047 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1048 [19853] stdio: Printing IBM long double in decimal with high precision is
1050 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1051 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1052 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1054 [19865] network: Assertion failure or memory leak in
1055 _nss_dns_getcanonname_r
1056 [19868] network: nss_dns: netent code does not skip over non-PTR records
1057 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1059 [19881] string: Improve x86-64 memset
1060 [19907] string: Incorrect memcpy tests
1061 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1063 [19925] libc: termios.h XCASE namespace
1064 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1065 [19929] libc: limits.h NL_NMAX namespace
1066 [19931] stdio: Memory leak in vfprintf
1067 [19957] libc: clone(CLONE_VM) access invalid parent memory
1068 [19963] localedata: en_IL: New locale
1069 [19989] stdio: stdio.h cuserid namespace
1070 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1072 [19996] locale: langinfo.h nl_langinfo_l namespace
1073 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1074 [20010] network: getaddrinfo: Stack overflow in hostent translation
1076 [20012] stdio: libio: fmemopen append mode failure
1077 [20014] stdio: stdio.h namespace for pre-threads POSIX
1078 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1079 [20023] libc: fcntl.h timespec namespace
1080 [20024] math: [x86_64] vectorized sincos trashes the stack
1081 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1082 [20041] time: sys/time.h timespec namespace
1083 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1084 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1085 [20051] libc: ttyslot in wrong header under wrong conditions
1086 [20054] libc: gethostname not declared for XPG4
1087 [20055] libc: termios.h missing tcgetsid for XPG4
1088 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1090 [20073] libc: sys/stat.h fchmod namespace
1091 [20074] libc: stdlib.h rand_r namespace
1092 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1093 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1095 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1096 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1098 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1099 [20119] libc: Wrong mask for processors level type from CPUID
1100 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1101 AS not supporting AVX512
1102 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1104 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1105 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1107 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1108 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1109 [20178] libc: posix_spawn{p} should not call exit
1110 [20191] stdio: libio: vtables hardening
1111 [20195] string: FMA4 detection requires CPUID execution with register
1113 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1114 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1116 [20212] math: acos (sNaN) returns sNaN
1117 [20213] math: asin (sNaN) returns sNaN
1118 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1119 [20218] math: [i386] asinhl (sNaN) returns sNaN
1120 [20219] math: [i386] atanhl (sNaN) returns sNaN
1121 [20222] stdio: fopencookie: Mangle function pointers
1122 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1123 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1124 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1126 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1127 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1128 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1129 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1130 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1131 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1132 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1133 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1134 [20240] math: modf (sNaN) returns sNaN
1135 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1136 [20250] math: frexp (sNaN) returns sNaN
1137 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1138 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1140 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1141 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1142 subscript is below array bounds [-Werror=array-bounds]
1143 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1144 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1146 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1147 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1149 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1150 "invalid" exceptions
1151 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1152 [20316] localedata: id_ID: Februari instead of Pebruari
1153 [20327] string: POWER8 strcasecmp returns incorrect result
1154 [20347] math: Failure: Test: j0_downward (0xap+0)
1155 [20348] libc: FAIL: misc/tst-preadvwritev64
1156 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1157 [20350] libc: There is no test for p{read,write}64
1158 [20357] math: Incorrect cos result for 1.5174239687223976
1159 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1163 * Unicode 8.0.0 Support: Character encoding, character type info, and
1164 transliteration tables are all updated to Unicode 8.0.0, using new
1165 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1166 These updates cause user visible changes, such as the fixes for bugs
1167 89, 16061, and 18568.
1169 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1170 kernel-internal CPU set size. This means that requests that change the
1171 CPU affinity which failed before (for example, an all-ones CPU mask) will
1172 now succeed. Applications that need to determine the effective CPU
1173 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1174 setting it because the kernel can adjust it (and the previous size check
1175 would not detect this in the majority of cases).
1177 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1178 the following new symbols are used: fts64_children, fts64_close,
1179 fts64_open, fts64_read and fts64_set.
1181 * getaddrinfo now detects certain invalid responses on an internal netlink
1182 socket. If such responses are received, an affected process will
1183 terminate with an error message of "Unexpected error <number> on netlink
1184 descriptor <number>" or "Unexpected netlink response of size <number> on
1185 descriptor <number>". The most likely cause for these errors is a
1186 multi-threaded application which erroneously closes and reuses the netlink
1187 file descriptor while it is used by getaddrinfo.
1189 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1190 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1191 unnecessary serialization of memory allocation requests across threads.
1192 The defect is now corrected. Users should see a substantial increase in
1193 the concurent throughput of allocation requests for applications which
1194 trigger this bug. Affected applications typically create and
1195 destroy threads frequently. (Bug 19048 was reported and analyzed by
1198 * There is now a --disable-timezone-tools configure option for disabling the
1199 building and installing of the timezone related utilities (zic, zdump, and
1200 tzselect). This is useful for people who build the timezone data and code
1201 independent of the GNU C Library.
1203 * The obsolete header <regexp.h> has been removed. Programs that require
1204 this header must be updated to use <regex.h> instead.
1206 * The obsolete functions bdflush, create_module, get_kernel_syms,
1207 query_module and uselib are no longer available to newly linked binaries;
1208 the header <sys/kdaemon.h> has been removed. These functions and header
1209 were specific to systems using the Linux kernel and could not usefully be
1210 used with the GNU C Library on systems with version 2.6 or later of the
1213 * Optimized string, wcsmbs and memory functions for IBM z13.
1214 Implemented by Stefan Liebler.
1216 * Newly linked programs that define a variable called signgam will no longer
1217 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1218 require signgam to be set by those functions must ensure that they use the
1219 variable provided by the GNU C Library and declared in <math.h>, without
1220 defining their own copy.
1222 * The minimum GCC version that can be used to build this version of the GNU
1223 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1224 still be used to compile programs using the GNU C Library.
1226 Security related changes:
1228 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1229 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1231 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1232 the pointer guard feature. It is always enabled. Previously,
1233 LD_POINTER_GUARD could be used to disable security hardening in binaries
1234 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1237 * An integer overflow in hcreate and hcreate_r could lead to an
1238 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1240 * The catopen function no longer has unbounded stack usage. Reported by
1241 Max. (CVE-2015-8779)
1243 * The nan, nanf and nanl functions no longer have unbounded stack usage
1244 depending on the length of the string passed as an argument to the
1245 functions. Reported by Joseph Myers. (CVE-2014-9761)
1247 * A stack-based buffer overflow was found in libresolv when invoked from
1248 libnss_dns, allowing specially crafted DNS responses to seize control
1249 of execution flow in the DNS client. The buffer overflow occurs in
1250 the functions send_dg (send datagram) and send_vc (send TCP) for the
1251 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1252 family. The use of AF_UNSPEC triggers the low-level resolver code to
1253 send out two parallel queries for A and AAAA. A mismanagement of the
1254 buffers used for those queries could result in the response of a query
1255 writing beyond the alloca allocated buffer created by
1256 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1257 the overflow. Thanks to the Google Security Team and Red Hat for
1258 reporting the security impact of this issue, and Robert Holiday of
1259 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1261 The following bugs are resolved with this release:
1263 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1264 [887] math: Math library function "logb" and "nextafter" inconsistent
1265 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1266 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1267 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1268 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1270 [4404] localedata: German translation of "Alarm clock" is misleading
1271 [6799] math: nextafter() and nexttoward() doen't set errno on
1272 overflow/underflow errors
1273 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1275 [10432] nis: _nss_nis_setnetgrent assertion failure
1276 [11460] libc: fts has no LFS support
1277 [12926] network: getaddrinfo()/make_request() may spin forever
1278 [13065] nptl: Race condition in pthread barriers
1279 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1280 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1282 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1283 [14912] libc: Rename non-installed bits/*.h headers
1284 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1285 [15367] math: Let gcc use __builtin_isinf
1286 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1287 [15421] math: lgamma wrongly sets signgam for ISO C
1288 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1289 FE_INVALID with argument out of range
1290 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1292 [15786] dynamic-link: ifunc resolver functions can smash function
1294 [15918] math: Unnecessary check for equality in hypotf()
1295 [16061] localedata: Review / update transliteration data
1296 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1298 [16141] time: strptime %z offset restriction
1299 [16171] math: drem should be alias of remainder
1300 [16296] math: fegetround is pure?
1301 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1302 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1304 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1305 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1307 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1308 "Exception "Inexact" set" on ppc32
1309 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1310 [16517] math: Missing underflow exception from tanf/tan/tanl
1311 [16519] math: Missing underflow exception from sinhf
1312 [16520] math: Missing underflow exception from tanhf
1313 [16521] math: Missing underflow exception from exp2
1314 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1316 [16734] stdio: fopen calls mmap to allocate its buffer
1317 [16961] math: nan function incorrect handling of bad sequences
1318 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1319 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1320 [16985] locale: localedef: confusing error message when opening output
1322 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1323 [17197] locale: Redundant shift character in iconv conversion output at
1325 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1326 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1327 [17250] dynamic-link: static linking breaks nss loading
1328 (getaddrinfo/getpwnam/etc...)
1329 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1331 [17441] math: isnan() should use __builtin_isnan() in GCC
1332 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1333 (related to lock elision)
1334 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1335 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1336 [17887] time: strptime should be able to parse "+01:00" style timezones
1337 [17905] libc: catopen() Multiple unbounded stack allocations
1339 [18084] libc: backtrace (..., 0) dumps core on x86
1340 [18086] libc: nice() sets errno to 0 on success
1341 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1342 is too large (CVE-2015-8778)
1343 [18251] dynamic-link: SONAME missing when audit modules provides path
1344 [18265] libc: add attributes for wchar string and memory functions
1345 [18370] math: csqrt missing underflows
1346 [18421] libc: [hppa] read-only segment has dynamic relocations
1347 [18472] libc: Obsolete syscall wrappers should be compat symbols
1348 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1349 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1350 [18525] localedata: Remove locale timezone information
1351 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1352 [18568] localedata: Update locale data to Unicode 8.0
1353 [18589] locale: sort-test.sh fails at random
1354 [18595] math: ctan, ctanh missing underflows
1355 [18604] libc: assert macro-expands its argument
1356 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1357 contains a vector instruction exception.
1358 [18611] math: j1, jn missing errno setting on underflow
1359 [18618] localedata: sync Chechen locale definitions with other *_RU
1361 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1362 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1363 [18665] network: In send_dg, the recvfrom function is NOT always using the
1364 buffer size of a newly created buffer (CVE-2015-7547)
1365 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1367 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1369 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1370 [18699] math: tilegx cproj() for various complex infinities does not yield
1372 [18724] libc: Harden put*ent functions against data injection
1373 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1374 [18755] build: build errors with -DNDEBUG
1375 [18757] stdio: fmemopen fails to set errno on failure
1376 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1378 [18781] libc: openat64 lacks O_LARGEFILE
1379 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1380 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1381 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1382 [18790] math: [ldbl-128ibm] tanhl inaccurate
1383 [18795] libc: stpncpy fortification misses buffer lengths that are
1384 statically too large
1385 [18796] build: build fails for --disable-mathvec
1386 [18803] math: hypot missing underflows
1387 [18820] stdio: fmemopen may leak memory on failure
1388 [18823] math: csqrt spurious underflows
1389 [18824] math: fma spurious underflows
1390 [18825] math: pow missing underflows
1391 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1392 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1393 [18870] build: sem_open.c fails to compile with missing symbol
1395 [18872] stdio: Fix memory leak in printf_positional
1396 [18873] libc: posix_fallocate overflow check ineffective
1397 [18875] math: Excess precision leads incorrect libm
1398 [18877] libc: arm: mmap offset regression
1399 [18887] libc: memory corruption when using getmntent on blank lines
1400 [18918] localedata: hu_HU: change time to HH:MM:SS format
1401 [18921] libc: Regression: extraneous stat() and fstat() performed by
1403 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1404 binaries (CVE-2015-8777)
1405 [18951] math: tgamma missing underflows
1406 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1408 [18953] localedata: lt_LT: change currency symbol to the euro
1409 [18956] math: powf inaccuracy
1410 [18961] math: [i386] exp missing underflows
1411 [18966] math: [i386] exp10 missing underflows
1412 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1413 [18969] build: multiple string test failures due to missing locale
1415 [18970] libc: Reference of pthread_setcancelstate in libc.a
1416 [18977] math: float / long double Bessel functions not in XSI POSIX
1417 [18980] math: i386 libm functions return with excess range and precision
1418 [18981] math: i386 scalb*, ldexp return with excess range and precision
1419 [18982] stdio: va_list and vprintf
1420 [18985] time: Passing out of range data to strftime() causes a segfault
1422 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1423 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1424 [19012] locale: iconv_open leaks memory on error path
1425 [19016] math: clog, clog10 inaccuracy
1426 [19018] nptl: Mangle function pointers in tls_dtor_list
1427 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1428 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1429 [19048] malloc: malloc: arena free list can become cyclic, increasing
1431 [19049] math: [powerpc] erfc incorrect zero sign
1432 [19050] math: [powerpc] log* incorrect zero sign
1433 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1434 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1435 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1436 [19074] network: Data race in _res_hconf_reorder_addrs
1437 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1438 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1439 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1441 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1443 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1444 [19086] manual: posix_fallocate64 documented argument order is wrong.
1445 [19088] math: lround, llround missing exceptions close to overflow
1447 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1448 [19095] math: dbl-64 lrint incorrect for 64-bit long
1449 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1450 [19124] dynamic-link: ld.so failed to build with older assmebler
1451 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1452 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1453 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1454 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1455 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1456 pthread_setaffinity_np
1457 [19156] math: [ldbl-128] j0l spurious underflows
1458 [19164] nptl: tst-getcpu fails with many possible CPUs
1459 [19168] math: math/test-ildoubl and math/test-ldouble failure
1460 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1461 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1463 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1464 (FE_NOMASK_ENV) do not clear SSE exceptions
1465 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1466 _int_new_arena/reused_arena
1467 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1468 [19201] math: dbl-64 remainder incorrect sign of zero result
1469 [19205] math: bits/math-finite.h conditions do not match math.h and
1471 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1472 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1473 for C99-based standards
1474 [19212] libc: features.h not -Wundef clean
1475 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1477 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1478 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1479 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1481 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1482 "inexact" exceptions
1483 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1485 [19242] libc: strtol incorrect in Turkish locales
1486 [19243] malloc: reused_arena can pick an arena on the free list, leading
1487 to an assertion failure and reference count corruption
1488 [19253] time: tzset() ineffective when temporary TZ did not include DST
1490 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1491 [19270] math: [hppa] Shared libm missing __isnanl
1492 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1493 MAP_HUGETLB and MAP_STACK defines
1494 [19313] nptl: Wrong __cpu_mask for x32
1495 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1496 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1497 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1498 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1499 [19363] time: x32: times() return value wrongly truncates/sign extends
1501 [19367] dynamic-link: Improve branch prediction on Silvermont
1502 [19369] network: Default domain name not reset by res_ninit when "search"
1503 / "domain" entry is removed from resolv.conf
1504 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1506 [19385] localedata: bg_BG: time separator should be colon, not comma
1507 [19408] libc: linux personality syscall wrapper may erroneously return an
1508 error on 32-bit architectures
1509 [19415] libc: dladdr returns wrong names on hppa
1510 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1511 IBM905, IBM907, and IBM909
1512 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1513 [19443] build: build failures with -DDEBUG
1514 [19451] build: Make check fails on test-double-vlen2
1515 [19462] libc: Glibc failed to build with -Os
1516 [19465] math: Wrong code with -Os
1517 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1519 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1521 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1522 [19486] math: S390: Math tests fail with "Exception Inexact set".
1523 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1524 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1526 [19590] math: Fail to build shared objects that use libmvec.so functions.
1530 * The following bugs are resolved with this release:
1532 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1533 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1534 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1535 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1536 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1537 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1538 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1539 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1540 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1541 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1542 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1543 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1544 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1545 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1546 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1547 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1548 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1549 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1550 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1551 18657, 18676, 18694, 18696, 18887.
1553 * Cache information can be queried via sysconf() function on s390 e.g. with
1554 _SC_LEVEL1_ICACHE_SIZE as argument.
1556 * A buffer overflow in gethostbyname_r and related functions performing DNS
1557 requests has been fixed. If the NSS functions were called with a
1558 misaligned buffer, the buffer length change due to pointer alignment was
1559 not taken into account. This could result in application crashes or,
1560 potentially arbitrary code execution, using crafted, but syntactically
1561 valid DNS responses. (CVE-2015-1781)
1563 * The time zone file parser has been made more robust against crafted time
1564 zone files, avoiding heap buffer overflows related to the processing of
1565 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1566 large time zone data files. Overly long time zone specifiers in the TZ
1567 variable no longer result in stack overflows and crashes.
1569 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1570 for LD and GD on x86 and x86-64, has been implemented. You will need
1571 binutils-2.24 or later to enable this optimization.
1573 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1574 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1575 Hat). These updates cause user visible changes, such as the fix for bug
1578 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1579 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1580 condition in some applications.
1582 * Added vector math library named libmvec with the following vectorized x86_64
1583 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1585 The library can be disabled with --disable-mathvec. Use of the functions is
1586 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1587 Shared library libmvec.so is linked in as needed when using -lm (no need to
1588 specify -lmvec explicitly for not static builds).
1589 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1591 * A new fmemopen implementation has been added with the goal of POSIX
1592 compliance. The new implementation fixes the following long-standing
1593 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1594 old implementation is still present for use be by existing binaries.
1596 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1597 releases. It has been fixed to match 2.19 and older, but binaries built
1598 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1600 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1601 Contributed by Roland McGrath (Google).
1603 * The header <regexp.h> is deprecated, and will be removed in a future
1604 release. Use of this header will trigger a deprecation warning.
1605 Application developers should update their code to use <regex.h> instead.
1607 This header was formerly part of SUS, but was deprecated in 1994 and
1608 removed from the standard in 2001. Also, the glibc implementation
1609 leaks memory. See BZ#18681 for more details.
1613 * The following bugs are resolved with this release:
1615 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1616 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1617 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1618 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1619 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1620 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1621 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1622 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1623 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1624 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1627 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1628 too little memory for the to-be-scanned arguments and overflow the
1629 allocated buffer. The implementation now correctly computes the required
1630 buffer size when using malloc, and switches to malloc from alloca as
1633 * A new semaphore algorithm has been implemented in generic C code for all
1634 machines. Previous custom assembly implementations of semaphore were
1635 difficult to reason about or ensure that they were safe. The new version
1636 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1637 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1638 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1640 * Port to Altera Nios II has been contributed by Mentor Graphics.
1642 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1643 implementations for powerpc64/powerpc64le.
1644 Implemented by Adhemerval Zanella (IBM).
1646 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1647 and powerpc64le. This may improve lock scaling of existing programs on
1648 HTM capable systems. The lock elision code is only enabled with
1649 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1650 powerpc will issue a transaction abort on every syscall to avoid side
1651 effects being visible outside transactions.
1653 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1654 AArch64. Contributed by ARM Ltd.
1656 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1658 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1659 under certain input conditions resulting in the execution of a shell for
1660 command substitution when the application did not request it. The
1661 implementation now checks WRDE_NOCMD immediately before executing the
1662 shell and returns the error WRDE_CMDSUB as expected.
1664 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1665 processing format strings with a large number of format specifiers.
1667 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1668 infinite loop if the DNS response contained a PTR record of an unexpected
1671 * The minimum GCC version that can be used to build this version of the GNU
1672 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1673 still be used to compile programs using the GNU C Library.
1675 * The GNU C Library is now built with -Werror by default. This can be
1676 disabled by configuring with --disable-werror.
1678 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1680 * The obsolete sigvec function has been removed. This was the original
1681 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1682 programs have been using instead for about 25 years. Of course, ABI
1683 compatibility for old binaries using sigvec remains intact.
1685 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1686 with newer versions of bison.
1688 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1689 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1690 registers overlay two consecutive single-precision registers. MIPS32R2
1691 introduced a new FPU mode (FR=1) where double-precision registers extend the
1692 corresponding single-precision registers which is incompatible with the
1693 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1694 require the use of FR=1 making a transition necessary. New o32 ABI
1695 extensions enable users to migrate over time from the original o32 ABI
1696 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1697 tracks the ABI of any loaded object and verifies that new objects are
1698 compatible. Mode transitions will also be requested as required and
1699 unsupportable objects will be rejected. The ABI checks include both soft and
1700 hard float ABIs for o32, n32 and n64.
1702 GCC 5 with GNU binutils 2.25 onwards:
1703 It is strongly recommended that all o32 system libraries are built using the
1704 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1705 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1706 compiler using --with-fp-32=xx to set this by default.
1710 * The following bugs are resolved with this release:
1712 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1713 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1714 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1715 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1716 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1717 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1718 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1719 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1720 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1721 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1722 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1723 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1724 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1725 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1726 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1728 * Reverted change of ABI data structures for s390 and s390x:
1729 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1730 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1731 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1732 are preserved pointing straight to the same implementation as the old ones.
1733 Given that, new callers will simply provide a too-big buffer to these
1734 functions. Any applications/libraries out there that embed jmp_buf or
1735 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1736 against 2.19 headers will have to rebuilt again. This is necessary in any
1737 case to revert the breakage in their ABI caused by the glibc change.
1739 * Support for file description locks is added to systems running the
1740 Linux kernel. The standard file locking interfaces are extended to
1741 operate on file descriptions, not file descriptors, via the use of
1742 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1743 are associated with an open file instead of a process.
1745 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1747 * The minimum Linux kernel version that this version of the GNU C Library
1748 can be used with is 2.6.32.
1750 * Running the testsuite no longer terminates as soon as a test fails.
1751 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1752 with PASS or FAIL lines for individual tests. A summary of the results is
1753 printed, including a list of failing lists, and "make check" exits with
1754 error status if there were any unexpected failures. "make check
1755 stop-on-test-failure=y" may be used to keep the old behavior.
1757 * The am33 port, which had not worked for several years, has been removed
1760 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1761 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1762 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1763 conflicted with POSIX (support for which was removed in 2.19), the
1764 interfaces those macros enabled remain available when compiling with
1765 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1766 test macros defined.
1768 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1770 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1771 This may improve lock scaling of existing programs on TX capable systems.
1772 The lock elision code is only built with --enable-lock-elision=yes and
1773 then requires a GCC version supporting the TX builtins. With lock elision
1774 default mutexes are elided via __builtin_tbegin, if the cpu supports
1775 transactions. By default lock elision is not enabled and the elision code
1778 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1779 copy the path argument. This allowed programs to cause posix_spawn to
1780 deference a dangling pointer, or use an unexpected pathname argument if
1781 the string was modified after the posix_spawn_file_actions_addopen
1784 * All supported architectures now use the main glibc sysdeps directory
1785 instead of some being in a separate "ports" directory (which was
1786 distributed separately before glibc 2.17).
1788 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1789 On configurations that support it (all Linux configurations), it's now
1790 used regardless of the --enable-add-ons switch to configure. It is no
1791 longer possible to build such configurations without pthreads support.
1793 * Locale names, including those obtained from environment variables (LANG
1794 and the LC_* variables), are more tightly checked for proper syntax.
1795 setlocale will now fail (with EINVAL) for locale names that are overly
1796 long, contain slashes without starting with a slash, or contain ".." path
1797 components. (CVE-2014-0475) Previously, some valid locale names were
1798 silently replaced with the "C" locale when running in AT_SECURE mode
1799 (e.g., in a SUID program). This is no longer necessary because of the
1802 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1803 with application code using Intel MPX instructions. (With all previous
1804 versions, the MPX register state could be clobbered when making calls
1805 into or out of a shared library.) Note that while the new dynamic
1806 linker is compatible with all known x86 hardware whether or not it
1807 supports Intel MPX, some x86 instruction-set emulators might fail to
1808 handle the new instruction encodings. This is known to affect Valgrind
1809 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1810 release), and might affect other tools that do instruction emulation.
1812 * Support for loadable gconv transliteration modules has been removed.
1813 The support for transliteration modules has been non-functional for
1814 over a decade, and the removal is prompted by security defects. The
1815 normal gconv conversion modules are still supported. Transliteration
1816 with //TRANSLIT is still possible, and the //IGNORE specifier
1817 continues to be supported. (CVE-2014-5119)
1819 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1820 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1821 resulting a denial-of-service security vulnerability in applications which
1822 use functions related to iconv. (CVE-2014-6040)
1826 * The following bugs are resolved with this release:
1828 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1829 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1830 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1831 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1832 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1833 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1834 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1835 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1836 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1837 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1838 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1839 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1840 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1841 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1842 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1843 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1844 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1845 16431, 16453, 16474, 16506, 16510, 16529
1847 * Slovenian translations for glibc messages have been contributed by the
1848 Translation Project's Slovenian team of translators.
1850 * The public headers no longer use __unused nor __block. This change is to
1851 support compiling programs that are derived from BSD sources and use
1852 __unused internally, and to support compiling with Clang's -fblock
1853 extension which uses __block.
1855 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1856 large collation sequences to optimize multiple passes. This cache
1857 computation may overflow for large collation sequences and may cause a
1858 stack or buffer overflow. This is now fixed to use a slower algorithm
1859 which does not use a cache if there is an integer overflow.
1861 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1862 rules for large collation sequences to optimize multiple passes and falls
1863 back to alloca if malloc fails, resulting in a possible stack overflow.
1864 The implementation now falls back to an uncached collation sequence lookup
1867 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1868 initialized for static applications resulting in the security feature
1869 being disabled. The pointer guard is now correctly initialized to a
1870 random value for static applications. Existing static applications need
1871 to be recompiled to take advantage of the fix (bug 15754).
1873 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1874 to the d_name member of struct dirent, or omit the terminating NUL
1875 character. (Bugzilla #14699).
1877 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1878 aligned_alloc functions could allocate too few bytes or corrupt the
1879 heap when passed very large allocation size values (Bugzilla #15855,
1882 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1883 for AF_INET6 has been fixed (Bugzilla #16072).
1885 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1886 pap_CW, quz_PE, the_NP.
1888 * Substantially revised locales: gd_GB, ht_HT
1890 * The LC_ADDRESS field was updated to support country_car for almost all
1893 * ISO 1427 definitions were updated.
1895 * ISO 3166 definitions were updated.
1897 * The localedef utility now supports --big-endian and --little-endian
1898 command-line options to generate locales for a different system from that
1899 for which the C library was built.
1901 * Binary locale files now only depend on the endianness of the system for
1902 which they are generated and not on other properties of that system. As a
1903 consequence, binary files generated with new localedef may be incompatible
1904 with old versions of the GNU C Library, and binary files generated with
1905 old localedef may be incompatible with this version of the GNU C Library,
1906 in the following circumstances:
1908 + Locale files may be incompatible on m68k systems.
1910 + Locale archive files (but not separate files for individual locales) may
1911 be incompatible on systems where plain "char" is signed.
1913 * The configure option --disable-versioning has been removed. Builds with
1914 --disable-versioning had not worked for several years.
1916 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1917 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1919 * SystemTap probes for malloc have been introduced.
1921 * SystemTap probes for slow multiple precision fallback paths of
1922 transcendental functions have been introduced.
1924 * Support for powerpc64le has been added.
1926 * The soft-float powerpc port now supports e500 processors.
1928 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1930 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1931 set of header declarations that are enabled by default, even when other
1932 feature test macros or compiler options such as -std=c99 would otherwise
1933 disable some of those declarations.
1935 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1936 conflict with POSIX. The libbsd-compat library (which was a dummy library
1937 that did nothing) has also been removed.
1939 * Preliminary documentation about Multi-Thread, Async-Signal and
1940 Async-Cancel Safety has been added.
1942 * Change of ABI data structures for s390 and s390x:
1943 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1944 allow for future hardware extensions. All library functions that accept or
1945 return these structures were versioned in order to provide backwards
1946 compatibility with existing code. However, not all cases can be handled
1947 with interface versioning. If an application embeds either structure into
1948 another structure and that structure is passed to another compilation unit
1949 which expects the newer larger-sized object then any access to the new
1950 fields will result in undefined behaviour. Similarly any access to fields
1951 that were moved by the enlarging of the structures will cause undefined
1952 behaviour. To fix the undefined behaviour all objects that are part of the
1953 final in-memory image must agree on the size of structures, and this may
1954 require recompilation.
1958 * The following bugs are resolved with this release:
1960 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1961 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1962 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1963 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1964 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1965 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1966 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1967 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1968 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1969 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1970 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1971 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1972 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1975 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1976 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1977 Distributions can re-enable building and using pt_chown via the new configure
1978 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1979 considerable security risks and should only be used if the distribution
1980 understands and accepts the risks.
1982 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1985 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1986 fixed (Bugzilla #15330).
1988 * Add support for calling C++11 thread_local object destructors on thread
1989 and program exit. This needs compiler support for offloading C++11
1990 destructor calls to glibc.
1992 * Improved worst case performance of libm functions with double inputs and
1995 * Support for priority inherited mutexes in pthread condition variables on
1996 non-x86 architectures.
1998 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2000 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2002 * Optimized string functions for ARM. Implemented by Will Newton and
2005 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2007 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2010 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2011 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2013 * Added a benchmark framework to track performance of functions in glibc.
2015 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2016 It is based on draft TS 18661 and currently enabled as a GNU extension.
2018 * On Linux, the clock function now uses the clock_gettime system call
2019 for improved precision, rather than old times system call.
2021 * Added support for version-3 tz format files. This is needed when using
2022 the tz database release 2013e or later, and affects a few unusual cases --
2023 currently only TZ='America/Godthab' for time stamps after 2037.
2025 * Added new API functions pthread_getattr_default_np and
2026 pthread_setattr_default_np to get and set the default pthread
2027 attributes of a process.
2029 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2030 This may improve lock scaling of existing programs on TSX capable systems.
2031 When the --enable-lock-elision=yes parameter is specified at configure
2032 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2035 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2036 availability). Implemented by Ryan S. Arnold.
2038 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2042 * The following bugs are resolved with this release:
2044 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2045 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2046 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2047 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2048 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2049 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2050 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2051 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2052 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2053 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2054 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2055 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2056 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2057 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2058 14879, 14889, 14893, 14898, 14914.
2060 * Optimization of memcpy for MIPS.
2062 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2063 EMFILE has been fixed (Bugzilla #14889).
2065 * The add-on ports collection is now distributed in the "ports" subdirectory
2066 of the main GNU C Library distribution, rather than separately.
2068 * Port to ARM AArch64 contributed by Linaro.
2070 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2071 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2073 Implemented by Andreas Krebbel.
2075 * The new function secure_getenv allows secure access to the environment,
2076 returning NULL if running in a SUID/SGID process. This function replaces
2077 the internal function __secure_getenv.
2079 * SystemTap static probes have been added into the dynamic linker.
2080 Implemented by Gary Benson.
2082 * Optimizations of string functions strstr, strcasestr and memmem.
2083 Implemented by Maxim Kuvyrkov.
2085 * The minimum Linux kernel version that this version of the GNU C Library
2086 can be used with is 2.6.16.
2088 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2089 powerpc POWER7. Implemented by Will Schmidt.
2091 * New configure option --disable-nscd builds the C library such that it
2092 never attempts to contact the Name Service Caching Daemon (nscd).
2093 New configure option --disable-build-nscd avoids building nscd itself;
2094 this is the default if --disable-nscd is used.
2096 * Improved support for cross-compilation, including cross-testing and
2097 bootstrap builds without a previously built glibc.
2099 * Several testsuite tests are now able to test multiple IFUNC variants of an
2100 interface, rather than just testing the one that would be chooen by
2103 * New configure options --with-bugurl and --with-pkgversion, for
2104 distributors to use to embed their bug-reporting and package version
2105 information in --help and --version output.
2107 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2108 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2109 allows creation of chroots without the procfs mounted on /proc.
2111 * The `crypt' function now fails if passed salt bytes that violate the
2112 specification for those values. On Linux, the `crypt' function will
2113 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2114 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2115 when the mode is enabled.
2117 * The `clock_*' suite of functions (declared in <time.h>) is now available
2118 directly in the main C library. Previously it was necessary to link with
2119 -lrt to use these functions. This change has the effect that a
2120 single-threaded program that uses a function such as `clock_gettime' (and
2121 is not linked with -lrt) will no longer implicitly load the pthreads
2122 library at runtime and so will not suffer the overheads associated with
2123 multi-thread support in other code such as the C++ runtime library.
2125 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2130 * The following bugs are resolved with this release:
2132 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2133 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2134 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2135 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2136 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2137 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2138 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2139 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2140 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2141 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2142 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2143 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2144 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2145 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2146 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2147 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2148 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2149 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2150 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2151 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2152 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2153 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2156 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2157 configuring glibc with:
2158 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2159 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2160 Implemented by H.J. Lu.
2164 + define static_assert
2166 + do not declare gets
2168 + declare at_quick_exit and quick_exit also for ISO C11
2170 + aligned_alloc. NB: The code is deliberately allows the size parameter
2171 to not be a multiple of the alignment. This is a moronic requirement
2172 in the standard but it is only a requirement on the caller, not the
2175 + timespec_get added
2177 + uchar.h support added
2179 + CMPLX, CMPLXF, CMPLXL added
2181 Implemented by Ulrich Drepper.
2183 * Support for the IA-64 has been moved to ports.
2185 * Remove support for anything but ELF binary format
2187 * Checking versions of poll, ppoll added.
2188 Implemented by Ulrich Drepper.
2190 * More generic and 64-bit performance optimizations to math functions.
2191 Implemented by Ulrich Drepper.
2193 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2194 headers and functions available at compile time as they were before
2195 version 2.14. This option will be removed at some time in the future
2196 after the TI-RPC library becomes fully sufficient for the needs of
2197 existing applications.
2199 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2200 Note that glibc is not expected to work with any Linux kernel version
2203 * New header <sys/auxv.h> and function getauxval allowing easy access to
2204 the AT_* key-value pairs passed from the Linux kernel. The header also
2205 defines the HWCAP_* bits associated with the AT_HWCAP key.
2207 * New locales: mag_IN
2209 * New configure option --enable-systemtap builds SystemTap static probes
2210 into libc for setjmp and longjmp and into libpthread for various operations.
2211 So far the setjmp/longjmp probes and some of the libpthread probes are
2212 provided only for i*86 and x86_64.
2213 Implemented by Roland McGrath and Rayson Ho.
2215 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2217 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2220 * More optimized functions for SPARC. Implemented by David S. Miller.
2222 * Improved support for cross-compilation, especially bootstrap builds
2223 without a previously built glibc.
2225 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2226 Chris Metcalf from Tilera.
2228 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2229 now supported for ARM processors.
2231 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2232 as the name of the dynamic linker, to distinguish it from the
2233 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2235 * Support for CRIS has been removed from ports.
2237 * A new class of installed header has been documented for low-level
2238 platform-specific functionality. PowerPC added the first instance with a
2239 function to provide time base register access. Contributed by Tulio
2240 Magno Quites Machado Filho.
2242 * ix86 configurations now install header files that are consistent with
2243 what x86-64 configurations install. These same header files can be used
2244 for -m32, -m64, or -mx32 builds.
2245 Contributed by H.J. Lu.
2247 * Math library bug fixes. A thorough audit of all open math library bugs was
2248 conducted by Joseph Myers. Significant progress was made on many math
2249 library bugs resulting in more accurate exceptions and function results.
2250 Many thanks to all those that contributed including Andreas Jaeger for his
2251 patch review and work on the x87 trigonometric instruction issues.
2253 * Timezone data is no longer installed. Timezone-related binaries and scripts
2254 will continue to be installed. Users should obtain their timezone data from
2255 their distribution provider or from the tzdata package at
2256 <ftp://munnari.oz.au/pub/>.
2260 * The following bugs are resolved with this release:
2262 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2263 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2264 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2265 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2266 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2267 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2268 13484, 13506, 13515, 13523, 13524, 13538, 13540
2270 * New program pldd to list loaded object of a process
2271 Implemented by Ulrich Drepper.
2273 * Add nss_db support back to glibc. No more dependency on Berkeley db
2274 and support for initgroups lookups.
2275 Implemented by Ulrich Drepper.
2277 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2278 Contributed by HJ Lu.
2280 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2281 Contributed by HJ Lu.
2283 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2284 on x86-32 and x86-64.
2285 Contributed by Liubov Dmitrieva.
2287 * Optimized strchr and strrchr for SSE on x86-32.
2288 Contributed by Liubov Dmitrieva.
2290 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2291 for x86-64 and x86-32.
2292 Contributed by Liubov Dmitrieva.
2294 * New interfaces: scandirat, scandirat64
2295 Implemented by Ulrich Drepper.
2297 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2298 Implemented by Ulrich Drepper.
2300 * nscd now also caches the netgroup database.
2301 Implemented by Ulrich Drepper.
2303 * Integrate libm with gcc's -ffinite-math-only option.
2304 Implemented by Ulrich Drepper.
2306 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2307 to math functions. Implemented by Ulrich Drepper.
2309 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2310 Implemented by Ulrich Drepper.
2312 * New Linux interfaces: process_vm_readv, process_vm_writev
2314 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2315 Implemented by Ulrich Drepper.
2317 * Optimized nearbyint and strcasecmp for PPC.
2318 Implemented by Adhemerval Zanella.
2320 * New locales: bho_IN, unm_US, es_CU, ta_LK
2324 * The following bugs are resolved with this release:
2326 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2327 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2328 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2329 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2330 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2331 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2332 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2333 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2334 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2335 12795, 12811, 12813, 12814, 12841
2337 * The RPC implementation in libc is obsoleted. Old programs keep working
2338 but new programs cannot be linked with the routines in libc anymore.
2339 Programs in need of RPC functionality must be linked against TI-RPC.
2340 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2342 Visible changes of this change include (obviously) the inability to link
2343 programs using RPC functions without referencing the TI-RPC library and the
2344 removal of the RPC headers from the glibc headers.
2345 Implemented by Ulrich Drepper.
2347 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2348 syncfs, setns, sendmmsg
2350 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2351 yue_HK, lij_IT, mhr_RU
2353 * New encodings: CP770, CP771, CP772, CP773, CP774
2355 * New program sotruss to trace calls through PLTs
2356 Implemented by Ulrich Drepper.
2358 * The malloc hook implementation is marked deprecated and will be removed
2359 from the default implementation in the next version. The design never
2360 worked ever since the introduction of threads. Even programs which do
2361 not create threads themselves can use multiple threads created internally.
2365 * The following bugs are resolved with this release:
2367 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2368 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2369 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2370 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2373 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2375 * POWER7 optimizations: memset, memcmp, strncmp
2377 * New optimized string functions for x86-64: strnlen (SSE2),
2378 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2379 Implemented by Ulrich Drepper.
2383 * The following bugs are resolved with this release:
2385 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2386 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2387 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2388 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2389 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2390 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2391 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2392 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2394 * New interfaces: pthread_getname_np, pthread_setname_np
2396 * New Linux interface: recvmmsg
2398 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2400 * The dynamic linker now recognizes supported ABI versions from the
2401 EI_ABIVERSION field in the ELF header.
2402 Implemented by Ulrich Drepper.
2404 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2405 will not be used to fill in password fields in the passwd.byname replies.
2406 Instead it is used to synthesize the shadow.byname table, should it be
2407 missing. This is a useful mode in some installations involving Solaris.
2408 Implemented by Ulrich Drepper.
2410 * New locales: kok_IN, sq_MK, cv_RU
2414 * The following bugs are resolved with this release:
2416 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2417 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2418 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2419 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2420 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2421 10789, 10817, 10823, 10840, 10847
2423 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2425 Implemented by Ulrich Drepper.
2427 * Checking version of longjmp added that fails if an uninitialized stack
2428 frame would be created. Implemented by Ulrich Drepper.
2430 * STT_GNU_IFUNC is now supported in static executables.
2431 Implemented by H.J. Lu.
2433 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2435 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2436 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2437 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2438 Contributed by H.J. Lu.
2440 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2441 Implemented by Ulrich Drepper.
2443 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2445 Contributed by H.J. Lu.
2447 * Support for fma instruction in AVX on x86-64.
2448 Implemented by H.J. Lu and Ulrich Drepper.
2450 * AVX support in x86-64 auditing support in ld.so.
2451 Implemented by H.J. Lu.
2453 * STB_GNU_UNIQUE support added.
2454 Implemented by Ulrich Drepper.
2456 * Implement second fallback mode for optimized DNS lookup for even more
2457 broken environments. If two requests from the same port are not
2458 handled correctly close the socket and open a new one before sending
2459 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2460 can be used to select this mode right away, instead of rediscovering the
2461 necessity is every process again.
2462 Implemented by Ulrich Drepper.
2464 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2465 Implemented by Adam Tkac.
2467 * Optimized iconv conversions for S390x.
2468 Implemented by Andreas Krebbel.
2470 * Using condvars with PI mutexes is now more efficient due to kernel
2471 support for requeueing to PI futexes. NPTL support added for x86-64.
2472 Implemented by Ulrich Drepper.
2474 * New locales: ps_AF, my_MM
2478 * The following bugs are resolved with this release:
2480 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2481 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2482 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2483 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2484 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2486 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2487 Implemented by Ulrich Drepper.
2489 * New Linux interfaces: accept4, fallocate, fallocate64.
2490 Implemented by Ulrich Drepper.
2492 * Correct declarations of string function when used in C++ code. This
2493 could lead to compile errors for invalid C++ code.
2495 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2498 * New POSIX 2008 interface: psiginfo
2499 Implemented by Ulrich Drepper.
2501 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2502 Implemented by Ulrich Drepper.
2504 * Support for selecting between multiple function definitions at runtime
2505 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2507 * The libcrypt library can now use the hash function implementations in
2508 NSS. Implemented by Ulrich Drepper.
2510 * The malloc implementation can be compiled to be less memory efficient
2511 but higher performing in multi-threaded programs.
2512 Implemented by Ulrich Drepper.
2514 * New locales: nan_TW@latin, ks_IN
2516 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2517 Implemented by Ulrich Drepper.
2519 * Extended printf hook support. It is possible to use user-defined types
2520 and extend existing format specifiers.
2521 Implemented by Ulrich Drepper.
2523 * Handling for group shadow files has been added.
2524 Implemented by Ulrich Drepper.
2526 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2527 servers (the case, e.g., for some people using the built-in DNS
2528 server in ADSL modems/routers). There is a once-per-process timeout
2529 in case of a broken server. To avoid it, users can run nscd or put
2530 'options single-request' in /etc/resolv.conf.
2531 Implemented by Ulrich Drepper.
2535 * The following bugs are resolved with this release:
2537 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2538 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2539 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2540 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2541 6968, 6974, 6980, 6995, 7008, 7009, 7029
2543 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2544 up at the same time. Implemented by Ulrich Drepper.
2546 * TLS descriptors for LD and GD on x86 and x86-64.
2547 Implemented by Alexandre Oliva.
2549 * getaddrinfo now handles DCCP and UDPlite.
2550 Implemented by Ulrich Drepper.
2552 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2553 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2554 Implemented by Ulrich Drepper.
2556 * New implementation of memmem, strstr, and strcasestr which is O(n).
2557 Implemented by Eric Blake.
2559 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2561 * Implement "e" option for popen to open file descriptor with the
2562 close-on-exec flag set. Implemented by Ulrich Drepper.
2564 * Implement "b" mode for fmemopen. In this mode writes writes don't
2565 implicitly add a NUL byte and seeks from the end of the buffer really
2566 use the buffer end, not the string length as the basis.
2567 Implemented by Ulrich Drepper.
2569 * Many functions, exported and internal, now atomically set the close-on-exec
2570 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2572 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2574 Implemented by Pravin Satpute.
2576 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2580 * The following bugs are resolved with this release:
2582 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2583 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2584 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2585 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2586 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2587 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2588 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2590 * New locales: bo_CN, bo_IN, shs_CA.
2592 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2594 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2595 Implemented by Pravin Satpute.
2597 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2599 * nscd caches now all timeouts for DNS entries
2600 Implemented by Ulrich Drepper.
2602 * nscd is more efficient and wakes up less often.
2603 Implemented by Ulrich Drepper.
2605 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2606 vdprintf, and obstack_vprintf.
2607 Implemented by Jakub Jelinek.
2609 * Faster memset for x86-64.
2610 Implemented by Harsha Jagasia and H.J. Lu.
2612 * Faster memcpy on x86.
2613 Implemented by Ulrich Drepper.
2615 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2616 Implemented by Ulrich Drepper.
2618 * Faster sqrt and sqrtf implemention for some PPC variants.
2619 Implemented by Steven Munroe.
2623 * The following bugs are resolved with this release:
2625 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2626 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2627 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2628 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2629 5071, 5103, 5104, 5112, 5113, 5184, 5186
2631 * More checking functions: fread, fread_unlocked, open*, mq_open.
2632 Implemented by Jakub Jelinek and Ulrich Drepper.
2634 * Extend fortification to C++. Implemented by Jakub Jelinek.
2636 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2637 by not recognizing 'a' as a modifier when those specs are requested.
2638 Implemented by Jakub Jelinek.
2640 * PPC optimizations to math and string functions.
2641 Implemented by Steven Munroe.
2643 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2644 options to be passed. Implemented by Ulrich Drepper.
2646 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2648 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2649 Implemented by Ulrich Drepper.
2651 * Handle private futexes in the NPTL implementation.
2652 Implemented by Jakub Jelinek and Ulrich Drepper.
2654 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2655 Implemented by Roland McGrath and Ulrich Drepper.
2657 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2659 * SHA-256 and SHA-512 based password encryption.
2660 Implemented by Ulrich Drepper.
2662 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2663 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2666 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2667 Implemented by Ulrich Drepper.
2671 * The following bugs are resolved with this release:
2673 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2674 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2675 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2676 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2677 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2678 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2679 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2682 * New Linux interfaces: epoll_pwait, sched_getcpu.
2684 * New generic interfaces: strerror_l.
2686 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2690 * The following bugs are resolved with this release:
2692 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2693 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2694 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2695 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2696 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2697 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2698 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2699 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2700 3225, 3251, 3252, 3253, 3273
2702 * For Linux, the sorting of addresses returned by getaddrinfo now also
2703 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2704 Implemented by Ulrich Drepper.
2706 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2707 Implemented by Ulrich Drepper.
2709 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2711 * New iconv module for MIK. Contributed by Alexander Shopov.
2713 * For sites with broken group and/or passwd database, the auto-propagate
2714 option of nscd can prevent creating ID lookup entries from the results
2715 of a name lookup and vice versa. This usually is no problem but some
2716 site might have problems with the default behavior.
2717 Implemented by Ulrich Drepper.
2719 * Iterating over entire database in NIS can be slow. With the
2720 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2721 to trade time for memory. The entire database will be read at once.
2722 Implemented by Ulrich Drepper.
2724 * The interfaces introduced in RFC 3542 have been implemented by
2727 * Support for the new ELF hash table format was added by Ulrich Drepper.
2729 * Support for priority inheritance mutexes added by Jakub Jelinek and
2732 * Support for priority protected mutexes added by Jakub Jelinek.
2734 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2735 or_IN, csb_PL, fy_NL, sr_ME.
2739 * More overflow detection functions.
2741 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2742 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2744 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2745 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2746 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2747 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2748 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2749 by Masahide Washizawa.
2751 * It is now possible to install audit libraries for the dynamic linker using
2752 LD_AUDIT. Implemented by Ulrich Drepper.
2754 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2755 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2756 For a libc and libpthread that works well on Linux 2.4 kernels, we
2757 recommend using the stable 2.3 branch.
2759 * The new function open_wmemstream acts like open_memstream,
2760 but uses a wchar_t wide character buffer.
2762 * The new function ppoll is an improved version of poll, similar to pselect.
2764 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2765 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2766 renameat, unlinkat, linkat, symlinkat, readlinkat.
2768 * New Linux kernel system calls: unshare,
2769 inotify_init, inotify_add_watch, inotify_rm_watch.
2771 * The euidaccess function is now also known by the alias eaccess,
2772 for compatibility with some other systems.
2774 * Timezone data updated to 2006b version.
2778 * The following bugs are resolved with this release:
2780 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2781 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2782 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2783 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2784 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2785 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2787 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2789 * As of this release, GCC 4 can be used to compile the C Library.
2791 * Timezone data updated to 2005m version.
2795 * The following bugs are resolved with this release:
2797 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2798 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2799 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2800 777, 787, 821, 822, 823, 825
2802 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2806 * Support for RFC 3678. Real implementations exist only for Linux so far.
2807 Implemented by Ulrich Drepper.
2809 * nscd can now cache entries persistently. Expiring entries are reloaded.
2810 For speedups the cache can be shared in memory with client processes.
2811 Implemented by Ulrich Drepper.
2813 * nscd can now perform SELinux checks.
2814 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2816 * getaddrinfo queries are now cached. Canonical name lookup is performed
2818 Implemented by Ulrich Drepper.
2820 * The nothrow function attribute is used when headers are used by gcc when
2821 compiling C code. This can avoid significant amounts of exception
2824 * The malloc functions perform more error checking and are stricter when
2825 it comes to reacting on errors. The default action is to terminate
2826 the process after showing an error message. Implemented by Ulrich Drepper.
2828 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2829 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2831 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2832 namespaces using the new function dlmopen(). This feature is of course,
2833 like most other dynamic loading functionality, not available in statically
2834 linked applications. Implemented by Ulrich Drepper.
2836 * Low-overhead boundary checking variants of string and some stdio functions
2837 were added. These are to be used in conjunction with a gcc patch by
2838 Jakub Jelinek which adds calls to these functions if possible.
2839 Implemented by Jakub Jelinek and Ulrich Drepper.
2841 * Old code for several operating systems and machine architectures that
2842 have not been in working condition in a long time have been removed from
2843 the main source tree maintained by the GNU C Library's maintainers.
2844 These files are now reside in the separate `ports' source module
2845 that is usable as an add-on when building the library.
2849 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2850 interrogate the dynamic linker, compatible with the Solaris interface.
2852 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2853 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2855 * getifaddrs now uses the netlink interface on Linux to get its information.
2856 Implemented by Thorsten Kukuk.
2858 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2859 Implemented by Ulrich Drepper.
2861 * support for non-executable stacks on x86 has been added. Changes mostly
2864 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2867 * getaddrinfo now performs destination address selection according to
2872 * Thread-safe interfaces for many functions that access locale data
2873 were added in version 2.3, but these features were omitted from NEWS.
2874 Many functions have variants with an `_l' suffix that take a `locale_t'
2875 object as a parameter rather than consulting the current locale.
2876 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2877 create and maintain `locale_t' objects. Additionally, the new function
2878 `uselocale' sets "the current locale" (as used by functions not so
2879 parameterized) set for an individual thread. These features were added
2880 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2882 * The functions getresuid, getresgid, setresuid, and setresgid, which
2883 have long been available on Linux, are now declared in <unistd.h>
2884 and are now also available on the Hurd.
2886 * ELF thread-local storage support (TLS) now works on x86-64.
2888 * The new dynamic string token $LIB is expanded in shared library names.
2889 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2891 * Aldy Hernandez contributed complete software floating point support for
2892 PowerPC machines with no FPU.
2894 * fexecve is implemented on Linux.
2896 * The `btowc' function should work at least twice as fast due to
2897 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2899 * With approriate thread add-ons cancelable functions are now implemented
2900 in libc.so as well. No need to call the function in libpthread. This
2901 change allowed to finally disable the incorrect and expensive handling
2902 of weak definition in ld.so.
2904 * Yet more PLT entries in libc.so have been removed. We finally arrived
2905 at the bare minimum. Startup times improved appropriately.
2907 * Support for the new Linux/x86 system call interface was added. The
2908 AT_SYSINFO auxiliary vector entry is recognized and handled.
2912 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2915 * iconv (the program and the interface) now accepts empty names (excluding
2916 options like //TRANSLIT) to mean "use charset of current locale".
2918 * localedef can now transliterate characters in strings which are not in
2919 the provided charmap. The information from the input locale is used.
2921 * Prelinking support was added for ELF targets. This requires additional
2922 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2924 * Read-only stdio streams now use mmap to speed up operation by eliminating
2925 copying and buffer underflows. To use add 'm' to the mode string of
2926 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2928 * The malloc functions were completely rewritten by Wolfram Gloger based
2929 on Doug Lea's malloc-2.7.0.c.
2931 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2932 implementation of regex.
2934 * Bruno Haible upgraded the iconv and locale implementation to support
2937 * Contents of the LC_* and LANG environment variables in the CEN style are
2938 not recognized anymore. It never was used. Change by Ulrich Drepper.
2940 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2941 thread-local storage (TLS) ABI on some platforms.
2942 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2944 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2945 EUC-JISX0213, and TSCII.
2947 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2948 BSD-compatible interface for getting all network interface addresses.
2949 Implementation for IPv4 by Roland McGrath.
2951 * Loading of locale data is faster due to the introduction of a locale
2952 archive. Implemented by Roland McGrath and Ulrich Drepper.
2954 * Startup times are significantly reduced by not using exported functions
2955 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2958 * Steven Munroe contributed a port to PowerPC64/Linux.
2962 * The Hurd now uses the GNU libio implementation of stdio rather than the
2963 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2965 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2966 with the same behavior as the Linux system call of the same name.
2970 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2971 128-bit long double format.
2973 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2974 IBM1161, and IBM1162 charsets.
2976 * Andreas Jaeger contributed a port to x86-64/Linux.
2978 * Peter Bruin contributed a port to PowerPC/Hurd.
2980 * libc functions using I/O streams now can handle wide-oriented streams
2983 * optimizations in the dynamic linker. Binaries created by recent binutils
2984 versions start up quicker due to reduced time spent on relocations.
2986 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2990 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2991 asinh, atanh, j0 for the 128-bit long double format.
2993 * Bruno Haible updated all the code handling Unicode in some form to
2994 support Unicode 3.1.
2996 * Speed of regex for single-byte locales is back to previous levels.
2997 Patch by Isamu Hasegawa.
2999 * Alpha, SPARC, and IA-64 now also using floating stacks.
3001 * Startup time of internationalized applications greatly improved through
3002 iconv cache. Use iconvconfig to generate the cache file.
3003 Contributed by Ulrich Drepper.
3005 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3006 severe performance problems. Patch by David Mosberger.
3008 * The Hurd port got a lot more functionality like AIO, various stdio
3009 extensions, etc. Mainly done by Roland McGrath.
3011 * mtrace can now lookup symbols in shared libraries.
3015 * Intel's IA-64 math library is largely integrated. It provides fast and
3016 accurate implementatations for most basic and standard math functions
3017 in float, double, and long double format.
3019 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3020 and asin for the 96-bit long double format and asin, log, tan for the
3021 128-bit long double format.
3023 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3024 The basic double functions exist today. Contributed by Abraham Ziv
3025 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3026 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3028 * An asynchronous name lookup library was added. The interface is designed
3029 after POSIX AIO. The proposal was circulated beforehand to get comments.
3030 No negative ones came in. Implemented by Ulrich Drepper.
3032 * Port to S390/64bit contributed by Martin Schwidefsky
3033 <schwidefsky@de.ibm.com>.
3035 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3036 of functions for Linux/IA-64.
3038 * The RPC code is now thread safe. Threads can now use the same service
3039 of different services at the same time. Patch by Eric Norum
3040 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3042 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3043 family of functions for Linux/S390.
3045 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3046 of functions for Linux/x86.
3048 * Port to Linux/CRIS contributed by Axis Communications.
3052 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3053 we can now check for namespace violations and missing declarations. The
3054 result is that almost all headers are now Unix-compliant (as defined in
3055 the upcoming XPG6). The negative side is that some programs might need
3056 corrections, too, if they depend on the incorrect form of the headers in
3057 previous versions which defined too many symbols and included too many
3060 * regex now handles multibyte character sets correctly.
3061 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3063 * iconv (the program) does now conform to the upcoming XPG6 and handles
3064 charmaps. Instead of the charset names the path of charmaps can be
3065 provided and the conversion happens based on this data.
3066 Contributed by Ulrich Drepper.
3068 * The locale program now provides detailed information about the installed
3073 only lists the names of the supported locales
3077 provides details such as country, language, and codeset name.
3078 Contributed by Ulrich Drepper.
3082 * The gencat program now parses the input file according to the charset
3083 selected by the LC_CTYPE category. This is important for stateful
3084 character sets. To make generating catalogs easier there is a way
3085 to overwrite the charset selected by the locale: before the first
3086 message or $ quote line the catalog can contain a line like
3088 $ codeset=ISO-8859-2
3090 to select the charset (ISO-8859-2 in this case).
3092 Implemented by Shinya Hanataka and Ulrich Drepper.
3094 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3095 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3096 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3098 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3099 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3101 * The LANGUAGE environment variable is now ignored unless the locale is
3102 changed from the default "C" locale.
3104 * The usual bug fixes.
3108 * Greg McGary added runtime support for bounds checking using gcc's
3109 new -fbounded-pointers option. ix86 target is complete. PowerPC
3112 * Thorsten Kukuk added secure mode support to nscd.
3114 * The Berkeley db libraries have been removed.
3116 Related, the nss_db module is now in a separate package since it
3117 obviously requires a database library being available.
3119 * Wide character I/O streams implemented by Ulrich Drepper.
3121 * Functions from the extended socket API added by Ulrich Drepper.
3123 * Functions feenableexcept and fedisableexcept to control the
3124 behaviour of individual exceptions have been added by Andreas Jaeger.
3126 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3128 * The resolver code has been updated from bind 8.2.3-T5B which supports
3129 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3132 This change could in some situations effect backward compatibility. Since
3133 now `_res' is a thread-local instead of a global variable, modifying it
3134 in one thread does not have any effect in other threads.
3136 The resolver library was also extended to allow IPv6 as the transport
3137 protocol for the requests. This work was done by Stig Venaas.
3139 * Compatibility code for K&R C compilers has been removed from the
3140 header files. A ISO C compiler is needed to use the library
3141 (conforming to either C89 or C99 standard).
3143 * Complete rewrite of the localedef program to support multibyte character
3144 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3145 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3146 Implemented by Ulrich Drepper.
3148 Bruno Haible significantly improved the generation and use of the data
3149 structures for the wide character tables.
3151 * Plural handling in gettext implemented by Ulrich Drepper.
3153 * The utmp daemon has been removed.
3155 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3157 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3160 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3162 * POSIX barriers implemented by Kaz Kylheku.
3164 * POSIX spawn function family implemented by Ulrich Drepper.
3166 * POSIX spinlocks are now available.
3168 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3170 * the configure option --enable-kernel=X.Y.Z allows to strip out
3171 compatibility for kernel versions before X.Y.Z. This is currently only
3172 implemented for Linux.
3174 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3175 field sin6_scope_id. This means that all programs using IPv6 should be
3176 recompiled. Don't expect binary compatibility with previous glibc
3179 * various conversion modules for IBM character sets contributed by
3182 * IA-64 port by Jes Sorensen and HJ Lu.
3196 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3198 * New xdr functions are added; some rpc functions are now 64bit clean.
3200 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3202 * Fixed known incompatibilities with glibc 2.0.
3204 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3206 * Optimized a number of functions (especially the ELF dynamic loader).
3208 * Update timezone data files.
3210 * lots of charmaps corrections
3212 * some new locale definitions and charmaps
3217 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3218 conform to POSIX member type requirements. Symbol versions have been
3219 adjusted as needed within the library, and for direct use by applications,
3220 but there is potential for problems if third-party libraries use
3221 struct timeval as part of their interface. This does not present
3222 a problem for X and other "essential" system libraries.
3224 * An additional locale model to support C++ Standard Library locale
3225 model and probably more was implemented by Ulrich Drepper.
3227 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3230 * Miles Bader provided the `argp' function family to support hierachical
3231 command line argument parsing, layered on top of getopt.
3233 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3235 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3238 * scanf recognizes the %a and %A format for scanning floating point numbers.
3240 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3241 information and interfaces for the available integer types.
3243 * about 130 new math functions were added to implement the ISO C9x math
3246 * the new header <complex.h> contains definitions of the complex math
3247 functions from ISO C 9X.
3249 * the new header <tgmath.h> defines generic macros to use complex or
3250 real valued functions.
3252 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3254 * Andreas Jaeger provided a test suite for the math library.
3256 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3258 * Ulrich Drepper added character set conversion functions (iconv).
3260 * Optimized string functions have been added.
3262 * The localedata addon is now part of glibc.
3264 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3266 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3267 daemon for NSS (nscd).
3269 Missing a better place here are some numbers on improvements. Under
3270 Linux 2.1.125 un-tar-ing the kernel sources takes
3274 using local files 12.19s 6.88s 22.91s
3276 using NIS 13.92s 8.91s 26.34s
3278 using NIS & nscd 10.37s 7.34s 25.30s
3280 using NIS+ 27.57s 30.37s 640.46s
3282 using NIS+ & nscd 10.25s 7.83s 26.51s
3284 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3286 Keep in mind that non-namelookup related operations dominate above times.
3287 It was just a common complain that using NIS+ unpacking the kernel is
3290 [1] The old nscd implementation is not available anymore since it was
3291 distributed with glibc up to version 2.0.98 and thus is now replaced.
3293 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3295 * Mark Kettenis provided a Hesiod NSS module.
3297 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3298 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3300 * rcmd can now handle netgroups (Dick Streefland).
3302 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3305 * Support for the IPv6 protocol has been added to the socket API, as per the
3306 latest draft standards.
3308 * Support for Linux 2.2 has been added.
3310 * Interface changes relative to the latest 2.0.x release:
3311 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3312 addseverity NEW: Unix98
3313 alphasort64 NEW: LFS
3314 argp_err_exit_status NEW: argp, GNU ext
3315 argp_error NEW: argp, GNU ext
3316 argp_failure NEW: argp, GNU ext
3317 argp_help NEW: argp, GNU ext
3318 argp_parse NEW: argp, GNU ext
3319 argp_program_bug_address NEW: argp, GNU ext
3320 argp_program_version NEW: argp, GNU ext
3321 argp_program_version_hook NEW: argp, GNU ext
3322 argp_state_help NEW: argp, GNU ext
3323 argp_usage NEW: argp, GNU ext
3324 authdes_create NEW: Secure RPC
3325 authdes_getucred NEW: Secure RPC
3326 authdes_pk_create NEW: Secure RPC
3327 backtrace NEW: GNU ext.
3328 backtrace_symbols NEW: GNU ext.
3329 backtrace_symbols_fd NEW: GNU ext.
3331 cacosf NEW: ISO C 9x
3332 cacosh NEW: ISO C 9x
3333 cacoshf NEW: ISO C 9x
3334 cacoshl NEW: ISO C 9x
3335 cacosl NEW: ISO C 9x
3342 casinf NEW: ISO C 9x
3343 casinh NEW: ISO C 9x
3344 casinhf NEW: ISO C 9x
3345 casinhl NEW: ISO C 9x
3346 casinl NEW: ISO C 9x
3348 catanf NEW: ISO C 9x
3349 catanh NEW: ISO C 9x
3350 catanhf NEW: ISO C 9x
3351 catanhl NEW: ISO C 9x
3352 catanl NEW: ISO C 9x
3353 cbc_crypt NEW: Secure RPC
3357 ccoshf NEW: ISO C 9x
3358 ccoshl NEW: ISO C 9x
3364 cimagf NEW: ISO C 9x
3365 cimagl NEW: ISO C 9x
3366 clearerr_locked REMOVED
3367 clntunix_create NEW: sunrpc ext
3369 clog10 NEW: ISO C 9x
3370 clog10f NEW: ISO C 9x
3371 clog10l NEW: ISO C 9x
3381 cprojf NEW: ISO C 9x
3382 cprojl NEW: ISO C 9x
3384 crealf NEW: ISO C 9x
3385 creall NEW: ISO C 9x
3390 csinhf NEW: ISO C 9x
3391 csinhl NEW: ISO C 9x
3394 csqrtf NEW: ISO C 9x
3395 csqrtl NEW: ISO C 9x
3399 ctanhf NEW: ISO C 9x
3400 ctanhl NEW: ISO C 9x
3402 des_setparity NEW: Secure RPC
3403 ecb_crypt NEW: Secure RPC
3404 endutxent NEW: Unix98
3406 exp10f NEW: ISO C 9x
3407 exp10l NEW: ISO C 9x
3411 fattach NEW: STREAMS
3412 fdetach NEW: STREAMS
3416 feclearexcept NEW: ISO C 9x
3417 fegetenv NEW: ISO C 9x
3418 fegetexceptflag NEW: ISO C 9x
3419 fegetround NEW: ISO C 9x
3420 feholdexcept NEW: ISO C 9x
3422 feraiseexcept NEW: ISO C 9x
3423 ferror_locked REMOVED
3424 fesetenv NEW: ISO C 9x
3425 fesetexceptflag NEW: ISO C 9x
3426 fesetround NEW: ISO C 9x
3427 fetestexcept NEW: ISO C 9x
3428 feupdateenv NEW: ISO C 9x
3429 fflush_locked REMOVED
3433 fgets_unlocked NEW: GNU ext.
3434 fileno_locked REMOVED
3446 fputc_locked REMOVED
3447 fputs_unlocked NEW: GNU ext.
3448 fread_unlocked NEW: GNU ext.
3453 fstatvfs NEW: Unix98
3457 ftruncate64 NEW: LFS
3459 fwrite_unlocked NEW: GNU ext.
3460 gai_strerror NEW: IPv6
3464 getchar_locked REMOVED
3466 getdate_err NEW: Unix98
3467 getdate_r NEW: GNU ext.
3469 getnameinfo NEW: IPv6
3470 getnetname NEW: Secure RPC
3471 getpmsg NEW: STREAMS
3472 getpt NEW: Unix98 PTY
3473 getrlimit64 NEW: LFS
3474 getutxent NEW: Unix98
3475 getutxid NEW: Unix98
3476 getutxline NEW: Unix98
3477 glob64 NEW: GNU ext.
3478 globfree64 NEW: GNU ext.
3479 gnu_get_libc_release NEW: GNU ext.
3480 gnu_get_libc_version NEW: GNU ext.
3481 grantpt NEW: Unix98 PTY
3482 host2netname NEW: Secure RPC
3484 iconv_close NEW: iconv
3485 iconv_open NEW: iconv
3486 if_freenameindex NEW: IPv6
3487 if_indextoname NEW: IPv6
3488 if_nameindex NEW: IPv6
3489 if_nametoindex NEW: IPv6
3490 in6addr_any NEW: IPv6
3491 in6addr_loopback NEW: IPv6
3492 inet6_isipv4mapped NEW: IPv6
3493 isastream NEW: STREAMS
3494 iswblank NEW: GNU ext.
3495 key_decryptsession NEW: Secure RPC
3496 key_decryptsession_pk NEW: Secure RPC
3497 key_encryptsession NEW: Secure RPC
3498 key_encryptsession_pk NEW: Secure RPC
3499 key_gendes NEW: Secure RPC
3500 key_get_conv NEW: Secure RPC
3501 key_secretkey_is_set NEW: Secure RPC
3502 key_setnet NEW: Secure RPC
3503 key_setsecret NEW: Secure RPC
3504 llrint NEW: ISO C 9x
3505 llrintf NEW: ISO C 9x
3506 llrintl NEW: ISO C 9x
3507 llround NEW: ISO C 9x
3508 llroundf NEW: ISO C 9x
3509 llroundl NEW: ISO C 9x
3514 lrintf NEW: ISO C 9x
3515 lrintl NEW: ISO C 9x
3516 lround NEW: ISO C 9x
3517 lroundf NEW: ISO C 9x
3518 lroundl NEW: ISO C 9x
3520 makecontext NEW: Unix98
3521 mempcpy NEW: GNU ext.
3524 modify_ldt NEW: kernel
3528 nearbyint NEW: ISO C 9x
3529 nearbyintf NEW: ISO C 9x
3530 nearbyintl NEW: ISO C 9x
3531 netname2host NEW: Secure RPC
3532 netname2user NEW: Secure RPC
3533 nexttoward NEW: ISO C 9x
3534 nexttowardf NEW: ISO C 9x
3535 nexttowardl NEW: ISO C 9x
3539 passwd2des NEW: Secure RPC
3541 pow10f NEW: GNU ext.
3542 pow10l NEW: GNU ext.
3545 printf_size NEW: GNU ext.
3546 printf_size_info NEW: GNU ext.
3547 profil_counter REMOVED
3548 pthread_mutexattr_getkind_np REPLACED
3549 pthread_mutexattr_setkind_np REPLACED
3550 ptsname NEW: Unix98 PTY
3551 ptsname_r NEW: Unix98 PTY
3553 putchar_locked REMOVED
3554 putgrent NEW: GNU ext.
3556 putpmsg NEW: STREAMS
3557 pututxline NEW: Unix98
3561 readdir64_r NEW: LFS
3562 remquo NEW: ISO C 9x
3563 remquof NEW: ISO C 9x
3564 remquol NEW: ISO C 9x
3566 roundf NEW: ISO C 9x
3567 roundl NEW: ISO C 9x
3569 scalbln NEW: ISO C 9x
3570 scalblnf NEW: ISO C 9x
3571 scalblnl NEW: ISO C 9x
3573 sendfile NEW: kernel
3574 setcontext NEW: Unix98
3575 setrlimit64 NEW: LFS
3576 setutxent NEW: Unix98
3578 sigignore NEW: Unix98
3579 sigqueue NEW: POSIX.1b
3580 sigrelse NEW: Unix98
3581 sigset NEW: POSIX.1b
3582 sigtimedwait NEW: POSIX.1b
3583 sigwaitinfo NEW: POSIX.1b
3584 sincos NEW: GNU ext.
3585 sincosf NEW: GNU ext.
3586 sincosl NEW: GNU ext.
3590 strcasestr NEW: GNU ext.
3591 strtoimax NEW: ISO C 9x
3592 strtoumax NEW: ISO C 9x
3593 strverscmp NEW: GNU ext.
3594 svcauthdes_stats NEW: Secure RPC
3595 svcunix_create NEW: sunrpc etx
3596 svcunixfd_create NEW: sunrpc ext
3597 swapcontext NEW: Unix98
3598 tcgetsid NEW: Unix98 PTY
3599 tdestroy NEW: GNU ext.
3600 tgamma NEW: ISO C 9x
3601 tgammaf NEW: ISO C 9x
3602 tgammal NEW: ISO C 9x
3606 truncf NEW: ISO C 9x
3607 truncl NEW: ISO C 9x
3609 unlockpt NEW: Unix98 PTY
3610 updwtmpx NEW: Unix98
3611 user2netname NEW: Secure RPC
3612 utmpxname NEW: Unix98
3613 versionsort NEW: GNU ext.
3614 versionsort64 NEW: GNU ext.
3616 wcscasecmp NEW: GNU ext.
3617 wcsncasecmp NEW: GNU ext.
3618 wcsnlen NEW: GNU ext.
3619 wcstoimax NEW: ISO C 9x
3620 wcstoll NEW: ISO C 9x
3621 wcstoull NEW: ISO C 9x
3622 wcstoumax NEW: ISO C 9x
3624 wordexp NEW: POSIX.2
3625 wordfree NEW: POSIX.2
3626 write_profiling REMOVED
3627 xdecrypt NEW: Secure RPC
3628 xdr_authdes_cred NEW: Secure RPC
3629 xdr_authdes_verf NEW: Secure RPC
3630 xdr_cryptkeyarg NEW: Secure RPC
3631 xdr_cryptkeyarg2 NEW: Secure RPC
3632 xdr_cryptkeyres NEW: Secure RPC
3633 xdr_getcredres NEW: Secure RPC
3634 xdr_key_netstarg NEW: Secure RPC
3635 xdr_key_netstres NEW: Secure RPC
3636 xdr_keybuf NEW: Secure RPC
3637 xdr_keystatus NEW: Secure RPC
3638 xdr_netnamestr NEW: Secure RPC
3639 xdr_sizeof NEW: Secure RPC
3640 xdr_unixcred NEW: sunrpc ext
3641 xencrypt NEW: Secure RPC
3642 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3653 * inet_ntoa is thread-safe
3655 * updwtmp is moved from libutil to libc
3657 * rewrite of cbrt function
3659 * update of timezone data
3673 * add atoll function
3675 * fix complex problems in Berkeley DB code
3677 * fix math functions
3681 * fixed lots of header problems (especially Linux/GNU specific)
3683 * dynamic loader preserves all registers
3685 * Roland McGrath provided support for handling of auxiliary objects in
3686 the ELF dynamic loader.
3688 * support for parallel builds is improved
3692 * GNU extensions are no longer declared by default. To enable them you
3693 must define the macro `_GNU_SOURCE' in your program or compile with
3696 * The library has changed from using GNU ld symbol aliases to using weak
3697 symbols where available. The ELF object file format supports weak
3698 symbols; GNU ld also supports weak symbols in the a.out format. (There
3699 is also now support for other GNU ld extensions in ELF. Use the
3700 `--with-elf' option to configure to indicate you have ELF, and
3701 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3702 of many files which contained only symbol aliases, reducing the size of
3703 the source and the compiled library; many other files were renamed to
3704 less cryptic names previously occupied by the symbol alias files.
3705 There is a new header file <elf.h> for programs which operate on
3706 files in the ELF format.
3708 * Converted to Autoconf version 2, so `configure' has more options.
3709 Run `configure --help' to see the details.
3711 * The library can now be configured to build profiling, highly-optimized
3712 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3713 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3714 `configure' enable building these extra libraries. The shared library is
3715 built by default when using both ELF and GNU ld. When shared libraries
3716 are enabled, the new library `-ldl' is available for arbitrary run-time
3717 loading of shared objects; its interface is defined in <dlfcn.h>. The
3718 new header file <link.h> gives access to the internals of the run-time
3719 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3720 application of same name on other systems and it provides information
3721 about dynamically linked binaries.
3723 * The C library now provides the run-time support code for profiling
3724 executables compiled with `-pg'. Programs can control the profiling code
3725 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3726 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3727 the support for this file format was contributed by David Mosberger-Tang.
3729 * The math code has been replaced with a math library based on fdlibm from
3730 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3731 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3732 functions. The math functions now reside in a separate library, so
3733 programs using them will need to use `-lm' their linking commands.
3735 * John C. Bowman contributed optimized ix87 assembler inline functions.
3737 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3738 mechanism similar to that found in Solaris 2. This is now used for the
3739 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3740 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3741 file controls what services are used for each individual database. This
3742 works by loading shared libraries with names specified in `nsswitch.conf',
3743 so service modules can be changed or added at any time without even
3744 relinking any program. Currently there are the file, db, and NIS based
3745 NSS services available.
3747 * The new functions `strtoq' and `strtouq' parse integer values from
3748 strings, like `strtol' and `strtoul', but they return `long long int' and
3749 `unsigned long long int' values, respectively (64-bit quantities).
3751 * The new functions `strtof' and `strtold' parse floating-point values from
3752 strings, like `strtod', but they return `float' and `long double' values,
3753 respectively (on some machines `double' and `long double' are the same).
3755 * Ulrich Drepper has contributed new implementations of the floating-point
3756 printing and reading code used in the `printf' family of functions and
3757 `strtod', `strtof', and `strtold'. These new functions are perfectly
3758 accurate, and much faster than the old ones.
3760 * The implementation of the POSIX locale model was completely rewritten by
3761 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3762 to compile the POSIX locale definition.
3764 * The former dummy implementations of the strcoll and strxfrm function are
3765 now replaced by fully functional code contributed by Ulrich Drepper. The
3766 collation information comes from the POSIX locale definitions.
3768 * The new header <langinfo.h> defines an interface for accessing
3769 various locale-dependent data (using the locale chosen with `setlocale').
3771 * Ulrich Drepper has contributed a new suite of functions for operation on
3772 wide-character and multibyte-character strings, in <wchar.h>;
3773 and classification and case conversion of wide characters, in <wctype.h>.
3774 These new functions are conforming to the ISO C, Amendement 1 specification.
3776 * There is now a second implementation of the standard I/O library available.
3777 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3778 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3780 * You can now use positional parameter specifications in format strings
3781 for the `printf' and `scanf' families of functions. For example,
3782 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3783 ``Number 6, Mr Jones''. This is mainly useful when providing different
3784 format strings for different languages, whose grammars may dictate
3785 different orderings of the values being printed. To support this
3786 feature, the interface for `register_printf_handler' has changed; see
3787 the header file <printf.h> for details.
3789 * The `printf' and `scanf' families of functions now understand a new
3790 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3791 to group numbers as indicated by the locale; for `scanf' and friends, this
3792 says to accept as valid only a number with all the proper grouping
3793 separators in the right places. In the default "C" locale, numbers are
3794 not grouped; but locales for specific countries will define the usual
3795 conventions (i.e. separate thousands with `,' in the US locale).
3797 * The pgrp functions have been regularized, slightly incompatibly but much
3798 less confusingly. The core functions are now `getpgid' and `setpgid',
3799 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3800 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3801 provided for compatibility. There is no longer an incompatible `getpgrp'
3802 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3804 * The new header file <fts.h> and suite of functions simplify programs that
3805 operate on directory trees. This code comes from 4.4 BSD.
3807 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3808 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3809 NSS scheme used in glibc.
3811 * There is a new malloc debugging hook `__memalign_hook'.
3813 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3814 `unsigned int' in <sys/types.h>. These are for compatibility only and
3815 their use is discouraged.
3817 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3818 done differently, so that it works even without GNU ld.
3820 * New function `euidaccess' checks allowed access to a file like `access',
3821 but using the effective IDs instead of the real IDs.
3823 * The time zone data files have been updated for the latest and greatest
3824 local time conventions of the countries of the world.
3826 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3829 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3830 for formatting floating-point numbers. They are provided only for
3831 compatibility; new programs should use `sprintf' instead. There are
3832 also equivalent function for the `long double' floating-point type and
3833 all functions also exist in a reentrant form.
3835 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3836 functions for maintaining the login-record files (primarily of use to
3837 system programs such as `login'), and convenient functions for
3838 allocating and initializing a pseudo-terminal (pty) device.
3840 * Ulrich Drepper has contributed new support for System V style
3841 shared memory and IPC on systems that support it.
3843 * Ulrich Drepper has contributed several miscellaneous new functions found
3844 in System V: The `hsearch' family of functions provide an effective
3845 implementation of hash tables; `a64l' and `l64a' provide a very simple
3846 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3849 * Ulrich Drepper has contributed new reentrant counterparts for the
3850 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3852 * Ulrich Drepper has contributed new, highly-optimized versions of several
3853 string functions for the i486/Pentium family of processors.
3855 * Ulrich Drepper has updated the Linux-specific code, based largely
3856 on work done in Hongjiu Lu's version of GNU libc for Linux.
3857 The GNU library now supports Linux versions 2.0.10 and later,
3858 using the ELF object file format (i[3456]86-*-linux).
3860 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3862 * David Mosberger-Tang and Richard Henderson have ported the C library
3863 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3864 dynamic linking support for ELF/Alpha.
3866 * Richard Henderson contributed several Alpha optimized assembler function
3867 for arithmetic and string handling.
3869 * Ulrich Drepper has contributed a new set of message catalog functions to
3870 support multiple languages using the <libintl.h> interface, for use with
3871 his new package GNU gettext. Translation volunteers have contributed
3872 catalogs of the library's messages in Spanish, German, and Korean.
3874 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3875 program and the `catgets' function for reading the catalog files it
3876 creates. (The <libintl.h> interface is preferred; we include the
3877 <nl_types.h> interface using `catgets' only for source compatibility with
3878 programs already written to use it.)
3880 * New header file <values.h> gives SVID-compatible names for <limits.h>
3883 * Various new macros, declarations, and small header files for compatibility
3886 * New function `group_member' is a convenient way to check if a process has
3887 a given effective group ID.
3889 * When using GCC 2.7 and later, the socket functions are now declared in a
3890 special way so that passing an argument of type `struct sockaddr_in *',
3891 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3892 `struct sockaddr *' type, does not generate a type-clash warning.
3894 * New function `error' declared in header file <error.h> is a convenient
3895 function for printing error messages and optionally exiting; this is the
3896 canonical function used in GNU programs. The new functions `err', `warn',
3897 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3898 doing the same thing.
3900 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3901 POSIX.2 `glob' function to do ~ and {...} expansion.
3903 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3904 `clearenv' which is used in POSIX.9 is also available.
3906 * New function `getsid' returns session ID number on systems that support it.
3908 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3909 files <db.h> and <mpool.h> provide a rich set of functions for several
3910 types of simple databases stored in memory and in files, and <ndbm.h> is
3911 an old `ndbm'-compatible interface using the `db' functions. Link with
3912 `-ldb' to get these functions.
3914 * New macro `strdupa' copies a string like `strdup', but uses local stack
3915 space from `alloca' instead of dynamic heap space from `malloc'.
3917 * New function `strnlen' is like `strlen' but searches only a given maximum
3918 number of characters for the null terminator. `stpncpy', `strndup' and
3919 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3922 * New function `statfs' in header <sys/statfs.h>.
3924 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3925 provide convenient functions for operating on blocks of null-terminated
3928 * A new suite of functions in <utmp.h> handle all the details of reading
3929 and writing the utmp file.
3931 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3934 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3935 completely POSIX compliant and also implemented the extended functionality
3936 to handle alternate digit representation and alternate era date formats.
3938 * Ulrich Drepper provided an implementation of the `strptime' function
3939 defined in XPG4.2 which transforms a string into a `struct tm' value.
3941 * Paul Eggert provided the tzselect shell script as part of the timezone
3942 code. The shell script makes it easy to select the correct timezone
3945 * The implementation of the malloc family of functions is completely replaced
3946 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3947 The implementation uses the mmap function (if available) and it is
3948 optimized for the use in multi threaded programs.
3950 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3951 functions. This new functionality is usable by specifying a special
3952 salt string and it is compatible with implementation on *BSD systems.
3954 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3955 `getsubopt' to handle second level command line options, `bsd_signal'
3956 to access BSD style `signal' functionality, the obsolete `regexp' style
3959 * the `lchown' function is available on system which support this
3962 * The implementation of the shadow password handling function was contributed
3965 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3967 * POSIX.1g support was added. The <sys/select.h> header is available,
3968 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3969 implementation of `getaddrinfo'.
3973 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3975 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3976 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3978 * The new function `daemon' from 4.4 BSD is useful for server programs that
3979 want to put themselves in the background.
3981 * Joel Sherrill has contributed support for several standalone boards that
3982 run without an operating system.
3984 * `printf', `scanf' and friends now accept a `q' type modifier for long
3985 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3987 * All of the code taken from BSD (notably most of the math and networking
3988 routines) has been updated from the BSD 4.4-Lite release.
3990 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3992 * The new functions `getdomainname' and `setdomainname' fetch or change the
3993 YP/NIS domain name. These are system calls which exist on systems which
3996 * The time zone data files have been updated for the latest international
3999 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4000 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4004 * The C library now includes support for Sun RPC, from Sun's free
4005 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4006 are included. (There is still no support for YP.)
4008 * Tom Quinn has contributed a port of the C library to SGI machines running
4009 Irix 4 (mips-sgi-irix4).
4011 * The new `lockf' function is a simplified interface to the locking
4012 facilities of `fcntl', included for compatibility.
4014 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4016 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4018 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4021 * The error message printed by `assert' for a failed assertion now includes
4022 the name of the program (if using GNU ld) and the name of the calling
4023 function (with versions of GCC that support this).
4025 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4027 * The library now includes the <sys/mman.h> header file and memory
4028 management functions `mmap', `munmap', `mprotect', `msync', and
4029 `madvise', on systems that support those facilities.
4031 * The interface for `mcheck' has changed slightly: the function called to
4032 abort the program when an allocation inconsistency is detected now takes
4033 an argument that indicates the type of failure. The new function
4034 `mprobe' lets you request a consistency check for a particular block at
4035 any time (checks are normally done only when you call `free' or `realloc'
4038 * It is now possible to easily cross-compile the C library, building on one
4039 system a library to run on another machine and/or operating system. All
4040 you need to do is set the variable `HOST_CC' in `configparms' to the
4041 native compiler for programs to run on the machine you are building on (a
4042 few generator programs are used on Unix systems); set `CC' to the
4045 * The new function `fexecve' (only implemented on the GNU system) executes
4046 a program file given a file descriptor already open on the file.
4050 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4051 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4053 * You can set the variable `libprefix' in `configparms' to specify a prefix
4054 to be prepended to installed library files; this makes it easy to install
4055 the GNU C library to be linked as `-lgnuc' or whatever.
4057 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4058 copies a limited number of characters from a string, and returns the
4059 address of the last character written.
4061 * You no longer need to check for whether the installed `stddef.h' is
4062 compatible with the GNU C library. configure now checks for you.
4064 * You can now define a per-stream `fileno' function to convert the
4065 stream's cookie into an integral file descriptor.
4067 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4068 allocates zero bytes of storage, and returns a unique pointer which
4069 you can pass to `realloc' or `free'. The behavior is undefined if
4070 you dereference this pointer.
4072 * The C library now runs on Sony NEWS m68k machines running either
4073 NewsOS 3 or NewsOS 4.
4075 * The new `syscall' function is a system-dependent primitive function
4076 for invoking system calls. It has the canonical behavior on Unix
4077 systems, including unreliable return values for some calls (such as
4078 `pipe', `fork' and `getppid').
4080 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4081 to `EAGAIN', which is the preferred name. On systems whose kernels
4082 use two distinct codes, the C library now translates EWOULDBLOCK to
4083 EAGAIN in every system call function.
4087 * The GNU C Library Reference Manual is now distributed with the library.
4088 `make dvi' will produce a DVI file of the printed manual.
4089 `make info' will produce Info files that you can read on line using C-h i
4090 in Emacs or the `info' program.
4091 Please send comments on the manual to bug-glibc-manual@gnu.org.
4093 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4095 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4097 * Jason Merrill has contributed a port to the Sequent Symmetry running
4098 Dynix version 3 (i386-sequent-dynix).
4100 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4101 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4103 * New function `memory_warnings' lets you arrange to get warnings when
4104 malloc is running out of memory to allocate, like Emacs gives you.
4106 * The C library now contains the relocating allocator used in Emacs 19 for
4107 its editing buffers. This allocator (ralloc) minimizes allocation
4108 overhead and fragmentation by moving allocated regions around whenever it
4109 needs to. You always refer to a ralloc'd region with a "handle" (a
4110 pointer to a pointer--an object of type `void **').
4112 * There is a new `printf' format: `%m' gives you the string corresponding
4113 to the error code in `errno'.
4115 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4116 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4117 pass, the `a' modifier says to fill in a `char **' you pass with a
4120 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4121 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4122 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4124 * `mkstemp' is a traditional Unix function to atomically create and open a
4125 uniquely-named temporary file.
4129 * The standard location for the file that says what the local timezone is
4130 has changed again. It is now `/usr/local/etc/localtime' (or more
4131 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4133 * The distribution no longer contains any files with names longer than 14
4136 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4137 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4139 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4143 * The configuration process has changed quite a bit. The `configure'
4144 script is now used just like the configuration scripts for other GNU
4145 packages. The `sysdeps' directory hierarchy is much rearranged.
4146 The file `INSTALL' explains the new scheme in detail.
4148 * The header files no longer need to be processed into ANSI C and
4149 traditional C versions. There is just one set of files to install, and
4150 it will work with ANSI or old C compilers (including `gcc -traditional').
4152 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4153 MIPS DECStation running Ultrix 4.
4155 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4156 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4157 made itself into a shared library.
4159 * Yet further improved support for the i386, running 4.3 BSD-like systems
4160 (such as Mach 3 with the Unix single-server), or System V.
4162 * New function `strncasecmp' to do case-insensitive string comparison
4163 with limited length.
4165 * New function `strsep' is a reentrant alternative to `strtok'.
4167 * New functions `scandir' and `alphasort' for searching directories.
4169 * New function `setenv' is a better interface to `putenv'.
4171 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4172 function for traversing a directory tree.
4174 * The GNU obstack package is now also part of the C library.
4175 The new function `open_obstack_stream' creates a stdio stream that
4176 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4177 formatted output directly to an obstack.
4179 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4180 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4182 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4184 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4185 things to your strings.
4187 * There are some new test programs: test-fseek, testmb, and testrand.
4189 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4190 These ports are not finished, but are a good starting place for really
4191 supporting those systems.
4193 * `/etc/localtime' is now the standard location for the file that says what
4194 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4195 This follows the general principle that `/etc' is the place for all local
4196 configuration files.
4198 * The C library header files now use `extern "C"' when used by the C++
4199 compiler, so the C library should now work with C++ code.
4201 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4202 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4205 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4206 function which uses the `merge sort' algorithm, and is said to be
4207 significantly faster than the old GNU `qsort' function. Merge sort is
4208 now the standard `qsort' function. The new algorithm can require a lot
4209 of temporary storage; so, the old sorting function is called when the
4210 required storage is not available.
4212 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4213 provides the Unix `crypt' function, plus some other entry points.
4215 * The code and header files taken from 4.4 BSD have been updated with the
4216 latest files released from Berkeley.
4218 ----------------------------------------------------------------------
4219 Copyright information:
4221 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4223 Permission is granted to anyone to make or distribute verbatim copies
4224 of this document as received, in any medium, provided that the
4225 copyright notice and this permission notice are preserved,
4226 thus giving the recipient permission to redistribute in turn.
4228 Permission is granted to distribute modified versions
4229 of this document, or of portions of it,
4230 under the above conditions, provided also that they
4231 carry prominent notices stating who last changed them.
4234 version-control: never