1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin and tan
13 with FMA, contributed by Arjan van de Ven and H.J. Lu from Intel.
15 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
17 * Optimized generic expf, exp2f, logf, log2f and powf.
19 * In order to support faster and safer process termination the malloc API
20 family of functions will no longer print a failure address and stack
21 backtrace after detecting heap corruption. The goal is to minimize the
22 amount of work done after corruption is detected and to avoid potential
23 security issues in continued process execution. Reducing shutdown time
24 leads to lower overall process restart latency, so there is benefit both
25 from a security and performance perspective.
27 * The abort function terminates the process immediately, without flushing
28 stdio streams. Previous glibc versions used to flush streams, resulting
29 in deadlocks and further data corruption. This change also affects
30 process aborts as the result of assertion failures.
32 * On platforms where long double has the IEEE binary128 format (aarch64,
33 alpha, mips64, s390 and sparc), the math library now implements _Float128
34 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
35 are the same interfaces added in version 2.26 for some platforms where
36 this format is supported but is not the format of long double.
38 Deprecated and removed features, and other changes affecting compatibility:
40 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
41 defined by <sys/ptrace.h>.
43 * libm no longer supports SVID error handling (calling a user-provided
44 matherr function on error) or the _LIB_VERSION variable to control error
45 handling. (SVID error handling and the _LIB_VERSION variable still work
46 for binaries linked against older versions of the GNU C Library.) The
47 libieee.a library is no longer provided. math.h no longer defines struct
48 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
49 TLOSS, PLOSS and HUGE.
51 * The libm functions pow10, pow10f and pow10l are no longer supported for
52 new programs. Programs should use the standard names exp10, exp10f and
53 exp10l for these functions instead.
55 * The mcontext_t type is no longer the same as struct sigcontext. On
56 platforms where it was previously the same, this changes the C++ name
57 mangling for interfaces involving this type.
59 * The add-ons mechanism for building additional packages at the same time as
60 glibc has been removed. The --enable-add-ons configure option is now
63 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
64 the specified string can be parsed as a domain name.
66 Changes to build and runtime requirements:
68 [Add changes to build and runtime requirements here]
70 Security related changes:
72 CVE-2009-5064: The ldd script would sometimes run the program under
73 examination directly, without preventing code execution through the
74 dynamic linker. (The glibc project disputes that this is a security
75 vulnerability; only trusted binaries must be examined using the ldd
78 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
79 suffered from a one-byte overflow during ~ operator processing (either
80 on the stack or the heap, depending on the length of the user name).
81 Reported by Tim Rühsen.
83 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
84 would sometimes fail to free memory allocated during ~ operator
85 processing, leading to a memory leak and, potentially, to a denial
88 The glob function, when invoked with GLOB_TILDE and without
89 GLOB_NOESCAPE, could write past the end of a buffer while
90 unescaping user names. Reported by Tim Rühsen.
92 The following bugs are resolved with this release:
94 [The release manager will add the list generated by
95 scripts/list-fixed-bugs.py just before the release.]
102 * A per-thread cache has been added to malloc. Access to the cache requires
103 no locks and therefore significantly accelerates the fast path to allocate
104 and free small amounts of memory. Refilling an empty cache requires locking
105 the underlying arena. Performance measurements show significant gains in a
106 wide variety of user workloads. Workloads were captured using a special
107 instrumented malloc and analyzed with a malloc simulator. Contributed by
108 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
110 * Unicode 10.0.0 Support: Character encoding, character type info, and
111 transliteration tables are all updated to Unicode 10.0.0, using
112 generator scripts contributed by Mike FABIAN (Red Hat).
113 These updates cause user visible changes, especially the changes in
114 wcwidth for many emoji characters cause problems when emoji sequences
115 are rendered with pango, see for example:
116 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
118 * Collation of Hungarian has been overhauled and is now consistent with "The
119 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
122 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
124 - The GNU C Library will now detect when /etc/resolv.conf has been
125 modified and reload the changed configuration. The new resolver option
126 “no-reload” (RES_NORELOAD) disables this behavior.
128 - The GNU C Library now supports an arbitrary number of search domains
129 (configured using the “search” directive in /etc/resolv.conf);
130 previously, there was a hard limit of six domains. For backward
131 compatibility, applications that directly modify the ‘_res’ global
132 object are still limited to six search domains.
134 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
135 Library will now randomly pick a name server from the configuration as a
136 starting point. (Previously, the second name server was always used.)
138 * The tunables feature is now enabled by default. This allows users to tweak
139 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
141 * New function reallocarray, which resizes an allocated block (like realloc)
142 to the product of two sizes, with a guaranteed clean failure upon integer
143 overflow in the multiplication. Originally from OpenBSD, contributed by
144 Dennis Wölfing and Rüdiger Sonderfeld.
146 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
147 These are extended versions of preadv and pwritev, respectively, taking an
148 additional flags argument. The set of supported flags depends on the
149 running kernel; full support currently requires kernel 4.7 or later.
151 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
152 create a new session ID for the spawned process. This feature is
153 scheduled to be added to the next major revision of POSIX; for the time
154 being, it is available under _GNU_SOURCE.
156 * errno.h is now safe to use from C-preprocessed assembly language on all
157 supported operating systems. In this context, it will only define the
158 Exxxx constants, as preprocessor macros expanding to integer literals.
160 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
161 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
162 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
163 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
165 To compile programs that use this feature, the compiler must support
166 128-bit floating point with the type name _Float128 (as defined by TS
167 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
168 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
169 must be defined to make the new interfaces visible.
171 The new functions and macros correspond to those present for other
172 floating-point types (except for a few obsolescent interfaces not
173 supported for the new type), with F128 or f128 suffixes; for example,
174 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
175 printf or scanf formats for the new type; the strfromf128 and strtof128
176 interfaces should be used instead.
178 Deprecated and removed features, and other changes affecting compatibility:
180 * The synchronization that pthread_spin_unlock performs has been changed to
181 now be equivalent to a C11 atomic store with release memory order to the
182 spin lock's memory location. Previously, several (but not all)
183 architectures used stronger synchronization (e.g., containing what is
184 often called a full barrier). This change can improve performance, but
185 may affect odd fringe uses of spin locks that depend on the previous
186 behavior (e.g., using spin locks as atomic variables to try to implement
187 Dekker's mutual exclusion algorithm).
189 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
192 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
193 will only be built and installed when the GNU C Library is configured with
194 --enable-obsolete-rpc. This allows alternative RPC implementations, such
195 as TIRPC or rpcsvc-proto, to be used.
197 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
198 libnss_compat, are deprecated, and will not be built or installed by
201 The NIS(+) support library, libnsl, is also deprecated. By default, a
202 compatibility shared library will be built and installed, but not headers
203 or development libraries. Only a few NIS-related programs require this
204 library. (In particular, the GNU C Library has never required programs
205 that use 'gethostbyname' to be linked with libnsl.)
207 Replacement implementations based on TIRPC, which additionally support
208 IPv6, are available from <https://github.com/thkukuk/>. The configure
209 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
210 name service modules, to be built and installed.
212 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
213 support is enabled, the configured recursive resolver must support EDNS.
214 (Responding to EDNS-enabled queries with responses which are not
215 EDNS-enabled is fine, but FORMERR responses are not.)
217 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
218 servers have not supported this opcode for a long time.
220 * The _res_opcodes variable has been removed from libresolv. It had been
221 exported by accident.
223 * <string.h> no longer includes inline versions of any string functions,
224 as this kind of optimization is better done by the compiler. The macros
225 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
227 * The nonstandard header <xlocale.h> has been removed. Most programs should
228 use <locale.h> instead. If you have a specific need for the definition of
229 locale_t with no other declarations, please contact
230 libc-alpha@sourceware.org and explain.
232 * The obsolete header <sys/ultrasound.h> has been removed.
234 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
236 * The obsolete function cfree has been removed. Applications should use
239 * The stack_t type no longer has the name struct sigaltstack. This changes
240 the C++ name mangling for interfaces involving this type.
242 * The ucontext_t type no longer has the name struct ucontext. This changes
243 the C++ name mangling for interfaces involving this type.
245 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
246 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
247 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
248 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
249 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
250 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
251 fpregset_t type no longer has the name struct fpu. This changes the C++
252 name mangling for interfaces involving those types.
254 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
255 synced with the kernel:
257 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
258 are not supported on this architecture and have been removed.
260 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
261 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
262 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
264 Programs that assume the GET/SETREGS ptrace requests are universally
265 available will now fail to build, instead of malfunctioning at runtime.
267 Changes to build and runtime requirements:
269 * Linux kernel 3.2 or later is required at runtime, on all architectures
270 supported by that kernel. (This is a change from version 2.25 only for
273 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
275 * On most architectures, GCC 4.9 or later is required to build the GNU C
276 Library. On powerpc64le, GCC 6.2 or later is required.
278 Older GCC versions and non-GNU compilers are still supported when
279 compiling programs that use the GNU C Library. (We do not know exactly
280 how old, and some GNU extensions to C may be _de facto_ required. If you
281 are interested in helping us make this statement less vague, please
282 contact libc-alpha@sourceware.org.)
284 Security related changes:
286 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
287 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
289 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
290 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
292 * Avoid printing a backtrace from the __stack_chk_fail function since it is
293 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
296 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
297 fixed (CVE-2017-12133).
299 The following bugs are resolved with this release:
301 [984] network: Respond to changed resolv.conf in gethostbyname
302 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
303 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
305 [12189] libc: __stack_chk_fail should not attempt a backtrace
307 [14096] time: Race condition on timezone/tst-timezone.out
308 [14172] localedata: az_IR: new locale
309 [14995] build: glibc fails to build if gold is the default linker, even if
311 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
312 [16637] network: inet_pton function is accepting IPv6 with bad format
313 [16640] string: string/strtok.c: undefined behaviour inconsistent between
314 x86 and other generic code
315 [16875] localedata: ko_KR: fix lang_name
316 [17225] localedata: ar_SY: localized month names for May and June are
318 [17297] localedata: da_DK: wrong date_fmt string
319 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
320 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
321 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
322 [19066] localedata: ar_SA abbreviated day and month names are in English
323 [19569] network: resolv: Support an arbitrary number of search domains
324 [19570] network: Implement random DNS server selection in the stub
326 [19838] locale: localedef fails on PA-RISC
327 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
328 order of 0D36 and 0D37
329 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
331 [20098] libc: FAIL: debug/backtrace-tst on hppa
332 [20257] network: sunrpc: clntudp_call does not enforce timeout when
334 [20275] localedata: locale day/abday/mon/abmon should not have trailing
336 [20313] localedata: Update locale data to Unicode 9.0
337 [20424] manual: Document how to provide a malloc replacement
338 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
340 [20686] locale: Add el_GR@euro to SUPPORTED.
341 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
342 failures consistently
343 [21015] dynamic-link: Document and fix --enable-bind-now
344 [21016] nptl: pthread_cond support is broken on hppa
345 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
346 frame-pointer on i386
347 [21049] libc: segfault in longjmp_chk() due to clobbered processor
349 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
350 [21088] libc: Build fails with --enable-static-nss
351 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
352 generic c code is used
353 [21109] libc: Tunables broken on big-endian
354 [21112] math: powf has large ulp errors with base close to 1 and exponent
356 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
358 [21120] malloc: glibc malloc is incompatible with GCC 7
359 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
361 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
362 (-0.0) when linking with -lieee
363 [21171] math: log10, log2 and lgamma return incorrect results
364 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
365 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
366 [21207] localedata: ce_RU: update weekdays from CLDR
367 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
368 [21217] localedata: Update months from CLDR-31
369 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
370 [21243] libc: support_delete_temp_file should issue warning for failed
372 [21244] libc: support resolv_test_start() socket fd close should be
374 [21253] libc: localedef randomly segfaults when using -fstack-check due to
375 new posix_spawn implementation
376 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
377 leads to lower CPU frequency
378 [21259] libc: [alpha] termios.h missing IXANY for POSIX
379 [21261] libc: [sparc64] bits/setjmp.h namespace
380 [21267] network: [mips] bits/socket.h IOC* namespace
381 [21268] libc: [alpha] termios.h NL2, NL3 namespace
382 [21270] libc: mmap64 silently truncates large offset values
383 [21275] libc: posix_spawn always crashes on ia64 now
384 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
385 [21280] math: [powerpc] logbl for POWER7 return incorrect results
386 [21289] libc: Incorrect declaration for 32-bit platforms with
387 _FILE_OFFSET_BITS=64 causes build error
388 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
389 not support gethostbyname4_r
390 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
392 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
394 [21340] libc: Support POSIX_SPAWN_SETSID
395 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
396 [21359] network: ns_name_pack needs additional byte in destination buffer
397 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
398 against fragmentation attacks (CVE-2017-12132)
399 [21369] network: resolv: Remove EDNS fallback
400 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
402 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
403 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
404 [21393] stdio: Missing dup3 error check in freopen, freopen64
405 [21396] libc: Use AVX2 memcpy/memset on Skylake server
406 [21399] localedata: Bad description for U00EC in
407 localedata/charmaps/CP1254
408 [21411] malloc: realloc documentation error
409 [21426] network: sys/socket.h uio.h namespace
410 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
411 [21445] libc: signal.h bsd_signal namespace
412 [21455] network: Network headers stdint.h namespace
413 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
415 [21475] network: resolv: Overlong search path is truncated mid-label
416 [21511] libc: sigstack namespace
417 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
418 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
419 bits/syscall.h: No such file or directory
420 [21517] libc: struct sigaltstack namespace
421 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
422 [21533] localedata: Update locale data to Unicode 10.0
424 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
425 junk at end of line, first unrecognized character is `@'
426 [21538] libc: SIG_HOLD missing for XPG4
427 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
428 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
429 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
430 [21543] libc: sigevent namespace
431 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
432 size for general purpose registers in mcontext_t structure
433 [21550] libc: sigwait namespace
434 [21552] libc: XPG4 bsd_signal namespace
435 [21554] libc: sigpause namespace
436 [21560] libc: sys/wait.h signal.h namespace
437 [21561] libc: waitid namespace
438 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
439 '/usr/include/stdlib.h': Permission denied
440 [21575] libc: sys/wait.h missing struct rusage definition
441 [21584] libc: sigaltstack etc namespace
442 [21597] libc: siginterrupt namespace
443 [21607] math: hppa: FAIL: math/test-tgmath
444 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
446 [21622] libc: [tile] missing SA_* for POSIX.1:2008
447 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
448 stack and heap (CVE-2017-1000366)
449 [21625] libc: wait3 namespace
450 [21654] nss: Incorrect pointer alignment in NSS group merge result
452 [21657] network: Parse interface zone id for node-local multicast
453 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
455 [21666] libc: .symver is used on common symbol
456 [21668] network: resolv: res_init cross-thread broadcast introduces race
458 [21687] math: tgmath.h totalorder, totalordermag return type
459 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
461 [21696] libc: Incorrect assumption of of __cpu_mask in
462 posix/sched_cpucount.c
463 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
464 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
466 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
467 leading to relocation crash
468 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
469 `resolv_conf_matches (resp, conf)' failed.
470 [21710] localedata: Added Samoan language locale for Samoa
471 [21711] localedata: Pashto yesstr/nostr locale are missing
472 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
473 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
474 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
475 [21723] localedata: yesstr/nostr missing for Chinese language locale
476 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
478 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
480 [21728] localedata: New Locale for Tongan language
481 [21729] localedata: incorrect LC_NAME fields for hi_IN
482 [21733] localedata: yesstr and nostr are missing for zh_HK
483 [21734] localedata: Missing yesstr and nostr are for kw_GB
484 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
485 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
486 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
487 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
489 [21744] libc: Tests failing on --enable-tunables --enable-stack-
491 [21749] localedata: Wrong abbreviated day name (“abday”) for
493 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
494 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
495 [21759] localedata: missing yesstr and nostr for Tigrinya
496 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
497 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
498 [21767] localedata: Missing Bislama locales
499 [21768] localedata: Missing yesstr and nostr for aa_ET
500 [21770] localedata: Missing Field in li_NL
501 [21778] nptl: Robust mutex may deadlock
502 [21779] libc: MicroBlaze segfaults when loading libpthread
503 [21783] localedata: Fix int_select international_call_prefixes
504 [21784] localedata: Inconsistency in country_isbn
505 [21788] localedata: Missing Country Postal Abbreviations
506 [21794] localedata: Added-country_isbn-for-Italy
507 [21795] localedata: Add/Fix country_isbn for France
508 [21796] localedata: Added country_isbn for Republic of Korea
509 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
510 [21799] localedata: Added int_select international_call_prefixes
511 [21801] localedata: Added int_select international_call_prefixes
512 [21804] nptl: Double semicolon in thread-shared-types.h
513 [21807] localedata: LC_ADDRESS fix for pap_CW
514 [21808] localedata: Fix LC_ADDRESS for pap_AW
515 [21821] localedata: Added country_name in mai_IN
516 [21822] localedata: Fix LC_TIME for mai_IN
517 [21823] localedata: missing yesstr, nostr for sa_IN
518 [21825] localedata: Fix name_mrs for mag_IN
519 [21828] localedata: 2.26 changelog should mention user visible changes
521 [21835] localedata: Added Maithili language locale for Nepal
522 [21838] localedata: Removed redundant data for the_NP
523 [21839] localedata: Fix LC_MONETARY for ta_LK
524 [21844] localedata: Fix Latin characters and Months Sequence.
525 [21848] localedata: Fix mai_NP Title Name
530 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
531 24731-2:2010, is supported to enable declarations of functions from that
532 TR. Note that not all functions from that TR are supported by the GNU C
535 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
536 18661-1:2014, is supported to enable declarations of functions and macros
537 from that TS. Note that not all features from that TS are supported by
540 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
541 18661-4:2015, is supported to enable declarations of functions and macros
542 from that TS. Note that most features from that TS are not supported by
545 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
546 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
547 Since the GNU C Library defaults to a much newer revision of POSIX, this
548 will only affect programs that specifically request an old conformance
549 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
550 see a change in the visible declarations, but a program compiled with
551 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
554 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
555 defined by all multithreaded code, but glibc has not required this for
558 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
559 means that in a future release, the macros “major”, “minor”, and “makedev”
560 will only be available from <sys/sysmacros.h>.
562 These macros are not part of POSIX nor XSI, and their names frequently
563 collide with user code; see for instance glibc bug 19239 and Red Hat bug
564 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
565 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
568 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
569 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
570 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
572 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
573 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
574 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
575 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
576 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
577 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
578 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
579 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
580 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
581 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
582 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
585 * New <math.h> features are added from TS 18661-1:2014:
587 - Signaling NaN macros: SNANF, SNAN, SNANL.
589 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
590 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
591 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
593 - llogb functions: the llogb, llogbf and llogbl functions, and the
594 FP_LLOGB0 and FP_LLOGBNAN macros.
596 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
599 - Comparison macros: iseqsig.
601 - Classification macros: iscanonical, issubnormal, iszero.
603 - Total order functions: totalorder, totalorderf, totalorderl,
604 totalordermag, totalordermagf, totalordermagl.
606 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
608 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
609 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
611 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
612 are added to libc. They convert a floating-point number into string.
614 * Most of glibc can now be built with the stack smashing protector enabled.
615 It is recommended to build glibc with --enable-stack-protector=strong.
616 Implemented by Nick Alcock (Oracle).
618 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
619 intended to be used instead of memset() to erase sensitive data after use;
620 the compiler will not optimize out calls to explicit_bzero even if they
621 are "unnecessary" (in the sense that no _correct_ program can observe the
622 effects of the memory clear).
624 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
625 to float instead of double. This does not affect the ABI of any libraries
626 that are part of the GNU C Library, but may affect the ABI of other
627 libraries that use this type in their interfaces.
629 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
630 float_t and double_t types are now defined to long double instead of float
631 and double. These options are not the default, and this does not affect
632 the ABI of any libraries that are part of the GNU C Library, but it may
633 affect the ABI of other libraries that use this type in their interfaces,
634 if they are compiled or used with those options.
636 * The getentropy and getrandom functions, and the <sys/random.h> header file
639 * The buffer size for byte-oriented stdio streams is now limited to 8192
640 bytes by default. Previously, on Linux, the default buffer size on most
641 file systems was 4096 bytes (and thus remains unchanged), except on
642 network file systems, where the buffer size was unpredictable and could be
643 as large as several megabytes.
645 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
646 for the Linux quota interface which predates kernel version 2.4.22 has
649 * The malloc_get_state and malloc_set_state functions have been removed.
650 Already-existing binaries that dynamically link to these functions will
651 get a hidden implementation in which malloc_get_state is a stub. As far
652 as we know, these functions are used only by GNU Emacs and this change
653 will not adversely affect already-built Emacs executables. Any undumped
654 Emacs executables, which normally exist only during an Emacs build, should
655 be rebuilt by re-running “./configure; make” in the Emacs build tree.
657 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
658 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
659 “no-ip6-dotint” had already been the default, and support for the
660 “ip6-dotint” option was removed from the Internet in 2006.
662 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
663 flag from <resolv.h> have been removed. The option relied on a
664 backwards-incompatible DNS extension which was never deployed on the
667 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
668 RES_BLAST defined in the <resolv.h> header file have been deprecated.
669 They were already unimplemented.
671 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
672 _res.flags are deprecated. The flag was standardized in RFC 2133, but
673 removed again from the IETF name lookup interface specification in RFC
674 2553. Applications should use getaddrinfo instead.
676 * DNSSEC-related declarations and definitions have been removed from the
677 <arpa/nameser.h> header file, and libresolv will no longer attempt to
678 decode the data part of DNSSEC record types. Previous versions of glibc
679 only implemented minimal support for the previous version of DNSSEC, which
680 is incompatible with the currently deployed version.
682 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
683 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
684 <arpa/nameser.h> header file because the distinction between RR types and
685 meta-RR types is not officially standardized, subject to revision, and
686 thus not suitable for encoding in a macro.
688 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
689 and rhook members of the res_state type in <resolv.h> have been removed.
690 The glibc stub resolver did not support these hooks, but the header file
691 did not reflect that.
693 * For multi-arch support it is recommended to use a GCC which has
694 been built with support for GNU indirect functions. This ensures
695 that correct debugging information is generated for functions
696 selected by IFUNC resolvers. This support can either be enabled by
697 configuring GCC with '--enable-gnu-indirect-function', or by
698 enabling it by default by setting 'default_gnu_indirect_function'
699 variable for a particular architecture in the GCC source file
702 * GDB pretty printers have been added for mutex and condition variable
703 structures in POSIX Threads. When installed and loaded in gdb these pretty
704 printers show various pthread variables in human-readable form when read
705 using the 'print' or 'display' commands in gdb.
707 * Tunables feature added to allow tweaking of the runtime for an application
708 program. This feature can be enabled with the '--enable-tunables' configure
709 flag. The GNU C Library manual has details on usage and README.tunables has
710 instructions on adding new tunables to the library.
712 * A new version of condition variables functions have been implemented in
713 the NPTL implementation of POSIX Threads to provide stronger ordering
716 * A new version of pthread_rwlock functions have been implemented to use a more
717 scalable algorithm primarily through not using a critical section anymore to
720 Security related changes:
722 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
723 have been created with makecontext could fail to terminate due to a
724 missing .cantunwind annotation. This has been observed to lead to a hang
725 (denial of service) in some Go applications compiled with gccgo. Reported
726 by Andreas Schwab. (CVE-2016-6323)
728 * The DNS stub resolver functions would crash due to a NULL pointer
729 dereference when processing a query with a valid DNS question type which
730 was used internally in the implementation. The stub resolver now uses a
731 question type which is outside the range of valid question type values.
734 The following bugs are resolved with this release:
736 [4099] stdio: Overly agressive caching by stream i/o functions.
737 [7065] build: Support building glibc with -fstack-protector or -fstack-
739 [9842] localedata: en_CA: incorrect date format
740 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
741 before it started waiting
742 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
743 [15765] nptl: sem_open is wrongly a cancellation point
744 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
745 [16458] libc: endian.h and netinet/in.h byte order macros return values of
747 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
748 library linked with pthread
749 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
751 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
752 [17252] libc: getrandom and getentropy syscall
753 [17730] malloc: thread-local storage is sometimes improperly free()'d
754 after being __libc_memalign()'d
755 [18241] stdio: failed fseek on memstream does not set errno and can fail
757 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
759 [18463] nptl: pthread_cond_broadcast issue when surrounded by
760 PTHREAD_PRIO_INHERIT mutex on ARM
761 [18784] network: res_query and related function crash for special record
762 type queries (CVE-2015-5180)
763 [19380] math: strtod does not raise "inexact"
764 [19387] string: Integer overflow in memchr
765 [19390] string: Integer overflow in strncat
766 [19398] build: linknamespace tests fail with massively parallel build
767 [19402] nptl: Deadlock with robust shared mutex and asynchronous
769 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
770 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
771 [19514] libc: [PATCH] Fix spelling errors in spelling
772 "implement"/"implementation" in several places
773 [19582] network: Deprecate RES_USE_INET6
774 [19673] manual: clog10 docs appear to be erroneous
775 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
776 causes a segmentation fault
777 [19826] libc: invalid pointer returned from __tls_get_addr with static
779 [20016] network: resolv: Remove hooks support from the API
780 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
782 [20033] math: [x86_64] vectorized math function don't call the __finite
784 [20116] nptl: use after free in pthread_create
785 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
786 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
788 [20311] nptl: please install proc_service.h
789 [20366] libc: Compilation errors in installed headers in strict-compliance
791 [20370] malloc: malloc: Arena free list management is still racy
792 (incorrect fix in bug 19243)
793 [20386] libc: assert (X = 0) does not result in GCC warning
794 [20432] malloc: malloc: Minimize interface required for interposition
795 [20435] libc: Missing unwind info in __startcontext causes infinite loop
796 in _Unwind_Backtrace (CVE-2016-6323)
797 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
798 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
800 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
801 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
802 [20477] network: resolv: incorrect double-checked locking related to
804 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
806 [20495] math: x86_64 performance degradation due to AVX/SSE transition
808 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
809 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
810 cause transition penalty
811 [20517] math: sparcv9 missing fdiml compat symbol
812 [20524] manual: strverscmp is inconsistent
813 [20525] libc: <sys/quota.h> should be based on kernel headers
814 [20539] math: GCC 7 -static -lm fails to link at -O3
815 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
817 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
819 [20591] network: Remove obsolete DNSSEC support
820 [20592] network: DNS resource record type classification macros in
821 <arpa/nameser.h> are incorrect
822 [20593] network: Update DNS RR type definitions
823 [20611] network: getaddrinfo accepts invalid numeric scope IDs
824 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
826 [20629] network: libresolv: Remove support for bitstring labels
828 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
829 assertion: "maxidx >= 2"
830 [20660] math: [arm] Use VSQRT
831 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
832 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
833 [20689] libc: Test for FMA should also check for AVX.
834 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
835 [20715] math: iszero macro breaks existing code
836 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
837 [20728] libc: powerpc: Missing TOC stub in clone
838 [20729] build: glibc-2.24 fails to build for i486 with -Os
839 [20750] build: Build fails with default PIE enabled gcc-6.x
840 [20768] math: [alpha] sqrt fegetenv namespace
841 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
843 [20787] math: float_t is defined as float incorrectly on x86_64 even if
844 __FLT_EVAL_METHOD__ is 2
845 [20790] network: rpcgen buffer overrun in get_prog_declaration
846 [20822] nptl: race condition in __lll_unlock_elision on powerpc
847 [20829] libc: crypt snprintf namespace
848 [20847] libc: tst-vfork3 failure
849 [20855] math: Default bits/mathdef.h has inappropriate float_t
850 [20859] math: [sh4] FP_ILOGB0 invalid
851 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
852 U+20AC), not same as GBK
853 [20915] dynamic-link: global-dynamic TLS broken on aarch64
854 [20916] math: pow handling of sNaN arguments
855 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
856 [20919] math: Bad pow (qNaN, 0) result with -lieee
857 [20940] math: hypot sNaN handling
858 [20947] math: fmax, fmin sNaN handling
859 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
860 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
862 [20971] string: powerpc64/power7 memchr overflows internal pointer check
863 [20973] nptl: robust mutexes: Lost wake-ups
864 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
865 [20978] nis: strlen on null pointer in nss_nisplus
866 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
868 [21014] string: i686 memchr overflows internal pointer check
869 [21019] libc: [mips] n32 lseek incorrect on overflow
870 [21022] libc: [microblaze] __backtrace get_frame_size namespace
871 [21026] libc: [MIPS] readahead syscall is broken on n64
872 [21028] math: Fallback fesetexceptflag should always succeed
873 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
874 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
876 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
877 [21061] librt: [microblaze] librt lost clock_* exports
878 [21073] libc: tunables: insecure environment variables passed to
879 subprocesses with AT_SECURE
880 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
884 * The minimum Linux kernel version that this version of the GNU C Library
885 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
886 kernel version 2.6.32 or later suffices (on architectures that already
887 required kernel versions more recent than 3.2, those requirements remain
888 unchanged). Linux 3.2 or later kernel headers are required on all
891 * The pap_AN locale has been deleted. This has been deprecated for a long
892 time. It has been replaced by pap_AW & pap_CW, both of which have long
893 been included in previous releases.
895 * The readdir_r and readdir64_r functions have been deprecated. It is
896 recommended to use readdir and readdir64 instead.
898 * The type “union wait” has been removed. It was deprecated in the early
899 1990s and never part of POSIX. Application code should use the int type
900 instead of “union wait”.
902 * A new NSS action is added to facilitate large distributed system
903 administration. The action, MERGE, allows remote user stores like LDAP
904 to be merged into local user stores like /etc/groups in order to provide
905 easy to use, updated, and managed sets of merged credentials. The new
906 action can be used by configuring it in /etc/nsswitch.conf:
907 group: files [SUCCESS=merge] nis
908 Implemented by Stephen Gallagher (Red Hat).
910 * The deprecated __malloc_initialize_hook variable has been removed from the
913 * The long unused localedef --old-style option has been removed. It hasn't
914 done anything in over 16 years. Scripts using this option can safely
917 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
918 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
919 return the next representable value in the direction of positive infinity
920 and the nextdown functions return the next representable value in the
921 direction of negative infinity. These are currently enabled as GNU
924 Security related changes:
926 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
927 could result in a stack overflow when getnetbyname was called with an
928 overly long name. (CVE-2016-3075)
930 * Previously, getaddrinfo copied large amounts of address data to the stack,
931 even after the fix for CVE-2013-4458 has been applied, potentially
932 resulting in a stack overflow. getaddrinfo now uses a heap allocation
933 instead. Reported by Michael Petlan. (CVE-2016-3706)
935 * The glob function suffered from a stack-based buffer overflow when it was
936 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
937 Reported by Alexander Cherepanov. (CVE-2016-1234)
939 * The Sun RPC UDP client could exhaust all available stack space when
940 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
941 alloca plugin for GCC. (CVE-2016-4429)
943 * The IPv6 name server management code in libresolv could result in a memory
944 leak for each thread which is created, performs a failing naming lookup,
945 and exits. Over time, this could result in a denial of service due to
946 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
948 The following bugs are resolved with this release:
950 [1170] localedata: ne_NP: update Nepali locale definition file
951 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
953 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
954 [6796] math: fdim() does not set errno on overflow
955 [10354] libc: posix_spawn should use vfork() in more cases than presently
956 [11213] localedata: localedata: add copyright disclaimer to locale files
957 [12143] localedata: chr_US: new Cherokee locale
958 [12450] localedata: sgs_LT: new locale
959 [12676] localedata: ln_CD: new locale
960 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
962 [13304] math: fma, fmaf, fmal produce wrong results
963 [14259] build: --localedir arg to configure is ignored
964 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
965 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
966 [14934] localedata: es_CL: wrong first weekday chilean locale
967 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
969 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
971 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
972 [15368] nptl: raise() is not async-signal-safe
973 [15479] math: ceil, floor, round and trunc raise inexact exception
974 [15578] localedata: kk_KZ: various updates
975 [16003] localedata: pap_AN: punt old locale
976 [16137] localedata: iw_IL: punt old locale
977 [16190] localedata: eo: new esperanto locale
978 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
979 [16742] malloc: race condition: pthread_atfork() called before first
980 malloc() results in unexpected locking behaviour/deadlocks
981 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
983 [16983] localedata: postal_fmt does not allow %l and %n modifiers
984 [17565] localedata: pt_PT: wrong (work-)week start
985 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
986 [17950] build: Build fails with -msse
987 [18205] localedata: be_BY*: wrong first_weekday and first_workday
988 [18433] libc: posix_spawn does not return correctly upon failure to
990 [18453] localedata: charmaps/IBM875: incorrect codes
991 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
993 [18896] localedata: he_IL: improvements for currency
994 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
996 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
998 [19056] libc: Deprecate readdir_r
999 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1001 [19198] localedata: nl_NL: small improvements for Dutch locales
1002 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1003 nameservers (CVE-2016-5417)
1004 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1006 [19400] locale: Language missing in "iso-639.def", trivial fix in
1008 [19431] malloc: Deadlock between fflush, getdelim, and fork
1009 [19505] libc: Incorrect file descriptor validity checks in
1010 posix_spawn_file_actions_add{open,close,dup2}
1011 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1012 when using RTLD_NEXT
1013 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1014 `intl/{gettextP,loadinfo}.h'
1015 [19534] libc: execle, execlp may use malloc
1016 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1017 [19573] network: res_nclose and __res_maybe_init disagree about name
1018 server initialization, breaking Hesiod
1019 [19575] localedata: Status of GB18030 tables
1020 [19581] localedata: sr_* date_fmt string contains additional newline
1021 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1023 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1024 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1025 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1026 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1028 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1030 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1032 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1033 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1034 [19633] locale: strfmon_l applies global locale to number formatting
1035 [19642] network: Memory leak in getnameinfo
1036 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1037 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1039 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1040 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1042 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1043 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1045 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1047 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1049 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1050 pointers and lengths in error-case.
1051 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1052 report errors on UTF-16 surrogates values.
1053 [19755] nscd: nscd assertion failure in gc
1054 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1055 [19759] libc: mempcpy shouldn't be inlined
1056 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1057 [19765] libc: s390 needs an optimized mempcpy
1058 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1059 NAME_MAX limit assumption (CVE-2016-1234)
1060 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1061 [19787] network: Missing and incorrect truncation checks in getnameinfo
1062 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1064 [19791] network: Assertion failure in res_query.c with un-connectable name
1066 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1067 [19822] math: libm.so install clobbers old version
1068 [19825] network: resolv: send_vc can return uninitialized data in second
1069 response to getaddrinfo
1070 [19830] network: nss_dns: should check RDATA length against buffer length
1071 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1072 confronted with A/AAAA records of invalid size
1073 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1074 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1075 [19853] stdio: Printing IBM long double in decimal with high precision is
1077 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1078 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1079 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1081 [19865] network: Assertion failure or memory leak in
1082 _nss_dns_getcanonname_r
1083 [19868] network: nss_dns: netent code does not skip over non-PTR records
1084 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1086 [19881] string: Improve x86-64 memset
1087 [19907] string: Incorrect memcpy tests
1088 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1090 [19925] libc: termios.h XCASE namespace
1091 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1092 [19929] libc: limits.h NL_NMAX namespace
1093 [19931] stdio: Memory leak in vfprintf
1094 [19957] libc: clone(CLONE_VM) access invalid parent memory
1095 [19963] localedata: en_IL: New locale
1096 [19989] stdio: stdio.h cuserid namespace
1097 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1099 [19996] locale: langinfo.h nl_langinfo_l namespace
1100 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1101 [20010] network: getaddrinfo: Stack overflow in hostent translation
1103 [20012] stdio: libio: fmemopen append mode failure
1104 [20014] stdio: stdio.h namespace for pre-threads POSIX
1105 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1106 [20023] libc: fcntl.h timespec namespace
1107 [20024] math: [x86_64] vectorized sincos trashes the stack
1108 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1109 [20041] time: sys/time.h timespec namespace
1110 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1111 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1112 [20051] libc: ttyslot in wrong header under wrong conditions
1113 [20054] libc: gethostname not declared for XPG4
1114 [20055] libc: termios.h missing tcgetsid for XPG4
1115 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1117 [20073] libc: sys/stat.h fchmod namespace
1118 [20074] libc: stdlib.h rand_r namespace
1119 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1120 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1122 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1123 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1125 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1126 [20119] libc: Wrong mask for processors level type from CPUID
1127 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1128 AS not supporting AVX512
1129 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1131 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1132 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1134 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1135 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1136 [20178] libc: posix_spawn{p} should not call exit
1137 [20191] stdio: libio: vtables hardening
1138 [20195] string: FMA4 detection requires CPUID execution with register
1140 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1141 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1143 [20212] math: acos (sNaN) returns sNaN
1144 [20213] math: asin (sNaN) returns sNaN
1145 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1146 [20218] math: [i386] asinhl (sNaN) returns sNaN
1147 [20219] math: [i386] atanhl (sNaN) returns sNaN
1148 [20222] stdio: fopencookie: Mangle function pointers
1149 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1150 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1151 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1153 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1154 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1155 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1156 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1157 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1158 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1159 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1160 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1161 [20240] math: modf (sNaN) returns sNaN
1162 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1163 [20250] math: frexp (sNaN) returns sNaN
1164 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1165 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1167 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1168 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1169 subscript is below array bounds [-Werror=array-bounds]
1170 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1171 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1173 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1174 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1176 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1177 "invalid" exceptions
1178 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1179 [20316] localedata: id_ID: Februari instead of Pebruari
1180 [20327] string: POWER8 strcasecmp returns incorrect result
1181 [20347] math: Failure: Test: j0_downward (0xap+0)
1182 [20348] libc: FAIL: misc/tst-preadvwritev64
1183 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1184 [20350] libc: There is no test for p{read,write}64
1185 [20357] math: Incorrect cos result for 1.5174239687223976
1186 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1190 * Unicode 8.0.0 Support: Character encoding, character type info, and
1191 transliteration tables are all updated to Unicode 8.0.0, using new
1192 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1193 These updates cause user visible changes, such as the fixes for bugs
1194 89, 16061, and 18568.
1196 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1197 kernel-internal CPU set size. This means that requests that change the
1198 CPU affinity which failed before (for example, an all-ones CPU mask) will
1199 now succeed. Applications that need to determine the effective CPU
1200 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1201 setting it because the kernel can adjust it (and the previous size check
1202 would not detect this in the majority of cases).
1204 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1205 the following new symbols are used: fts64_children, fts64_close,
1206 fts64_open, fts64_read and fts64_set.
1208 * getaddrinfo now detects certain invalid responses on an internal netlink
1209 socket. If such responses are received, an affected process will
1210 terminate with an error message of "Unexpected error <number> on netlink
1211 descriptor <number>" or "Unexpected netlink response of size <number> on
1212 descriptor <number>". The most likely cause for these errors is a
1213 multi-threaded application which erroneously closes and reuses the netlink
1214 file descriptor while it is used by getaddrinfo.
1216 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1217 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1218 unnecessary serialization of memory allocation requests across threads.
1219 The defect is now corrected. Users should see a substantial increase in
1220 the concurent throughput of allocation requests for applications which
1221 trigger this bug. Affected applications typically create and
1222 destroy threads frequently. (Bug 19048 was reported and analyzed by
1225 * There is now a --disable-timezone-tools configure option for disabling the
1226 building and installing of the timezone related utilities (zic, zdump, and
1227 tzselect). This is useful for people who build the timezone data and code
1228 independent of the GNU C Library.
1230 * The obsolete header <regexp.h> has been removed. Programs that require
1231 this header must be updated to use <regex.h> instead.
1233 * The obsolete functions bdflush, create_module, get_kernel_syms,
1234 query_module and uselib are no longer available to newly linked binaries;
1235 the header <sys/kdaemon.h> has been removed. These functions and header
1236 were specific to systems using the Linux kernel and could not usefully be
1237 used with the GNU C Library on systems with version 2.6 or later of the
1240 * Optimized string, wcsmbs and memory functions for IBM z13.
1241 Implemented by Stefan Liebler.
1243 * Newly linked programs that define a variable called signgam will no longer
1244 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1245 require signgam to be set by those functions must ensure that they use the
1246 variable provided by the GNU C Library and declared in <math.h>, without
1247 defining their own copy.
1249 * The minimum GCC version that can be used to build this version of the GNU
1250 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1251 still be used to compile programs using the GNU C Library.
1253 Security related changes:
1255 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1256 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1258 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1259 the pointer guard feature. It is always enabled. Previously,
1260 LD_POINTER_GUARD could be used to disable security hardening in binaries
1261 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1264 * An integer overflow in hcreate and hcreate_r could lead to an
1265 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1267 * The catopen function no longer has unbounded stack usage. Reported by
1268 Max. (CVE-2015-8779)
1270 * The nan, nanf and nanl functions no longer have unbounded stack usage
1271 depending on the length of the string passed as an argument to the
1272 functions. Reported by Joseph Myers. (CVE-2014-9761)
1274 * A stack-based buffer overflow was found in libresolv when invoked from
1275 libnss_dns, allowing specially crafted DNS responses to seize control
1276 of execution flow in the DNS client. The buffer overflow occurs in
1277 the functions send_dg (send datagram) and send_vc (send TCP) for the
1278 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1279 family. The use of AF_UNSPEC triggers the low-level resolver code to
1280 send out two parallel queries for A and AAAA. A mismanagement of the
1281 buffers used for those queries could result in the response of a query
1282 writing beyond the alloca allocated buffer created by
1283 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1284 the overflow. Thanks to the Google Security Team and Red Hat for
1285 reporting the security impact of this issue, and Robert Holiday of
1286 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1288 The following bugs are resolved with this release:
1290 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1291 [887] math: Math library function "logb" and "nextafter" inconsistent
1292 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1293 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1294 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1295 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1297 [4404] localedata: German translation of "Alarm clock" is misleading
1298 [6799] math: nextafter() and nexttoward() doen't set errno on
1299 overflow/underflow errors
1300 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1302 [10432] nis: _nss_nis_setnetgrent assertion failure
1303 [11460] libc: fts has no LFS support
1304 [12926] network: getaddrinfo()/make_request() may spin forever
1305 [13065] nptl: Race condition in pthread barriers
1306 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1307 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1309 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1310 [14912] libc: Rename non-installed bits/*.h headers
1311 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1312 [15367] math: Let gcc use __builtin_isinf
1313 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1314 [15421] math: lgamma wrongly sets signgam for ISO C
1315 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1316 FE_INVALID with argument out of range
1317 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1319 [15786] dynamic-link: ifunc resolver functions can smash function
1321 [15918] math: Unnecessary check for equality in hypotf()
1322 [16061] localedata: Review / update transliteration data
1323 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1325 [16141] time: strptime %z offset restriction
1326 [16171] math: drem should be alias of remainder
1327 [16296] math: fegetround is pure?
1328 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1329 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1331 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1332 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1334 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1335 "Exception "Inexact" set" on ppc32
1336 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1337 [16517] math: Missing underflow exception from tanf/tan/tanl
1338 [16519] math: Missing underflow exception from sinhf
1339 [16520] math: Missing underflow exception from tanhf
1340 [16521] math: Missing underflow exception from exp2
1341 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1343 [16734] stdio: fopen calls mmap to allocate its buffer
1344 [16961] math: nan function incorrect handling of bad sequences
1345 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1346 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1347 [16985] locale: localedef: confusing error message when opening output
1349 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1350 [17197] locale: Redundant shift character in iconv conversion output at
1352 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1353 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1354 [17250] dynamic-link: static linking breaks nss loading
1355 (getaddrinfo/getpwnam/etc...)
1356 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1358 [17441] math: isnan() should use __builtin_isnan() in GCC
1359 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1360 (related to lock elision)
1361 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1362 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1363 [17887] time: strptime should be able to parse "+01:00" style timezones
1364 [17905] libc: catopen() Multiple unbounded stack allocations
1366 [18084] libc: backtrace (..., 0) dumps core on x86
1367 [18086] libc: nice() sets errno to 0 on success
1368 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1369 is too large (CVE-2015-8778)
1370 [18251] dynamic-link: SONAME missing when audit modules provides path
1371 [18265] libc: add attributes for wchar string and memory functions
1372 [18370] math: csqrt missing underflows
1373 [18421] libc: [hppa] read-only segment has dynamic relocations
1374 [18472] libc: Obsolete syscall wrappers should be compat symbols
1375 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1376 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1377 [18525] localedata: Remove locale timezone information
1378 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1379 [18568] localedata: Update locale data to Unicode 8.0
1380 [18589] locale: sort-test.sh fails at random
1381 [18595] math: ctan, ctanh missing underflows
1382 [18604] libc: assert macro-expands its argument
1383 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1384 contains a vector instruction exception.
1385 [18611] math: j1, jn missing errno setting on underflow
1386 [18618] localedata: sync Chechen locale definitions with other *_RU
1388 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1389 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1390 [18665] network: In send_dg, the recvfrom function is NOT always using the
1391 buffer size of a newly created buffer (CVE-2015-7547)
1392 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1394 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1396 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1397 [18699] math: tilegx cproj() for various complex infinities does not yield
1399 [18724] libc: Harden put*ent functions against data injection
1400 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1401 [18755] build: build errors with -DNDEBUG
1402 [18757] stdio: fmemopen fails to set errno on failure
1403 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1405 [18781] libc: openat64 lacks O_LARGEFILE
1406 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1407 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1408 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1409 [18790] math: [ldbl-128ibm] tanhl inaccurate
1410 [18795] libc: stpncpy fortification misses buffer lengths that are
1411 statically too large
1412 [18796] build: build fails for --disable-mathvec
1413 [18803] math: hypot missing underflows
1414 [18820] stdio: fmemopen may leak memory on failure
1415 [18823] math: csqrt spurious underflows
1416 [18824] math: fma spurious underflows
1417 [18825] math: pow missing underflows
1418 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1419 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1420 [18870] build: sem_open.c fails to compile with missing symbol
1422 [18872] stdio: Fix memory leak in printf_positional
1423 [18873] libc: posix_fallocate overflow check ineffective
1424 [18875] math: Excess precision leads incorrect libm
1425 [18877] libc: arm: mmap offset regression
1426 [18887] libc: memory corruption when using getmntent on blank lines
1427 [18918] localedata: hu_HU: change time to HH:MM:SS format
1428 [18921] libc: Regression: extraneous stat() and fstat() performed by
1430 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1431 binaries (CVE-2015-8777)
1432 [18951] math: tgamma missing underflows
1433 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1435 [18953] localedata: lt_LT: change currency symbol to the euro
1436 [18956] math: powf inaccuracy
1437 [18961] math: [i386] exp missing underflows
1438 [18966] math: [i386] exp10 missing underflows
1439 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1440 [18969] build: multiple string test failures due to missing locale
1442 [18970] libc: Reference of pthread_setcancelstate in libc.a
1443 [18977] math: float / long double Bessel functions not in XSI POSIX
1444 [18980] math: i386 libm functions return with excess range and precision
1445 [18981] math: i386 scalb*, ldexp return with excess range and precision
1446 [18982] stdio: va_list and vprintf
1447 [18985] time: Passing out of range data to strftime() causes a segfault
1449 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1450 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1451 [19012] locale: iconv_open leaks memory on error path
1452 [19016] math: clog, clog10 inaccuracy
1453 [19018] nptl: Mangle function pointers in tls_dtor_list
1454 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1455 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1456 [19048] malloc: malloc: arena free list can become cyclic, increasing
1458 [19049] math: [powerpc] erfc incorrect zero sign
1459 [19050] math: [powerpc] log* incorrect zero sign
1460 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1461 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1462 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1463 [19074] network: Data race in _res_hconf_reorder_addrs
1464 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1465 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1466 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1468 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1470 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1471 [19086] manual: posix_fallocate64 documented argument order is wrong.
1472 [19088] math: lround, llround missing exceptions close to overflow
1474 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1475 [19095] math: dbl-64 lrint incorrect for 64-bit long
1476 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1477 [19124] dynamic-link: ld.so failed to build with older assmebler
1478 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1479 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1480 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1481 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1482 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1483 pthread_setaffinity_np
1484 [19156] math: [ldbl-128] j0l spurious underflows
1485 [19164] nptl: tst-getcpu fails with many possible CPUs
1486 [19168] math: math/test-ildoubl and math/test-ldouble failure
1487 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1488 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1490 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1491 (FE_NOMASK_ENV) do not clear SSE exceptions
1492 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1493 _int_new_arena/reused_arena
1494 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1495 [19201] math: dbl-64 remainder incorrect sign of zero result
1496 [19205] math: bits/math-finite.h conditions do not match math.h and
1498 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1499 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1500 for C99-based standards
1501 [19212] libc: features.h not -Wundef clean
1502 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1504 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1505 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1506 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1508 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1509 "inexact" exceptions
1510 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1512 [19242] libc: strtol incorrect in Turkish locales
1513 [19243] malloc: reused_arena can pick an arena on the free list, leading
1514 to an assertion failure and reference count corruption
1515 [19253] time: tzset() ineffective when temporary TZ did not include DST
1517 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1518 [19270] math: [hppa] Shared libm missing __isnanl
1519 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1520 MAP_HUGETLB and MAP_STACK defines
1521 [19313] nptl: Wrong __cpu_mask for x32
1522 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1523 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1524 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1525 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1526 [19363] time: x32: times() return value wrongly truncates/sign extends
1528 [19367] dynamic-link: Improve branch prediction on Silvermont
1529 [19369] network: Default domain name not reset by res_ninit when "search"
1530 / "domain" entry is removed from resolv.conf
1531 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1533 [19385] localedata: bg_BG: time separator should be colon, not comma
1534 [19408] libc: linux personality syscall wrapper may erroneously return an
1535 error on 32-bit architectures
1536 [19415] libc: dladdr returns wrong names on hppa
1537 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1538 IBM905, IBM907, and IBM909
1539 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1540 [19443] build: build failures with -DDEBUG
1541 [19451] build: Make check fails on test-double-vlen2
1542 [19462] libc: Glibc failed to build with -Os
1543 [19465] math: Wrong code with -Os
1544 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1546 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1548 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1549 [19486] math: S390: Math tests fail with "Exception Inexact set".
1550 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1551 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1553 [19590] math: Fail to build shared objects that use libmvec.so functions.
1557 * The following bugs are resolved with this release:
1559 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1560 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1561 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1562 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1563 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1564 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1565 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1566 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1567 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1568 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1569 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1570 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1571 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1572 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1573 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1574 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1575 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1576 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1577 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1578 18657, 18676, 18694, 18696, 18887.
1580 * Cache information can be queried via sysconf() function on s390 e.g. with
1581 _SC_LEVEL1_ICACHE_SIZE as argument.
1583 * A buffer overflow in gethostbyname_r and related functions performing DNS
1584 requests has been fixed. If the NSS functions were called with a
1585 misaligned buffer, the buffer length change due to pointer alignment was
1586 not taken into account. This could result in application crashes or,
1587 potentially arbitrary code execution, using crafted, but syntactically
1588 valid DNS responses. (CVE-2015-1781)
1590 * The time zone file parser has been made more robust against crafted time
1591 zone files, avoiding heap buffer overflows related to the processing of
1592 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1593 large time zone data files. Overly long time zone specifiers in the TZ
1594 variable no longer result in stack overflows and crashes.
1596 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1597 for LD and GD on x86 and x86-64, has been implemented. You will need
1598 binutils-2.24 or later to enable this optimization.
1600 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1601 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1602 Hat). These updates cause user visible changes, such as the fix for bug
1605 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1606 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1607 condition in some applications.
1609 * Added vector math library named libmvec with the following vectorized x86_64
1610 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1612 The library can be disabled with --disable-mathvec. Use of the functions is
1613 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1614 Shared library libmvec.so is linked in as needed when using -lm (no need to
1615 specify -lmvec explicitly for not static builds).
1616 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1618 * A new fmemopen implementation has been added with the goal of POSIX
1619 compliance. The new implementation fixes the following long-standing
1620 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1621 old implementation is still present for use be by existing binaries.
1623 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1624 releases. It has been fixed to match 2.19 and older, but binaries built
1625 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1627 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1628 Contributed by Roland McGrath (Google).
1630 * The header <regexp.h> is deprecated, and will be removed in a future
1631 release. Use of this header will trigger a deprecation warning.
1632 Application developers should update their code to use <regex.h> instead.
1634 This header was formerly part of SUS, but was deprecated in 1994 and
1635 removed from the standard in 2001. Also, the glibc implementation
1636 leaks memory. See BZ#18681 for more details.
1640 * The following bugs are resolved with this release:
1642 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1643 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1644 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1645 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1646 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1647 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1648 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1649 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1650 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1651 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1654 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1655 too little memory for the to-be-scanned arguments and overflow the
1656 allocated buffer. The implementation now correctly computes the required
1657 buffer size when using malloc, and switches to malloc from alloca as
1660 * A new semaphore algorithm has been implemented in generic C code for all
1661 machines. Previous custom assembly implementations of semaphore were
1662 difficult to reason about or ensure that they were safe. The new version
1663 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1664 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1665 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1667 * Port to Altera Nios II has been contributed by Mentor Graphics.
1669 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1670 implementations for powerpc64/powerpc64le.
1671 Implemented by Adhemerval Zanella (IBM).
1673 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1674 and powerpc64le. This may improve lock scaling of existing programs on
1675 HTM capable systems. The lock elision code is only enabled with
1676 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1677 powerpc will issue a transaction abort on every syscall to avoid side
1678 effects being visible outside transactions.
1680 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1681 AArch64. Contributed by ARM Ltd.
1683 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1685 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1686 under certain input conditions resulting in the execution of a shell for
1687 command substitution when the application did not request it. The
1688 implementation now checks WRDE_NOCMD immediately before executing the
1689 shell and returns the error WRDE_CMDSUB as expected.
1691 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1692 processing format strings with a large number of format specifiers.
1694 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1695 infinite loop if the DNS response contained a PTR record of an unexpected
1698 * The minimum GCC version that can be used to build this version of the GNU
1699 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1700 still be used to compile programs using the GNU C Library.
1702 * The GNU C Library is now built with -Werror by default. This can be
1703 disabled by configuring with --disable-werror.
1705 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1707 * The obsolete sigvec function has been removed. This was the original
1708 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1709 programs have been using instead for about 25 years. Of course, ABI
1710 compatibility for old binaries using sigvec remains intact.
1712 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1713 with newer versions of bison.
1715 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1716 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1717 registers overlay two consecutive single-precision registers. MIPS32R2
1718 introduced a new FPU mode (FR=1) where double-precision registers extend the
1719 corresponding single-precision registers which is incompatible with the
1720 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1721 require the use of FR=1 making a transition necessary. New o32 ABI
1722 extensions enable users to migrate over time from the original o32 ABI
1723 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1724 tracks the ABI of any loaded object and verifies that new objects are
1725 compatible. Mode transitions will also be requested as required and
1726 unsupportable objects will be rejected. The ABI checks include both soft and
1727 hard float ABIs for o32, n32 and n64.
1729 GCC 5 with GNU binutils 2.25 onwards:
1730 It is strongly recommended that all o32 system libraries are built using the
1731 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1732 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1733 compiler using --with-fp-32=xx to set this by default.
1737 * The following bugs are resolved with this release:
1739 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1740 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1741 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1742 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1743 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1744 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1745 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1746 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1747 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1748 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1749 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1750 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1751 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1752 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1753 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1755 * Reverted change of ABI data structures for s390 and s390x:
1756 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1757 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1758 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1759 are preserved pointing straight to the same implementation as the old ones.
1760 Given that, new callers will simply provide a too-big buffer to these
1761 functions. Any applications/libraries out there that embed jmp_buf or
1762 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1763 against 2.19 headers will have to rebuilt again. This is necessary in any
1764 case to revert the breakage in their ABI caused by the glibc change.
1766 * Support for file description locks is added to systems running the
1767 Linux kernel. The standard file locking interfaces are extended to
1768 operate on file descriptions, not file descriptors, via the use of
1769 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1770 are associated with an open file instead of a process.
1772 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1774 * The minimum Linux kernel version that this version of the GNU C Library
1775 can be used with is 2.6.32.
1777 * Running the testsuite no longer terminates as soon as a test fails.
1778 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1779 with PASS or FAIL lines for individual tests. A summary of the results is
1780 printed, including a list of failing lists, and "make check" exits with
1781 error status if there were any unexpected failures. "make check
1782 stop-on-test-failure=y" may be used to keep the old behavior.
1784 * The am33 port, which had not worked for several years, has been removed
1787 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1788 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1789 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1790 conflicted with POSIX (support for which was removed in 2.19), the
1791 interfaces those macros enabled remain available when compiling with
1792 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1793 test macros defined.
1795 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1797 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1798 This may improve lock scaling of existing programs on TX capable systems.
1799 The lock elision code is only built with --enable-lock-elision=yes and
1800 then requires a GCC version supporting the TX builtins. With lock elision
1801 default mutexes are elided via __builtin_tbegin, if the cpu supports
1802 transactions. By default lock elision is not enabled and the elision code
1805 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1806 copy the path argument. This allowed programs to cause posix_spawn to
1807 deference a dangling pointer, or use an unexpected pathname argument if
1808 the string was modified after the posix_spawn_file_actions_addopen
1811 * All supported architectures now use the main glibc sysdeps directory
1812 instead of some being in a separate "ports" directory (which was
1813 distributed separately before glibc 2.17).
1815 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1816 On configurations that support it (all Linux configurations), it's now
1817 used regardless of the --enable-add-ons switch to configure. It is no
1818 longer possible to build such configurations without pthreads support.
1820 * Locale names, including those obtained from environment variables (LANG
1821 and the LC_* variables), are more tightly checked for proper syntax.
1822 setlocale will now fail (with EINVAL) for locale names that are overly
1823 long, contain slashes without starting with a slash, or contain ".." path
1824 components. (CVE-2014-0475) Previously, some valid locale names were
1825 silently replaced with the "C" locale when running in AT_SECURE mode
1826 (e.g., in a SUID program). This is no longer necessary because of the
1829 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1830 with application code using Intel MPX instructions. (With all previous
1831 versions, the MPX register state could be clobbered when making calls
1832 into or out of a shared library.) Note that while the new dynamic
1833 linker is compatible with all known x86 hardware whether or not it
1834 supports Intel MPX, some x86 instruction-set emulators might fail to
1835 handle the new instruction encodings. This is known to affect Valgrind
1836 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1837 release), and might affect other tools that do instruction emulation.
1839 * Support for loadable gconv transliteration modules has been removed.
1840 The support for transliteration modules has been non-functional for
1841 over a decade, and the removal is prompted by security defects. The
1842 normal gconv conversion modules are still supported. Transliteration
1843 with //TRANSLIT is still possible, and the //IGNORE specifier
1844 continues to be supported. (CVE-2014-5119)
1846 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1847 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1848 resulting a denial-of-service security vulnerability in applications which
1849 use functions related to iconv. (CVE-2014-6040)
1853 * The following bugs are resolved with this release:
1855 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1856 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1857 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1858 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1859 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1860 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1861 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1862 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1863 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1864 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1865 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1866 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1867 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1868 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1869 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1870 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1871 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1872 16431, 16453, 16474, 16506, 16510, 16529
1874 * Slovenian translations for glibc messages have been contributed by the
1875 Translation Project's Slovenian team of translators.
1877 * The public headers no longer use __unused nor __block. This change is to
1878 support compiling programs that are derived from BSD sources and use
1879 __unused internally, and to support compiling with Clang's -fblock
1880 extension which uses __block.
1882 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1883 large collation sequences to optimize multiple passes. This cache
1884 computation may overflow for large collation sequences and may cause a
1885 stack or buffer overflow. This is now fixed to use a slower algorithm
1886 which does not use a cache if there is an integer overflow.
1888 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1889 rules for large collation sequences to optimize multiple passes and falls
1890 back to alloca if malloc fails, resulting in a possible stack overflow.
1891 The implementation now falls back to an uncached collation sequence lookup
1894 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1895 initialized for static applications resulting in the security feature
1896 being disabled. The pointer guard is now correctly initialized to a
1897 random value for static applications. Existing static applications need
1898 to be recompiled to take advantage of the fix (bug 15754).
1900 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1901 to the d_name member of struct dirent, or omit the terminating NUL
1902 character. (Bugzilla #14699).
1904 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1905 aligned_alloc functions could allocate too few bytes or corrupt the
1906 heap when passed very large allocation size values (Bugzilla #15855,
1909 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1910 for AF_INET6 has been fixed (Bugzilla #16072).
1912 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1913 pap_CW, quz_PE, the_NP.
1915 * Substantially revised locales: gd_GB, ht_HT
1917 * The LC_ADDRESS field was updated to support country_car for almost all
1920 * ISO 1427 definitions were updated.
1922 * ISO 3166 definitions were updated.
1924 * The localedef utility now supports --big-endian and --little-endian
1925 command-line options to generate locales for a different system from that
1926 for which the C library was built.
1928 * Binary locale files now only depend on the endianness of the system for
1929 which they are generated and not on other properties of that system. As a
1930 consequence, binary files generated with new localedef may be incompatible
1931 with old versions of the GNU C Library, and binary files generated with
1932 old localedef may be incompatible with this version of the GNU C Library,
1933 in the following circumstances:
1935 + Locale files may be incompatible on m68k systems.
1937 + Locale archive files (but not separate files for individual locales) may
1938 be incompatible on systems where plain "char" is signed.
1940 * The configure option --disable-versioning has been removed. Builds with
1941 --disable-versioning had not worked for several years.
1943 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1944 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1946 * SystemTap probes for malloc have been introduced.
1948 * SystemTap probes for slow multiple precision fallback paths of
1949 transcendental functions have been introduced.
1951 * Support for powerpc64le has been added.
1953 * The soft-float powerpc port now supports e500 processors.
1955 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1957 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1958 set of header declarations that are enabled by default, even when other
1959 feature test macros or compiler options such as -std=c99 would otherwise
1960 disable some of those declarations.
1962 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1963 conflict with POSIX. The libbsd-compat library (which was a dummy library
1964 that did nothing) has also been removed.
1966 * Preliminary documentation about Multi-Thread, Async-Signal and
1967 Async-Cancel Safety has been added.
1969 * Change of ABI data structures for s390 and s390x:
1970 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1971 allow for future hardware extensions. All library functions that accept or
1972 return these structures were versioned in order to provide backwards
1973 compatibility with existing code. However, not all cases can be handled
1974 with interface versioning. If an application embeds either structure into
1975 another structure and that structure is passed to another compilation unit
1976 which expects the newer larger-sized object then any access to the new
1977 fields will result in undefined behaviour. Similarly any access to fields
1978 that were moved by the enlarging of the structures will cause undefined
1979 behaviour. To fix the undefined behaviour all objects that are part of the
1980 final in-memory image must agree on the size of structures, and this may
1981 require recompilation.
1985 * The following bugs are resolved with this release:
1987 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1988 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1989 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1990 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1991 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1992 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1993 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1994 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1995 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1996 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1997 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1998 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1999 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2002 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2003 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2004 Distributions can re-enable building and using pt_chown via the new configure
2005 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2006 considerable security risks and should only be used if the distribution
2007 understands and accepts the risks.
2009 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2012 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2013 fixed (Bugzilla #15330).
2015 * Add support for calling C++11 thread_local object destructors on thread
2016 and program exit. This needs compiler support for offloading C++11
2017 destructor calls to glibc.
2019 * Improved worst case performance of libm functions with double inputs and
2022 * Support for priority inherited mutexes in pthread condition variables on
2023 non-x86 architectures.
2025 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2027 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2029 * Optimized string functions for ARM. Implemented by Will Newton and
2032 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2034 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2037 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2038 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2040 * Added a benchmark framework to track performance of functions in glibc.
2042 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2043 It is based on draft TS 18661 and currently enabled as a GNU extension.
2045 * On Linux, the clock function now uses the clock_gettime system call
2046 for improved precision, rather than old times system call.
2048 * Added support for version-3 tz format files. This is needed when using
2049 the tz database release 2013e or later, and affects a few unusual cases --
2050 currently only TZ='America/Godthab' for time stamps after 2037.
2052 * Added new API functions pthread_getattr_default_np and
2053 pthread_setattr_default_np to get and set the default pthread
2054 attributes of a process.
2056 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2057 This may improve lock scaling of existing programs on TSX capable systems.
2058 When the --enable-lock-elision=yes parameter is specified at configure
2059 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2062 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2063 availability). Implemented by Ryan S. Arnold.
2065 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2069 * The following bugs are resolved with this release:
2071 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2072 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2073 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2074 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2075 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2076 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2077 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2078 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2079 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2080 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2081 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2082 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2083 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2084 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2085 14879, 14889, 14893, 14898, 14914.
2087 * Optimization of memcpy for MIPS.
2089 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2090 EMFILE has been fixed (Bugzilla #14889).
2092 * The add-on ports collection is now distributed in the "ports" subdirectory
2093 of the main GNU C Library distribution, rather than separately.
2095 * Port to ARM AArch64 contributed by Linaro.
2097 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2098 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2100 Implemented by Andreas Krebbel.
2102 * The new function secure_getenv allows secure access to the environment,
2103 returning NULL if running in a SUID/SGID process. This function replaces
2104 the internal function __secure_getenv.
2106 * SystemTap static probes have been added into the dynamic linker.
2107 Implemented by Gary Benson.
2109 * Optimizations of string functions strstr, strcasestr and memmem.
2110 Implemented by Maxim Kuvyrkov.
2112 * The minimum Linux kernel version that this version of the GNU C Library
2113 can be used with is 2.6.16.
2115 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2116 powerpc POWER7. Implemented by Will Schmidt.
2118 * New configure option --disable-nscd builds the C library such that it
2119 never attempts to contact the Name Service Caching Daemon (nscd).
2120 New configure option --disable-build-nscd avoids building nscd itself;
2121 this is the default if --disable-nscd is used.
2123 * Improved support for cross-compilation, including cross-testing and
2124 bootstrap builds without a previously built glibc.
2126 * Several testsuite tests are now able to test multiple IFUNC variants of an
2127 interface, rather than just testing the one that would be chooen by
2130 * New configure options --with-bugurl and --with-pkgversion, for
2131 distributors to use to embed their bug-reporting and package version
2132 information in --help and --version output.
2134 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2135 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2136 allows creation of chroots without the procfs mounted on /proc.
2138 * The `crypt' function now fails if passed salt bytes that violate the
2139 specification for those values. On Linux, the `crypt' function will
2140 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2141 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2142 when the mode is enabled.
2144 * The `clock_*' suite of functions (declared in <time.h>) is now available
2145 directly in the main C library. Previously it was necessary to link with
2146 -lrt to use these functions. This change has the effect that a
2147 single-threaded program that uses a function such as `clock_gettime' (and
2148 is not linked with -lrt) will no longer implicitly load the pthreads
2149 library at runtime and so will not suffer the overheads associated with
2150 multi-thread support in other code such as the C++ runtime library.
2152 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2157 * The following bugs are resolved with this release:
2159 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2160 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2161 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2162 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2163 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2164 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2165 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2166 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2167 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2168 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2169 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2170 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2171 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2172 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2173 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2174 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2175 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2176 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2177 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2178 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2179 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2180 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2183 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2184 configuring glibc with:
2185 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2186 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2187 Implemented by H.J. Lu.
2191 + define static_assert
2193 + do not declare gets
2195 + declare at_quick_exit and quick_exit also for ISO C11
2197 + aligned_alloc. NB: The code is deliberately allows the size parameter
2198 to not be a multiple of the alignment. This is a moronic requirement
2199 in the standard but it is only a requirement on the caller, not the
2202 + timespec_get added
2204 + uchar.h support added
2206 + CMPLX, CMPLXF, CMPLXL added
2208 Implemented by Ulrich Drepper.
2210 * Support for the IA-64 has been moved to ports.
2212 * Remove support for anything but ELF binary format
2214 * Checking versions of poll, ppoll added.
2215 Implemented by Ulrich Drepper.
2217 * More generic and 64-bit performance optimizations to math functions.
2218 Implemented by Ulrich Drepper.
2220 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2221 headers and functions available at compile time as they were before
2222 version 2.14. This option will be removed at some time in the future
2223 after the TI-RPC library becomes fully sufficient for the needs of
2224 existing applications.
2226 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2227 Note that glibc is not expected to work with any Linux kernel version
2230 * New header <sys/auxv.h> and function getauxval allowing easy access to
2231 the AT_* key-value pairs passed from the Linux kernel. The header also
2232 defines the HWCAP_* bits associated with the AT_HWCAP key.
2234 * New locales: mag_IN
2236 * New configure option --enable-systemtap builds SystemTap static probes
2237 into libc for setjmp and longjmp and into libpthread for various operations.
2238 So far the setjmp/longjmp probes and some of the libpthread probes are
2239 provided only for i*86 and x86_64.
2240 Implemented by Roland McGrath and Rayson Ho.
2242 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2244 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2247 * More optimized functions for SPARC. Implemented by David S. Miller.
2249 * Improved support for cross-compilation, especially bootstrap builds
2250 without a previously built glibc.
2252 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2253 Chris Metcalf from Tilera.
2255 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2256 now supported for ARM processors.
2258 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2259 as the name of the dynamic linker, to distinguish it from the
2260 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2262 * Support for CRIS has been removed from ports.
2264 * A new class of installed header has been documented for low-level
2265 platform-specific functionality. PowerPC added the first instance with a
2266 function to provide time base register access. Contributed by Tulio
2267 Magno Quites Machado Filho.
2269 * ix86 configurations now install header files that are consistent with
2270 what x86-64 configurations install. These same header files can be used
2271 for -m32, -m64, or -mx32 builds.
2272 Contributed by H.J. Lu.
2274 * Math library bug fixes. A thorough audit of all open math library bugs was
2275 conducted by Joseph Myers. Significant progress was made on many math
2276 library bugs resulting in more accurate exceptions and function results.
2277 Many thanks to all those that contributed including Andreas Jaeger for his
2278 patch review and work on the x87 trigonometric instruction issues.
2280 * Timezone data is no longer installed. Timezone-related binaries and scripts
2281 will continue to be installed. Users should obtain their timezone data from
2282 their distribution provider or from the tzdata package at
2283 <ftp://munnari.oz.au/pub/>.
2287 * The following bugs are resolved with this release:
2289 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2290 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2291 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2292 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2293 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2294 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2295 13484, 13506, 13515, 13523, 13524, 13538, 13540
2297 * New program pldd to list loaded object of a process
2298 Implemented by Ulrich Drepper.
2300 * Add nss_db support back to glibc. No more dependency on Berkeley db
2301 and support for initgroups lookups.
2302 Implemented by Ulrich Drepper.
2304 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2305 Contributed by HJ Lu.
2307 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2308 Contributed by HJ Lu.
2310 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2311 on x86-32 and x86-64.
2312 Contributed by Liubov Dmitrieva.
2314 * Optimized strchr and strrchr for SSE on x86-32.
2315 Contributed by Liubov Dmitrieva.
2317 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2318 for x86-64 and x86-32.
2319 Contributed by Liubov Dmitrieva.
2321 * New interfaces: scandirat, scandirat64
2322 Implemented by Ulrich Drepper.
2324 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2325 Implemented by Ulrich Drepper.
2327 * nscd now also caches the netgroup database.
2328 Implemented by Ulrich Drepper.
2330 * Integrate libm with gcc's -ffinite-math-only option.
2331 Implemented by Ulrich Drepper.
2333 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2334 to math functions. Implemented by Ulrich Drepper.
2336 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2337 Implemented by Ulrich Drepper.
2339 * New Linux interfaces: process_vm_readv, process_vm_writev
2341 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2342 Implemented by Ulrich Drepper.
2344 * Optimized nearbyint and strcasecmp for PPC.
2345 Implemented by Adhemerval Zanella.
2347 * New locales: bho_IN, unm_US, es_CU, ta_LK
2351 * The following bugs are resolved with this release:
2353 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2354 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2355 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2356 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2357 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2358 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2359 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2360 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2361 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2362 12795, 12811, 12813, 12814, 12841
2364 * The RPC implementation in libc is obsoleted. Old programs keep working
2365 but new programs cannot be linked with the routines in libc anymore.
2366 Programs in need of RPC functionality must be linked against TI-RPC.
2367 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2369 Visible changes of this change include (obviously) the inability to link
2370 programs using RPC functions without referencing the TI-RPC library and the
2371 removal of the RPC headers from the glibc headers.
2372 Implemented by Ulrich Drepper.
2374 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2375 syncfs, setns, sendmmsg
2377 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2378 yue_HK, lij_IT, mhr_RU
2380 * New encodings: CP770, CP771, CP772, CP773, CP774
2382 * New program sotruss to trace calls through PLTs
2383 Implemented by Ulrich Drepper.
2385 * The malloc hook implementation is marked deprecated and will be removed
2386 from the default implementation in the next version. The design never
2387 worked ever since the introduction of threads. Even programs which do
2388 not create threads themselves can use multiple threads created internally.
2392 * The following bugs are resolved with this release:
2394 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2395 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2396 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2397 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2400 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2402 * POWER7 optimizations: memset, memcmp, strncmp
2404 * New optimized string functions for x86-64: strnlen (SSE2),
2405 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2406 Implemented by Ulrich Drepper.
2410 * The following bugs are resolved with this release:
2412 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2413 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2414 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2415 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2416 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2417 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2418 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2419 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2421 * New interfaces: pthread_getname_np, pthread_setname_np
2423 * New Linux interface: recvmmsg
2425 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2427 * The dynamic linker now recognizes supported ABI versions from the
2428 EI_ABIVERSION field in the ELF header.
2429 Implemented by Ulrich Drepper.
2431 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2432 will not be used to fill in password fields in the passwd.byname replies.
2433 Instead it is used to synthesize the shadow.byname table, should it be
2434 missing. This is a useful mode in some installations involving Solaris.
2435 Implemented by Ulrich Drepper.
2437 * New locales: kok_IN, sq_MK, cv_RU
2441 * The following bugs are resolved with this release:
2443 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2444 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2445 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2446 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2447 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2448 10789, 10817, 10823, 10840, 10847
2450 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2452 Implemented by Ulrich Drepper.
2454 * Checking version of longjmp added that fails if an uninitialized stack
2455 frame would be created. Implemented by Ulrich Drepper.
2457 * STT_GNU_IFUNC is now supported in static executables.
2458 Implemented by H.J. Lu.
2460 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2462 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2463 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2464 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2465 Contributed by H.J. Lu.
2467 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2468 Implemented by Ulrich Drepper.
2470 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2472 Contributed by H.J. Lu.
2474 * Support for fma instruction in AVX on x86-64.
2475 Implemented by H.J. Lu and Ulrich Drepper.
2477 * AVX support in x86-64 auditing support in ld.so.
2478 Implemented by H.J. Lu.
2480 * STB_GNU_UNIQUE support added.
2481 Implemented by Ulrich Drepper.
2483 * Implement second fallback mode for optimized DNS lookup for even more
2484 broken environments. If two requests from the same port are not
2485 handled correctly close the socket and open a new one before sending
2486 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2487 can be used to select this mode right away, instead of rediscovering the
2488 necessity is every process again.
2489 Implemented by Ulrich Drepper.
2491 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2492 Implemented by Adam Tkac.
2494 * Optimized iconv conversions for S390x.
2495 Implemented by Andreas Krebbel.
2497 * Using condvars with PI mutexes is now more efficient due to kernel
2498 support for requeueing to PI futexes. NPTL support added for x86-64.
2499 Implemented by Ulrich Drepper.
2501 * New locales: ps_AF, my_MM
2505 * The following bugs are resolved with this release:
2507 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2508 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2509 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2510 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2511 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2513 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2514 Implemented by Ulrich Drepper.
2516 * New Linux interfaces: accept4, fallocate, fallocate64.
2517 Implemented by Ulrich Drepper.
2519 * Correct declarations of string function when used in C++ code. This
2520 could lead to compile errors for invalid C++ code.
2522 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2525 * New POSIX 2008 interface: psiginfo
2526 Implemented by Ulrich Drepper.
2528 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2529 Implemented by Ulrich Drepper.
2531 * Support for selecting between multiple function definitions at runtime
2532 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2534 * The libcrypt library can now use the hash function implementations in
2535 NSS. Implemented by Ulrich Drepper.
2537 * The malloc implementation can be compiled to be less memory efficient
2538 but higher performing in multi-threaded programs.
2539 Implemented by Ulrich Drepper.
2541 * New locales: nan_TW@latin, ks_IN
2543 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2544 Implemented by Ulrich Drepper.
2546 * Extended printf hook support. It is possible to use user-defined types
2547 and extend existing format specifiers.
2548 Implemented by Ulrich Drepper.
2550 * Handling for group shadow files has been added.
2551 Implemented by Ulrich Drepper.
2553 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2554 servers (the case, e.g., for some people using the built-in DNS
2555 server in ADSL modems/routers). There is a once-per-process timeout
2556 in case of a broken server. To avoid it, users can run nscd or put
2557 'options single-request' in /etc/resolv.conf.
2558 Implemented by Ulrich Drepper.
2562 * The following bugs are resolved with this release:
2564 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2565 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2566 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2567 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2568 6968, 6974, 6980, 6995, 7008, 7009, 7029
2570 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2571 up at the same time. Implemented by Ulrich Drepper.
2573 * TLS descriptors for LD and GD on x86 and x86-64.
2574 Implemented by Alexandre Oliva.
2576 * getaddrinfo now handles DCCP and UDPlite.
2577 Implemented by Ulrich Drepper.
2579 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2580 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2581 Implemented by Ulrich Drepper.
2583 * New implementation of memmem, strstr, and strcasestr which is O(n).
2584 Implemented by Eric Blake.
2586 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2588 * Implement "e" option for popen to open file descriptor with the
2589 close-on-exec flag set. Implemented by Ulrich Drepper.
2591 * Implement "b" mode for fmemopen. In this mode writes writes don't
2592 implicitly add a NUL byte and seeks from the end of the buffer really
2593 use the buffer end, not the string length as the basis.
2594 Implemented by Ulrich Drepper.
2596 * Many functions, exported and internal, now atomically set the close-on-exec
2597 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2599 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2601 Implemented by Pravin Satpute.
2603 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2607 * The following bugs are resolved with this release:
2609 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2610 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2611 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2612 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2613 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2614 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2615 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2617 * New locales: bo_CN, bo_IN, shs_CA.
2619 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2621 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2622 Implemented by Pravin Satpute.
2624 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2626 * nscd caches now all timeouts for DNS entries
2627 Implemented by Ulrich Drepper.
2629 * nscd is more efficient and wakes up less often.
2630 Implemented by Ulrich Drepper.
2632 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2633 vdprintf, and obstack_vprintf.
2634 Implemented by Jakub Jelinek.
2636 * Faster memset for x86-64.
2637 Implemented by Harsha Jagasia and H.J. Lu.
2639 * Faster memcpy on x86.
2640 Implemented by Ulrich Drepper.
2642 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2643 Implemented by Ulrich Drepper.
2645 * Faster sqrt and sqrtf implemention for some PPC variants.
2646 Implemented by Steven Munroe.
2650 * The following bugs are resolved with this release:
2652 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2653 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2654 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2655 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2656 5071, 5103, 5104, 5112, 5113, 5184, 5186
2658 * More checking functions: fread, fread_unlocked, open*, mq_open.
2659 Implemented by Jakub Jelinek and Ulrich Drepper.
2661 * Extend fortification to C++. Implemented by Jakub Jelinek.
2663 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2664 by not recognizing 'a' as a modifier when those specs are requested.
2665 Implemented by Jakub Jelinek.
2667 * PPC optimizations to math and string functions.
2668 Implemented by Steven Munroe.
2670 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2671 options to be passed. Implemented by Ulrich Drepper.
2673 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2675 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2676 Implemented by Ulrich Drepper.
2678 * Handle private futexes in the NPTL implementation.
2679 Implemented by Jakub Jelinek and Ulrich Drepper.
2681 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2682 Implemented by Roland McGrath and Ulrich Drepper.
2684 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2686 * SHA-256 and SHA-512 based password encryption.
2687 Implemented by Ulrich Drepper.
2689 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2690 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2693 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2694 Implemented by Ulrich Drepper.
2698 * The following bugs are resolved with this release:
2700 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2701 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2702 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2703 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2704 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2705 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2706 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2709 * New Linux interfaces: epoll_pwait, sched_getcpu.
2711 * New generic interfaces: strerror_l.
2713 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2717 * The following bugs are resolved with this release:
2719 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2720 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2721 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2722 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2723 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2724 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2725 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2726 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2727 3225, 3251, 3252, 3253, 3273
2729 * For Linux, the sorting of addresses returned by getaddrinfo now also
2730 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2731 Implemented by Ulrich Drepper.
2733 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2734 Implemented by Ulrich Drepper.
2736 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2738 * New iconv module for MIK. Contributed by Alexander Shopov.
2740 * For sites with broken group and/or passwd database, the auto-propagate
2741 option of nscd can prevent creating ID lookup entries from the results
2742 of a name lookup and vice versa. This usually is no problem but some
2743 site might have problems with the default behavior.
2744 Implemented by Ulrich Drepper.
2746 * Iterating over entire database in NIS can be slow. With the
2747 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2748 to trade time for memory. The entire database will be read at once.
2749 Implemented by Ulrich Drepper.
2751 * The interfaces introduced in RFC 3542 have been implemented by
2754 * Support for the new ELF hash table format was added by Ulrich Drepper.
2756 * Support for priority inheritance mutexes added by Jakub Jelinek and
2759 * Support for priority protected mutexes added by Jakub Jelinek.
2761 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2762 or_IN, csb_PL, fy_NL, sr_ME.
2766 * More overflow detection functions.
2768 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2769 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2771 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2772 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2773 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2774 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2775 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2776 by Masahide Washizawa.
2778 * It is now possible to install audit libraries for the dynamic linker using
2779 LD_AUDIT. Implemented by Ulrich Drepper.
2781 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2782 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2783 For a libc and libpthread that works well on Linux 2.4 kernels, we
2784 recommend using the stable 2.3 branch.
2786 * The new function open_wmemstream acts like open_memstream,
2787 but uses a wchar_t wide character buffer.
2789 * The new function ppoll is an improved version of poll, similar to pselect.
2791 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2792 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2793 renameat, unlinkat, linkat, symlinkat, readlinkat.
2795 * New Linux kernel system calls: unshare,
2796 inotify_init, inotify_add_watch, inotify_rm_watch.
2798 * The euidaccess function is now also known by the alias eaccess,
2799 for compatibility with some other systems.
2801 * Timezone data updated to 2006b version.
2805 * The following bugs are resolved with this release:
2807 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2808 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2809 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2810 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2811 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2812 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2814 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2816 * As of this release, GCC 4 can be used to compile the C Library.
2818 * Timezone data updated to 2005m version.
2822 * The following bugs are resolved with this release:
2824 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2825 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2826 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2827 777, 787, 821, 822, 823, 825
2829 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2833 * Support for RFC 3678. Real implementations exist only for Linux so far.
2834 Implemented by Ulrich Drepper.
2836 * nscd can now cache entries persistently. Expiring entries are reloaded.
2837 For speedups the cache can be shared in memory with client processes.
2838 Implemented by Ulrich Drepper.
2840 * nscd can now perform SELinux checks.
2841 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2843 * getaddrinfo queries are now cached. Canonical name lookup is performed
2845 Implemented by Ulrich Drepper.
2847 * The nothrow function attribute is used when headers are used by gcc when
2848 compiling C code. This can avoid significant amounts of exception
2851 * The malloc functions perform more error checking and are stricter when
2852 it comes to reacting on errors. The default action is to terminate
2853 the process after showing an error message. Implemented by Ulrich Drepper.
2855 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2856 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2858 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2859 namespaces using the new function dlmopen(). This feature is of course,
2860 like most other dynamic loading functionality, not available in statically
2861 linked applications. Implemented by Ulrich Drepper.
2863 * Low-overhead boundary checking variants of string and some stdio functions
2864 were added. These are to be used in conjunction with a gcc patch by
2865 Jakub Jelinek which adds calls to these functions if possible.
2866 Implemented by Jakub Jelinek and Ulrich Drepper.
2868 * Old code for several operating systems and machine architectures that
2869 have not been in working condition in a long time have been removed from
2870 the main source tree maintained by the GNU C Library's maintainers.
2871 These files are now reside in the separate `ports' source module
2872 that is usable as an add-on when building the library.
2876 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2877 interrogate the dynamic linker, compatible with the Solaris interface.
2879 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2880 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2882 * getifaddrs now uses the netlink interface on Linux to get its information.
2883 Implemented by Thorsten Kukuk.
2885 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2886 Implemented by Ulrich Drepper.
2888 * support for non-executable stacks on x86 has been added. Changes mostly
2891 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2894 * getaddrinfo now performs destination address selection according to
2899 * Thread-safe interfaces for many functions that access locale data
2900 were added in version 2.3, but these features were omitted from NEWS.
2901 Many functions have variants with an `_l' suffix that take a `locale_t'
2902 object as a parameter rather than consulting the current locale.
2903 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2904 create and maintain `locale_t' objects. Additionally, the new function
2905 `uselocale' sets "the current locale" (as used by functions not so
2906 parameterized) set for an individual thread. These features were added
2907 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2909 * The functions getresuid, getresgid, setresuid, and setresgid, which
2910 have long been available on Linux, are now declared in <unistd.h>
2911 and are now also available on the Hurd.
2913 * ELF thread-local storage support (TLS) now works on x86-64.
2915 * The new dynamic string token $LIB is expanded in shared library names.
2916 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2918 * Aldy Hernandez contributed complete software floating point support for
2919 PowerPC machines with no FPU.
2921 * fexecve is implemented on Linux.
2923 * The `btowc' function should work at least twice as fast due to
2924 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2926 * With approriate thread add-ons cancelable functions are now implemented
2927 in libc.so as well. No need to call the function in libpthread. This
2928 change allowed to finally disable the incorrect and expensive handling
2929 of weak definition in ld.so.
2931 * Yet more PLT entries in libc.so have been removed. We finally arrived
2932 at the bare minimum. Startup times improved appropriately.
2934 * Support for the new Linux/x86 system call interface was added. The
2935 AT_SYSINFO auxiliary vector entry is recognized and handled.
2939 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2942 * iconv (the program and the interface) now accepts empty names (excluding
2943 options like //TRANSLIT) to mean "use charset of current locale".
2945 * localedef can now transliterate characters in strings which are not in
2946 the provided charmap. The information from the input locale is used.
2948 * Prelinking support was added for ELF targets. This requires additional
2949 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2951 * Read-only stdio streams now use mmap to speed up operation by eliminating
2952 copying and buffer underflows. To use add 'm' to the mode string of
2953 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2955 * The malloc functions were completely rewritten by Wolfram Gloger based
2956 on Doug Lea's malloc-2.7.0.c.
2958 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2959 implementation of regex.
2961 * Bruno Haible upgraded the iconv and locale implementation to support
2964 * Contents of the LC_* and LANG environment variables in the CEN style are
2965 not recognized anymore. It never was used. Change by Ulrich Drepper.
2967 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2968 thread-local storage (TLS) ABI on some platforms.
2969 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2971 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2972 EUC-JISX0213, and TSCII.
2974 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2975 BSD-compatible interface for getting all network interface addresses.
2976 Implementation for IPv4 by Roland McGrath.
2978 * Loading of locale data is faster due to the introduction of a locale
2979 archive. Implemented by Roland McGrath and Ulrich Drepper.
2981 * Startup times are significantly reduced by not using exported functions
2982 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2985 * Steven Munroe contributed a port to PowerPC64/Linux.
2989 * The Hurd now uses the GNU libio implementation of stdio rather than the
2990 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2992 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2993 with the same behavior as the Linux system call of the same name.
2997 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2998 128-bit long double format.
3000 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3001 IBM1161, and IBM1162 charsets.
3003 * Andreas Jaeger contributed a port to x86-64/Linux.
3005 * Peter Bruin contributed a port to PowerPC/Hurd.
3007 * libc functions using I/O streams now can handle wide-oriented streams
3010 * optimizations in the dynamic linker. Binaries created by recent binutils
3011 versions start up quicker due to reduced time spent on relocations.
3013 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3017 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3018 asinh, atanh, j0 for the 128-bit long double format.
3020 * Bruno Haible updated all the code handling Unicode in some form to
3021 support Unicode 3.1.
3023 * Speed of regex for single-byte locales is back to previous levels.
3024 Patch by Isamu Hasegawa.
3026 * Alpha, SPARC, and IA-64 now also using floating stacks.
3028 * Startup time of internationalized applications greatly improved through
3029 iconv cache. Use iconvconfig to generate the cache file.
3030 Contributed by Ulrich Drepper.
3032 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3033 severe performance problems. Patch by David Mosberger.
3035 * The Hurd port got a lot more functionality like AIO, various stdio
3036 extensions, etc. Mainly done by Roland McGrath.
3038 * mtrace can now lookup symbols in shared libraries.
3042 * Intel's IA-64 math library is largely integrated. It provides fast and
3043 accurate implementatations for most basic and standard math functions
3044 in float, double, and long double format.
3046 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3047 and asin for the 96-bit long double format and asin, log, tan for the
3048 128-bit long double format.
3050 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3051 The basic double functions exist today. Contributed by Abraham Ziv
3052 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3053 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3055 * An asynchronous name lookup library was added. The interface is designed
3056 after POSIX AIO. The proposal was circulated beforehand to get comments.
3057 No negative ones came in. Implemented by Ulrich Drepper.
3059 * Port to S390/64bit contributed by Martin Schwidefsky
3060 <schwidefsky@de.ibm.com>.
3062 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3063 of functions for Linux/IA-64.
3065 * The RPC code is now thread safe. Threads can now use the same service
3066 of different services at the same time. Patch by Eric Norum
3067 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3069 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3070 family of functions for Linux/S390.
3072 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3073 of functions for Linux/x86.
3075 * Port to Linux/CRIS contributed by Axis Communications.
3079 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3080 we can now check for namespace violations and missing declarations. The
3081 result is that almost all headers are now Unix-compliant (as defined in
3082 the upcoming XPG6). The negative side is that some programs might need
3083 corrections, too, if they depend on the incorrect form of the headers in
3084 previous versions which defined too many symbols and included too many
3087 * regex now handles multibyte character sets correctly.
3088 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3090 * iconv (the program) does now conform to the upcoming XPG6 and handles
3091 charmaps. Instead of the charset names the path of charmaps can be
3092 provided and the conversion happens based on this data.
3093 Contributed by Ulrich Drepper.
3095 * The locale program now provides detailed information about the installed
3100 only lists the names of the supported locales
3104 provides details such as country, language, and codeset name.
3105 Contributed by Ulrich Drepper.
3109 * The gencat program now parses the input file according to the charset
3110 selected by the LC_CTYPE category. This is important for stateful
3111 character sets. To make generating catalogs easier there is a way
3112 to overwrite the charset selected by the locale: before the first
3113 message or $ quote line the catalog can contain a line like
3115 $ codeset=ISO-8859-2
3117 to select the charset (ISO-8859-2 in this case).
3119 Implemented by Shinya Hanataka and Ulrich Drepper.
3121 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3122 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3123 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3125 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3126 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3128 * The LANGUAGE environment variable is now ignored unless the locale is
3129 changed from the default "C" locale.
3131 * The usual bug fixes.
3135 * Greg McGary added runtime support for bounds checking using gcc's
3136 new -fbounded-pointers option. ix86 target is complete. PowerPC
3139 * Thorsten Kukuk added secure mode support to nscd.
3141 * The Berkeley db libraries have been removed.
3143 Related, the nss_db module is now in a separate package since it
3144 obviously requires a database library being available.
3146 * Wide character I/O streams implemented by Ulrich Drepper.
3148 * Functions from the extended socket API added by Ulrich Drepper.
3150 * Functions feenableexcept and fedisableexcept to control the
3151 behaviour of individual exceptions have been added by Andreas Jaeger.
3153 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3155 * The resolver code has been updated from bind 8.2.3-T5B which supports
3156 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3159 This change could in some situations effect backward compatibility. Since
3160 now `_res' is a thread-local instead of a global variable, modifying it
3161 in one thread does not have any effect in other threads.
3163 The resolver library was also extended to allow IPv6 as the transport
3164 protocol for the requests. This work was done by Stig Venaas.
3166 * Compatibility code for K&R C compilers has been removed from the
3167 header files. A ISO C compiler is needed to use the library
3168 (conforming to either C89 or C99 standard).
3170 * Complete rewrite of the localedef program to support multibyte character
3171 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3172 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3173 Implemented by Ulrich Drepper.
3175 Bruno Haible significantly improved the generation and use of the data
3176 structures for the wide character tables.
3178 * Plural handling in gettext implemented by Ulrich Drepper.
3180 * The utmp daemon has been removed.
3182 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3184 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3187 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3189 * POSIX barriers implemented by Kaz Kylheku.
3191 * POSIX spawn function family implemented by Ulrich Drepper.
3193 * POSIX spinlocks are now available.
3195 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3197 * the configure option --enable-kernel=X.Y.Z allows to strip out
3198 compatibility for kernel versions before X.Y.Z. This is currently only
3199 implemented for Linux.
3201 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3202 field sin6_scope_id. This means that all programs using IPv6 should be
3203 recompiled. Don't expect binary compatibility with previous glibc
3206 * various conversion modules for IBM character sets contributed by
3209 * IA-64 port by Jes Sorensen and HJ Lu.
3223 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3225 * New xdr functions are added; some rpc functions are now 64bit clean.
3227 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3229 * Fixed known incompatibilities with glibc 2.0.
3231 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3233 * Optimized a number of functions (especially the ELF dynamic loader).
3235 * Update timezone data files.
3237 * lots of charmaps corrections
3239 * some new locale definitions and charmaps
3244 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3245 conform to POSIX member type requirements. Symbol versions have been
3246 adjusted as needed within the library, and for direct use by applications,
3247 but there is potential for problems if third-party libraries use
3248 struct timeval as part of their interface. This does not present
3249 a problem for X and other "essential" system libraries.
3251 * An additional locale model to support C++ Standard Library locale
3252 model and probably more was implemented by Ulrich Drepper.
3254 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3257 * Miles Bader provided the `argp' function family to support hierachical
3258 command line argument parsing, layered on top of getopt.
3260 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3262 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3265 * scanf recognizes the %a and %A format for scanning floating point numbers.
3267 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3268 information and interfaces for the available integer types.
3270 * about 130 new math functions were added to implement the ISO C9x math
3273 * the new header <complex.h> contains definitions of the complex math
3274 functions from ISO C 9X.
3276 * the new header <tgmath.h> defines generic macros to use complex or
3277 real valued functions.
3279 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3281 * Andreas Jaeger provided a test suite for the math library.
3283 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3285 * Ulrich Drepper added character set conversion functions (iconv).
3287 * Optimized string functions have been added.
3289 * The localedata addon is now part of glibc.
3291 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3293 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3294 daemon for NSS (nscd).
3296 Missing a better place here are some numbers on improvements. Under
3297 Linux 2.1.125 un-tar-ing the kernel sources takes
3301 using local files 12.19s 6.88s 22.91s
3303 using NIS 13.92s 8.91s 26.34s
3305 using NIS & nscd 10.37s 7.34s 25.30s
3307 using NIS+ 27.57s 30.37s 640.46s
3309 using NIS+ & nscd 10.25s 7.83s 26.51s
3311 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3313 Keep in mind that non-namelookup related operations dominate above times.
3314 It was just a common complain that using NIS+ unpacking the kernel is
3317 [1] The old nscd implementation is not available anymore since it was
3318 distributed with glibc up to version 2.0.98 and thus is now replaced.
3320 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3322 * Mark Kettenis provided a Hesiod NSS module.
3324 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3325 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3327 * rcmd can now handle netgroups (Dick Streefland).
3329 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3332 * Support for the IPv6 protocol has been added to the socket API, as per the
3333 latest draft standards.
3335 * Support for Linux 2.2 has been added.
3337 * Interface changes relative to the latest 2.0.x release:
3338 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3339 addseverity NEW: Unix98
3340 alphasort64 NEW: LFS
3341 argp_err_exit_status NEW: argp, GNU ext
3342 argp_error NEW: argp, GNU ext
3343 argp_failure NEW: argp, GNU ext
3344 argp_help NEW: argp, GNU ext
3345 argp_parse NEW: argp, GNU ext
3346 argp_program_bug_address NEW: argp, GNU ext
3347 argp_program_version NEW: argp, GNU ext
3348 argp_program_version_hook NEW: argp, GNU ext
3349 argp_state_help NEW: argp, GNU ext
3350 argp_usage NEW: argp, GNU ext
3351 authdes_create NEW: Secure RPC
3352 authdes_getucred NEW: Secure RPC
3353 authdes_pk_create NEW: Secure RPC
3354 backtrace NEW: GNU ext.
3355 backtrace_symbols NEW: GNU ext.
3356 backtrace_symbols_fd NEW: GNU ext.
3358 cacosf NEW: ISO C 9x
3359 cacosh NEW: ISO C 9x
3360 cacoshf NEW: ISO C 9x
3361 cacoshl NEW: ISO C 9x
3362 cacosl NEW: ISO C 9x
3369 casinf NEW: ISO C 9x
3370 casinh NEW: ISO C 9x
3371 casinhf NEW: ISO C 9x
3372 casinhl NEW: ISO C 9x
3373 casinl NEW: ISO C 9x
3375 catanf NEW: ISO C 9x
3376 catanh NEW: ISO C 9x
3377 catanhf NEW: ISO C 9x
3378 catanhl NEW: ISO C 9x
3379 catanl NEW: ISO C 9x
3380 cbc_crypt NEW: Secure RPC
3384 ccoshf NEW: ISO C 9x
3385 ccoshl NEW: ISO C 9x
3391 cimagf NEW: ISO C 9x
3392 cimagl NEW: ISO C 9x
3393 clearerr_locked REMOVED
3394 clntunix_create NEW: sunrpc ext
3396 clog10 NEW: ISO C 9x
3397 clog10f NEW: ISO C 9x
3398 clog10l NEW: ISO C 9x
3408 cprojf NEW: ISO C 9x
3409 cprojl NEW: ISO C 9x
3411 crealf NEW: ISO C 9x
3412 creall NEW: ISO C 9x
3417 csinhf NEW: ISO C 9x
3418 csinhl NEW: ISO C 9x
3421 csqrtf NEW: ISO C 9x
3422 csqrtl NEW: ISO C 9x
3426 ctanhf NEW: ISO C 9x
3427 ctanhl NEW: ISO C 9x
3429 des_setparity NEW: Secure RPC
3430 ecb_crypt NEW: Secure RPC
3431 endutxent NEW: Unix98
3433 exp10f NEW: ISO C 9x
3434 exp10l NEW: ISO C 9x
3438 fattach NEW: STREAMS
3439 fdetach NEW: STREAMS
3443 feclearexcept NEW: ISO C 9x
3444 fegetenv NEW: ISO C 9x
3445 fegetexceptflag NEW: ISO C 9x
3446 fegetround NEW: ISO C 9x
3447 feholdexcept NEW: ISO C 9x
3449 feraiseexcept NEW: ISO C 9x
3450 ferror_locked REMOVED
3451 fesetenv NEW: ISO C 9x
3452 fesetexceptflag NEW: ISO C 9x
3453 fesetround NEW: ISO C 9x
3454 fetestexcept NEW: ISO C 9x
3455 feupdateenv NEW: ISO C 9x
3456 fflush_locked REMOVED
3460 fgets_unlocked NEW: GNU ext.
3461 fileno_locked REMOVED
3473 fputc_locked REMOVED
3474 fputs_unlocked NEW: GNU ext.
3475 fread_unlocked NEW: GNU ext.
3480 fstatvfs NEW: Unix98
3484 ftruncate64 NEW: LFS
3486 fwrite_unlocked NEW: GNU ext.
3487 gai_strerror NEW: IPv6
3491 getchar_locked REMOVED
3493 getdate_err NEW: Unix98
3494 getdate_r NEW: GNU ext.
3496 getnameinfo NEW: IPv6
3497 getnetname NEW: Secure RPC
3498 getpmsg NEW: STREAMS
3499 getpt NEW: Unix98 PTY
3500 getrlimit64 NEW: LFS
3501 getutxent NEW: Unix98
3502 getutxid NEW: Unix98
3503 getutxline NEW: Unix98
3504 glob64 NEW: GNU ext.
3505 globfree64 NEW: GNU ext.
3506 gnu_get_libc_release NEW: GNU ext.
3507 gnu_get_libc_version NEW: GNU ext.
3508 grantpt NEW: Unix98 PTY
3509 host2netname NEW: Secure RPC
3511 iconv_close NEW: iconv
3512 iconv_open NEW: iconv
3513 if_freenameindex NEW: IPv6
3514 if_indextoname NEW: IPv6
3515 if_nameindex NEW: IPv6
3516 if_nametoindex NEW: IPv6
3517 in6addr_any NEW: IPv6
3518 in6addr_loopback NEW: IPv6
3519 inet6_isipv4mapped NEW: IPv6
3520 isastream NEW: STREAMS
3521 iswblank NEW: GNU ext.
3522 key_decryptsession NEW: Secure RPC
3523 key_decryptsession_pk NEW: Secure RPC
3524 key_encryptsession NEW: Secure RPC
3525 key_encryptsession_pk NEW: Secure RPC
3526 key_gendes NEW: Secure RPC
3527 key_get_conv NEW: Secure RPC
3528 key_secretkey_is_set NEW: Secure RPC
3529 key_setnet NEW: Secure RPC
3530 key_setsecret NEW: Secure RPC
3531 llrint NEW: ISO C 9x
3532 llrintf NEW: ISO C 9x
3533 llrintl NEW: ISO C 9x
3534 llround NEW: ISO C 9x
3535 llroundf NEW: ISO C 9x
3536 llroundl NEW: ISO C 9x
3541 lrintf NEW: ISO C 9x
3542 lrintl NEW: ISO C 9x
3543 lround NEW: ISO C 9x
3544 lroundf NEW: ISO C 9x
3545 lroundl NEW: ISO C 9x
3547 makecontext NEW: Unix98
3548 mempcpy NEW: GNU ext.
3551 modify_ldt NEW: kernel
3555 nearbyint NEW: ISO C 9x
3556 nearbyintf NEW: ISO C 9x
3557 nearbyintl NEW: ISO C 9x
3558 netname2host NEW: Secure RPC
3559 netname2user NEW: Secure RPC
3560 nexttoward NEW: ISO C 9x
3561 nexttowardf NEW: ISO C 9x
3562 nexttowardl NEW: ISO C 9x
3566 passwd2des NEW: Secure RPC
3568 pow10f NEW: GNU ext.
3569 pow10l NEW: GNU ext.
3572 printf_size NEW: GNU ext.
3573 printf_size_info NEW: GNU ext.
3574 profil_counter REMOVED
3575 pthread_mutexattr_getkind_np REPLACED
3576 pthread_mutexattr_setkind_np REPLACED
3577 ptsname NEW: Unix98 PTY
3578 ptsname_r NEW: Unix98 PTY
3580 putchar_locked REMOVED
3581 putgrent NEW: GNU ext.
3583 putpmsg NEW: STREAMS
3584 pututxline NEW: Unix98
3588 readdir64_r NEW: LFS
3589 remquo NEW: ISO C 9x
3590 remquof NEW: ISO C 9x
3591 remquol NEW: ISO C 9x
3593 roundf NEW: ISO C 9x
3594 roundl NEW: ISO C 9x
3596 scalbln NEW: ISO C 9x
3597 scalblnf NEW: ISO C 9x
3598 scalblnl NEW: ISO C 9x
3600 sendfile NEW: kernel
3601 setcontext NEW: Unix98
3602 setrlimit64 NEW: LFS
3603 setutxent NEW: Unix98
3605 sigignore NEW: Unix98
3606 sigqueue NEW: POSIX.1b
3607 sigrelse NEW: Unix98
3608 sigset NEW: POSIX.1b
3609 sigtimedwait NEW: POSIX.1b
3610 sigwaitinfo NEW: POSIX.1b
3611 sincos NEW: GNU ext.
3612 sincosf NEW: GNU ext.
3613 sincosl NEW: GNU ext.
3617 strcasestr NEW: GNU ext.
3618 strtoimax NEW: ISO C 9x
3619 strtoumax NEW: ISO C 9x
3620 strverscmp NEW: GNU ext.
3621 svcauthdes_stats NEW: Secure RPC
3622 svcunix_create NEW: sunrpc etx
3623 svcunixfd_create NEW: sunrpc ext
3624 swapcontext NEW: Unix98
3625 tcgetsid NEW: Unix98 PTY
3626 tdestroy NEW: GNU ext.
3627 tgamma NEW: ISO C 9x
3628 tgammaf NEW: ISO C 9x
3629 tgammal NEW: ISO C 9x
3633 truncf NEW: ISO C 9x
3634 truncl NEW: ISO C 9x
3636 unlockpt NEW: Unix98 PTY
3637 updwtmpx NEW: Unix98
3638 user2netname NEW: Secure RPC
3639 utmpxname NEW: Unix98
3640 versionsort NEW: GNU ext.
3641 versionsort64 NEW: GNU ext.
3643 wcscasecmp NEW: GNU ext.
3644 wcsncasecmp NEW: GNU ext.
3645 wcsnlen NEW: GNU ext.
3646 wcstoimax NEW: ISO C 9x
3647 wcstoll NEW: ISO C 9x
3648 wcstoull NEW: ISO C 9x
3649 wcstoumax NEW: ISO C 9x
3651 wordexp NEW: POSIX.2
3652 wordfree NEW: POSIX.2
3653 write_profiling REMOVED
3654 xdecrypt NEW: Secure RPC
3655 xdr_authdes_cred NEW: Secure RPC
3656 xdr_authdes_verf NEW: Secure RPC
3657 xdr_cryptkeyarg NEW: Secure RPC
3658 xdr_cryptkeyarg2 NEW: Secure RPC
3659 xdr_cryptkeyres NEW: Secure RPC
3660 xdr_getcredres NEW: Secure RPC
3661 xdr_key_netstarg NEW: Secure RPC
3662 xdr_key_netstres NEW: Secure RPC
3663 xdr_keybuf NEW: Secure RPC
3664 xdr_keystatus NEW: Secure RPC
3665 xdr_netnamestr NEW: Secure RPC
3666 xdr_sizeof NEW: Secure RPC
3667 xdr_unixcred NEW: sunrpc ext
3668 xencrypt NEW: Secure RPC
3669 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3680 * inet_ntoa is thread-safe
3682 * updwtmp is moved from libutil to libc
3684 * rewrite of cbrt function
3686 * update of timezone data
3700 * add atoll function
3702 * fix complex problems in Berkeley DB code
3704 * fix math functions
3708 * fixed lots of header problems (especially Linux/GNU specific)
3710 * dynamic loader preserves all registers
3712 * Roland McGrath provided support for handling of auxiliary objects in
3713 the ELF dynamic loader.
3715 * support for parallel builds is improved
3719 * GNU extensions are no longer declared by default. To enable them you
3720 must define the macro `_GNU_SOURCE' in your program or compile with
3723 * The library has changed from using GNU ld symbol aliases to using weak
3724 symbols where available. The ELF object file format supports weak
3725 symbols; GNU ld also supports weak symbols in the a.out format. (There
3726 is also now support for other GNU ld extensions in ELF. Use the
3727 `--with-elf' option to configure to indicate you have ELF, and
3728 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3729 of many files which contained only symbol aliases, reducing the size of
3730 the source and the compiled library; many other files were renamed to
3731 less cryptic names previously occupied by the symbol alias files.
3732 There is a new header file <elf.h> for programs which operate on
3733 files in the ELF format.
3735 * Converted to Autoconf version 2, so `configure' has more options.
3736 Run `configure --help' to see the details.
3738 * The library can now be configured to build profiling, highly-optimized
3739 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3740 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3741 `configure' enable building these extra libraries. The shared library is
3742 built by default when using both ELF and GNU ld. When shared libraries
3743 are enabled, the new library `-ldl' is available for arbitrary run-time
3744 loading of shared objects; its interface is defined in <dlfcn.h>. The
3745 new header file <link.h> gives access to the internals of the run-time
3746 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3747 application of same name on other systems and it provides information
3748 about dynamically linked binaries.
3750 * The C library now provides the run-time support code for profiling
3751 executables compiled with `-pg'. Programs can control the profiling code
3752 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3753 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3754 the support for this file format was contributed by David Mosberger-Tang.
3756 * The math code has been replaced with a math library based on fdlibm from
3757 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3758 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3759 functions. The math functions now reside in a separate library, so
3760 programs using them will need to use `-lm' their linking commands.
3762 * John C. Bowman contributed optimized ix87 assembler inline functions.
3764 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3765 mechanism similar to that found in Solaris 2. This is now used for the
3766 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3767 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3768 file controls what services are used for each individual database. This
3769 works by loading shared libraries with names specified in `nsswitch.conf',
3770 so service modules can be changed or added at any time without even
3771 relinking any program. Currently there are the file, db, and NIS based
3772 NSS services available.
3774 * The new functions `strtoq' and `strtouq' parse integer values from
3775 strings, like `strtol' and `strtoul', but they return `long long int' and
3776 `unsigned long long int' values, respectively (64-bit quantities).
3778 * The new functions `strtof' and `strtold' parse floating-point values from
3779 strings, like `strtod', but they return `float' and `long double' values,
3780 respectively (on some machines `double' and `long double' are the same).
3782 * Ulrich Drepper has contributed new implementations of the floating-point
3783 printing and reading code used in the `printf' family of functions and
3784 `strtod', `strtof', and `strtold'. These new functions are perfectly
3785 accurate, and much faster than the old ones.
3787 * The implementation of the POSIX locale model was completely rewritten by
3788 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3789 to compile the POSIX locale definition.
3791 * The former dummy implementations of the strcoll and strxfrm function are
3792 now replaced by fully functional code contributed by Ulrich Drepper. The
3793 collation information comes from the POSIX locale definitions.
3795 * The new header <langinfo.h> defines an interface for accessing
3796 various locale-dependent data (using the locale chosen with `setlocale').
3798 * Ulrich Drepper has contributed a new suite of functions for operation on
3799 wide-character and multibyte-character strings, in <wchar.h>;
3800 and classification and case conversion of wide characters, in <wctype.h>.
3801 These new functions are conforming to the ISO C, Amendement 1 specification.
3803 * There is now a second implementation of the standard I/O library available.
3804 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3805 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3807 * You can now use positional parameter specifications in format strings
3808 for the `printf' and `scanf' families of functions. For example,
3809 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3810 ``Number 6, Mr Jones''. This is mainly useful when providing different
3811 format strings for different languages, whose grammars may dictate
3812 different orderings of the values being printed. To support this
3813 feature, the interface for `register_printf_handler' has changed; see
3814 the header file <printf.h> for details.
3816 * The `printf' and `scanf' families of functions now understand a new
3817 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3818 to group numbers as indicated by the locale; for `scanf' and friends, this
3819 says to accept as valid only a number with all the proper grouping
3820 separators in the right places. In the default "C" locale, numbers are
3821 not grouped; but locales for specific countries will define the usual
3822 conventions (i.e. separate thousands with `,' in the US locale).
3824 * The pgrp functions have been regularized, slightly incompatibly but much
3825 less confusingly. The core functions are now `getpgid' and `setpgid',
3826 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3827 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3828 provided for compatibility. There is no longer an incompatible `getpgrp'
3829 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3831 * The new header file <fts.h> and suite of functions simplify programs that
3832 operate on directory trees. This code comes from 4.4 BSD.
3834 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3835 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3836 NSS scheme used in glibc.
3838 * There is a new malloc debugging hook `__memalign_hook'.
3840 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3841 `unsigned int' in <sys/types.h>. These are for compatibility only and
3842 their use is discouraged.
3844 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3845 done differently, so that it works even without GNU ld.
3847 * New function `euidaccess' checks allowed access to a file like `access',
3848 but using the effective IDs instead of the real IDs.
3850 * The time zone data files have been updated for the latest and greatest
3851 local time conventions of the countries of the world.
3853 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3856 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3857 for formatting floating-point numbers. They are provided only for
3858 compatibility; new programs should use `sprintf' instead. There are
3859 also equivalent function for the `long double' floating-point type and
3860 all functions also exist in a reentrant form.
3862 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3863 functions for maintaining the login-record files (primarily of use to
3864 system programs such as `login'), and convenient functions for
3865 allocating and initializing a pseudo-terminal (pty) device.
3867 * Ulrich Drepper has contributed new support for System V style
3868 shared memory and IPC on systems that support it.
3870 * Ulrich Drepper has contributed several miscellaneous new functions found
3871 in System V: The `hsearch' family of functions provide an effective
3872 implementation of hash tables; `a64l' and `l64a' provide a very simple
3873 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3876 * Ulrich Drepper has contributed new reentrant counterparts for the
3877 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3879 * Ulrich Drepper has contributed new, highly-optimized versions of several
3880 string functions for the i486/Pentium family of processors.
3882 * Ulrich Drepper has updated the Linux-specific code, based largely
3883 on work done in Hongjiu Lu's version of GNU libc for Linux.
3884 The GNU library now supports Linux versions 2.0.10 and later,
3885 using the ELF object file format (i[3456]86-*-linux).
3887 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3889 * David Mosberger-Tang and Richard Henderson have ported the C library
3890 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3891 dynamic linking support for ELF/Alpha.
3893 * Richard Henderson contributed several Alpha optimized assembler function
3894 for arithmetic and string handling.
3896 * Ulrich Drepper has contributed a new set of message catalog functions to
3897 support multiple languages using the <libintl.h> interface, for use with
3898 his new package GNU gettext. Translation volunteers have contributed
3899 catalogs of the library's messages in Spanish, German, and Korean.
3901 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3902 program and the `catgets' function for reading the catalog files it
3903 creates. (The <libintl.h> interface is preferred; we include the
3904 <nl_types.h> interface using `catgets' only for source compatibility with
3905 programs already written to use it.)
3907 * New header file <values.h> gives SVID-compatible names for <limits.h>
3910 * Various new macros, declarations, and small header files for compatibility
3913 * New function `group_member' is a convenient way to check if a process has
3914 a given effective group ID.
3916 * When using GCC 2.7 and later, the socket functions are now declared in a
3917 special way so that passing an argument of type `struct sockaddr_in *',
3918 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3919 `struct sockaddr *' type, does not generate a type-clash warning.
3921 * New function `error' declared in header file <error.h> is a convenient
3922 function for printing error messages and optionally exiting; this is the
3923 canonical function used in GNU programs. The new functions `err', `warn',
3924 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3925 doing the same thing.
3927 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3928 POSIX.2 `glob' function to do ~ and {...} expansion.
3930 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3931 `clearenv' which is used in POSIX.9 is also available.
3933 * New function `getsid' returns session ID number on systems that support it.
3935 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3936 files <db.h> and <mpool.h> provide a rich set of functions for several
3937 types of simple databases stored in memory and in files, and <ndbm.h> is
3938 an old `ndbm'-compatible interface using the `db' functions. Link with
3939 `-ldb' to get these functions.
3941 * New macro `strdupa' copies a string like `strdup', but uses local stack
3942 space from `alloca' instead of dynamic heap space from `malloc'.
3944 * New function `strnlen' is like `strlen' but searches only a given maximum
3945 number of characters for the null terminator. `stpncpy', `strndup' and
3946 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3949 * New function `statfs' in header <sys/statfs.h>.
3951 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3952 provide convenient functions for operating on blocks of null-terminated
3955 * A new suite of functions in <utmp.h> handle all the details of reading
3956 and writing the utmp file.
3958 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3961 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3962 completely POSIX compliant and also implemented the extended functionality
3963 to handle alternate digit representation and alternate era date formats.
3965 * Ulrich Drepper provided an implementation of the `strptime' function
3966 defined in XPG4.2 which transforms a string into a `struct tm' value.
3968 * Paul Eggert provided the tzselect shell script as part of the timezone
3969 code. The shell script makes it easy to select the correct timezone
3972 * The implementation of the malloc family of functions is completely replaced
3973 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3974 The implementation uses the mmap function (if available) and it is
3975 optimized for the use in multi threaded programs.
3977 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3978 functions. This new functionality is usable by specifying a special
3979 salt string and it is compatible with implementation on *BSD systems.
3981 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3982 `getsubopt' to handle second level command line options, `bsd_signal'
3983 to access BSD style `signal' functionality, the obsolete `regexp' style
3986 * the `lchown' function is available on system which support this
3989 * The implementation of the shadow password handling function was contributed
3992 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3994 * POSIX.1g support was added. The <sys/select.h> header is available,
3995 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3996 implementation of `getaddrinfo'.
4000 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4002 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4003 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4005 * The new function `daemon' from 4.4 BSD is useful for server programs that
4006 want to put themselves in the background.
4008 * Joel Sherrill has contributed support for several standalone boards that
4009 run without an operating system.
4011 * `printf', `scanf' and friends now accept a `q' type modifier for long
4012 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4014 * All of the code taken from BSD (notably most of the math and networking
4015 routines) has been updated from the BSD 4.4-Lite release.
4017 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4019 * The new functions `getdomainname' and `setdomainname' fetch or change the
4020 YP/NIS domain name. These are system calls which exist on systems which
4023 * The time zone data files have been updated for the latest international
4026 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4027 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4031 * The C library now includes support for Sun RPC, from Sun's free
4032 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4033 are included. (There is still no support for YP.)
4035 * Tom Quinn has contributed a port of the C library to SGI machines running
4036 Irix 4 (mips-sgi-irix4).
4038 * The new `lockf' function is a simplified interface to the locking
4039 facilities of `fcntl', included for compatibility.
4041 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4043 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4045 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4048 * The error message printed by `assert' for a failed assertion now includes
4049 the name of the program (if using GNU ld) and the name of the calling
4050 function (with versions of GCC that support this).
4052 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4054 * The library now includes the <sys/mman.h> header file and memory
4055 management functions `mmap', `munmap', `mprotect', `msync', and
4056 `madvise', on systems that support those facilities.
4058 * The interface for `mcheck' has changed slightly: the function called to
4059 abort the program when an allocation inconsistency is detected now takes
4060 an argument that indicates the type of failure. The new function
4061 `mprobe' lets you request a consistency check for a particular block at
4062 any time (checks are normally done only when you call `free' or `realloc'
4065 * It is now possible to easily cross-compile the C library, building on one
4066 system a library to run on another machine and/or operating system. All
4067 you need to do is set the variable `HOST_CC' in `configparms' to the
4068 native compiler for programs to run on the machine you are building on (a
4069 few generator programs are used on Unix systems); set `CC' to the
4072 * The new function `fexecve' (only implemented on the GNU system) executes
4073 a program file given a file descriptor already open on the file.
4077 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4078 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4080 * You can set the variable `libprefix' in `configparms' to specify a prefix
4081 to be prepended to installed library files; this makes it easy to install
4082 the GNU C library to be linked as `-lgnuc' or whatever.
4084 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4085 copies a limited number of characters from a string, and returns the
4086 address of the last character written.
4088 * You no longer need to check for whether the installed `stddef.h' is
4089 compatible with the GNU C library. configure now checks for you.
4091 * You can now define a per-stream `fileno' function to convert the
4092 stream's cookie into an integral file descriptor.
4094 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4095 allocates zero bytes of storage, and returns a unique pointer which
4096 you can pass to `realloc' or `free'. The behavior is undefined if
4097 you dereference this pointer.
4099 * The C library now runs on Sony NEWS m68k machines running either
4100 NewsOS 3 or NewsOS 4.
4102 * The new `syscall' function is a system-dependent primitive function
4103 for invoking system calls. It has the canonical behavior on Unix
4104 systems, including unreliable return values for some calls (such as
4105 `pipe', `fork' and `getppid').
4107 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4108 to `EAGAIN', which is the preferred name. On systems whose kernels
4109 use two distinct codes, the C library now translates EWOULDBLOCK to
4110 EAGAIN in every system call function.
4114 * The GNU C Library Reference Manual is now distributed with the library.
4115 `make dvi' will produce a DVI file of the printed manual.
4116 `make info' will produce Info files that you can read on line using C-h i
4117 in Emacs or the `info' program.
4118 Please send comments on the manual to bug-glibc-manual@gnu.org.
4120 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4122 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4124 * Jason Merrill has contributed a port to the Sequent Symmetry running
4125 Dynix version 3 (i386-sequent-dynix).
4127 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4128 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4130 * New function `memory_warnings' lets you arrange to get warnings when
4131 malloc is running out of memory to allocate, like Emacs gives you.
4133 * The C library now contains the relocating allocator used in Emacs 19 for
4134 its editing buffers. This allocator (ralloc) minimizes allocation
4135 overhead and fragmentation by moving allocated regions around whenever it
4136 needs to. You always refer to a ralloc'd region with a "handle" (a
4137 pointer to a pointer--an object of type `void **').
4139 * There is a new `printf' format: `%m' gives you the string corresponding
4140 to the error code in `errno'.
4142 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4143 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4144 pass, the `a' modifier says to fill in a `char **' you pass with a
4147 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4148 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4149 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4151 * `mkstemp' is a traditional Unix function to atomically create and open a
4152 uniquely-named temporary file.
4156 * The standard location for the file that says what the local timezone is
4157 has changed again. It is now `/usr/local/etc/localtime' (or more
4158 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4160 * The distribution no longer contains any files with names longer than 14
4163 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4164 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4166 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4170 * The configuration process has changed quite a bit. The `configure'
4171 script is now used just like the configuration scripts for other GNU
4172 packages. The `sysdeps' directory hierarchy is much rearranged.
4173 The file `INSTALL' explains the new scheme in detail.
4175 * The header files no longer need to be processed into ANSI C and
4176 traditional C versions. There is just one set of files to install, and
4177 it will work with ANSI or old C compilers (including `gcc -traditional').
4179 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4180 MIPS DECStation running Ultrix 4.
4182 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4183 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4184 made itself into a shared library.
4186 * Yet further improved support for the i386, running 4.3 BSD-like systems
4187 (such as Mach 3 with the Unix single-server), or System V.
4189 * New function `strncasecmp' to do case-insensitive string comparison
4190 with limited length.
4192 * New function `strsep' is a reentrant alternative to `strtok'.
4194 * New functions `scandir' and `alphasort' for searching directories.
4196 * New function `setenv' is a better interface to `putenv'.
4198 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4199 function for traversing a directory tree.
4201 * The GNU obstack package is now also part of the C library.
4202 The new function `open_obstack_stream' creates a stdio stream that
4203 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4204 formatted output directly to an obstack.
4206 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4207 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4209 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4211 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4212 things to your strings.
4214 * There are some new test programs: test-fseek, testmb, and testrand.
4216 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4217 These ports are not finished, but are a good starting place for really
4218 supporting those systems.
4220 * `/etc/localtime' is now the standard location for the file that says what
4221 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4222 This follows the general principle that `/etc' is the place for all local
4223 configuration files.
4225 * The C library header files now use `extern "C"' when used by the C++
4226 compiler, so the C library should now work with C++ code.
4228 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4229 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4232 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4233 function which uses the `merge sort' algorithm, and is said to be
4234 significantly faster than the old GNU `qsort' function. Merge sort is
4235 now the standard `qsort' function. The new algorithm can require a lot
4236 of temporary storage; so, the old sorting function is called when the
4237 required storage is not available.
4239 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4240 provides the Unix `crypt' function, plus some other entry points.
4242 * The code and header files taken from 4.4 BSD have been updated with the
4243 latest files released from Berkeley.
4245 ----------------------------------------------------------------------
4246 Copyright information:
4248 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4250 Permission is granted to anyone to make or distribute verbatim copies
4251 of this document as received, in any medium, provided that the
4252 copyright notice and this permission notice are preserved,
4253 thus giving the recipient permission to redistribute in turn.
4255 Permission is granted to distribute modified versions
4256 of this document, or of portions of it,
4257 under the above conditions, provided also that they
4258 carry prominent notices stating who last changed them.
4261 version-control: never