1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * 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 * glibc now implements the memfd_create function on Linux.
40 Deprecated and removed features, and other changes affecting compatibility:
42 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
43 defined by <sys/ptrace.h>.
45 * libm no longer supports SVID error handling (calling a user-provided
46 matherr function on error) or the _LIB_VERSION variable to control error
47 handling. (SVID error handling and the _LIB_VERSION variable still work
48 for binaries linked against older versions of the GNU C Library.) The
49 libieee.a library is no longer provided. math.h no longer defines struct
50 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
51 TLOSS, PLOSS and HUGE.
53 * The libm functions pow10, pow10f and pow10l are no longer supported for
54 new programs. Programs should use the standard names exp10, exp10f and
55 exp10l for these functions instead.
57 * The mcontext_t type is no longer the same as struct sigcontext. On
58 platforms where it was previously the same, this changes the C++ name
59 mangling for interfaces involving this type.
61 * The add-ons mechanism for building additional packages at the same time as
62 glibc has been removed. The --enable-add-ons configure option is now
65 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
66 the specified string can be parsed as a domain name.
68 * In the malloc_info output, the <heap> element may contain another <aspace>
69 element, "subheaps", which contains the number of sub-heaps.
71 * The libresolv function p_secstodate is no longer supported for new
74 Changes to build and runtime requirements:
76 [Add changes to build and runtime requirements here]
78 Security related changes:
80 CVE-2009-5064: The ldd script would sometimes run the program under
81 examination directly, without preventing code execution through the
82 dynamic linker. (The glibc project disputes that this is a security
83 vulnerability; only trusted binaries must be examined using the ldd
86 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
87 suffered from a one-byte overflow during ~ operator processing (either
88 on the stack or the heap, depending on the length of the user name).
89 Reported by Tim Rühsen.
91 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
92 would sometimes fail to free memory allocated during ~ operator
93 processing, leading to a memory leak and, potentially, to a denial
96 The glob function, when invoked with GLOB_TILDE and without
97 GLOB_NOESCAPE, could write past the end of a buffer while
98 unescaping user names. Reported by Tim Rühsen.
100 The following bugs are resolved with this release:
102 [The release manager will add the list generated by
103 scripts/list-fixed-bugs.py just before the release.]
110 * A per-thread cache has been added to malloc. Access to the cache requires
111 no locks and therefore significantly accelerates the fast path to allocate
112 and free small amounts of memory. Refilling an empty cache requires locking
113 the underlying arena. Performance measurements show significant gains in a
114 wide variety of user workloads. Workloads were captured using a special
115 instrumented malloc and analyzed with a malloc simulator. Contributed by
116 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
118 * Unicode 10.0.0 Support: Character encoding, character type info, and
119 transliteration tables are all updated to Unicode 10.0.0, using
120 generator scripts contributed by Mike FABIAN (Red Hat).
121 These updates cause user visible changes, especially the changes in
122 wcwidth for many emoji characters cause problems when emoji sequences
123 are rendered with pango, see for example:
124 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
126 * Collation of Hungarian has been overhauled and is now consistent with "The
127 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
130 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
132 - The GNU C Library will now detect when /etc/resolv.conf has been
133 modified and reload the changed configuration. The new resolver option
134 “no-reload” (RES_NORELOAD) disables this behavior.
136 - The GNU C Library now supports an arbitrary number of search domains
137 (configured using the “search” directive in /etc/resolv.conf);
138 previously, there was a hard limit of six domains. For backward
139 compatibility, applications that directly modify the ‘_res’ global
140 object are still limited to six search domains.
142 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
143 Library will now randomly pick a name server from the configuration as a
144 starting point. (Previously, the second name server was always used.)
146 * The tunables feature is now enabled by default. This allows users to tweak
147 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
149 * New function reallocarray, which resizes an allocated block (like realloc)
150 to the product of two sizes, with a guaranteed clean failure upon integer
151 overflow in the multiplication. Originally from OpenBSD, contributed by
152 Dennis Wölfing and Rüdiger Sonderfeld.
154 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
155 These are extended versions of preadv and pwritev, respectively, taking an
156 additional flags argument. The set of supported flags depends on the
157 running kernel; full support currently requires kernel 4.7 or later.
159 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
160 create a new session ID for the spawned process. This feature is
161 scheduled to be added to the next major revision of POSIX; for the time
162 being, it is available under _GNU_SOURCE.
164 * errno.h is now safe to use from C-preprocessed assembly language on all
165 supported operating systems. In this context, it will only define the
166 Exxxx constants, as preprocessor macros expanding to integer literals.
168 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
169 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
170 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
171 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
173 To compile programs that use this feature, the compiler must support
174 128-bit floating point with the type name _Float128 (as defined by TS
175 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
176 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
177 must be defined to make the new interfaces visible.
179 The new functions and macros correspond to those present for other
180 floating-point types (except for a few obsolescent interfaces not
181 supported for the new type), with F128 or f128 suffixes; for example,
182 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
183 printf or scanf formats for the new type; the strfromf128 and strtof128
184 interfaces should be used instead.
186 Deprecated and removed features, and other changes affecting compatibility:
188 * The synchronization that pthread_spin_unlock performs has been changed to
189 now be equivalent to a C11 atomic store with release memory order to the
190 spin lock's memory location. Previously, several (but not all)
191 architectures used stronger synchronization (e.g., containing what is
192 often called a full barrier). This change can improve performance, but
193 may affect odd fringe uses of spin locks that depend on the previous
194 behavior (e.g., using spin locks as atomic variables to try to implement
195 Dekker's mutual exclusion algorithm).
197 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
200 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
201 will only be built and installed when the GNU C Library is configured with
202 --enable-obsolete-rpc. This allows alternative RPC implementations, such
203 as TIRPC or rpcsvc-proto, to be used.
205 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
206 libnss_compat, are deprecated, and will not be built or installed by
209 The NIS(+) support library, libnsl, is also deprecated. By default, a
210 compatibility shared library will be built and installed, but not headers
211 or development libraries. Only a few NIS-related programs require this
212 library. (In particular, the GNU C Library has never required programs
213 that use 'gethostbyname' to be linked with libnsl.)
215 Replacement implementations based on TIRPC, which additionally support
216 IPv6, are available from <https://github.com/thkukuk/>. The configure
217 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
218 name service modules, to be built and installed.
220 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
221 support is enabled, the configured recursive resolver must support EDNS.
222 (Responding to EDNS-enabled queries with responses which are not
223 EDNS-enabled is fine, but FORMERR responses are not.)
225 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
226 servers have not supported this opcode for a long time.
228 * The _res_opcodes variable has been removed from libresolv. It had been
229 exported by accident.
231 * <string.h> no longer includes inline versions of any string functions,
232 as this kind of optimization is better done by the compiler. The macros
233 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
235 * The nonstandard header <xlocale.h> has been removed. Most programs should
236 use <locale.h> instead. If you have a specific need for the definition of
237 locale_t with no other declarations, please contact
238 libc-alpha@sourceware.org and explain.
240 * The obsolete header <sys/ultrasound.h> has been removed.
242 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
244 * The obsolete function cfree has been removed. Applications should use
247 * The stack_t type no longer has the name struct sigaltstack. This changes
248 the C++ name mangling for interfaces involving this type.
250 * The ucontext_t type no longer has the name struct ucontext. This changes
251 the C++ name mangling for interfaces involving this type.
253 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
254 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
255 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
256 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
257 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
258 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
259 fpregset_t type no longer has the name struct fpu. This changes the C++
260 name mangling for interfaces involving those types.
262 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
263 synced with the kernel:
265 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
266 are not supported on this architecture and have been removed.
268 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
269 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
270 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
272 Programs that assume the GET/SETREGS ptrace requests are universally
273 available will now fail to build, instead of malfunctioning at runtime.
275 Changes to build and runtime requirements:
277 * Linux kernel 3.2 or later is required at runtime, on all architectures
278 supported by that kernel. (This is a change from version 2.25 only for
281 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
283 * On most architectures, GCC 4.9 or later is required to build the GNU C
284 Library. On powerpc64le, GCC 6.2 or later is required.
286 Older GCC versions and non-GNU compilers are still supported when
287 compiling programs that use the GNU C Library. (We do not know exactly
288 how old, and some GNU extensions to C may be _de facto_ required. If you
289 are interested in helping us make this statement less vague, please
290 contact libc-alpha@sourceware.org.)
292 Security related changes:
294 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
295 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
297 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
298 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
300 * Avoid printing a backtrace from the __stack_chk_fail function since it is
301 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
304 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
305 fixed (CVE-2017-12133).
307 The following bugs are resolved with this release:
309 [984] network: Respond to changed resolv.conf in gethostbyname
310 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
311 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
313 [12189] libc: __stack_chk_fail should not attempt a backtrace
315 [14096] time: Race condition on timezone/tst-timezone.out
316 [14172] localedata: az_IR: new locale
317 [14995] build: glibc fails to build if gold is the default linker, even if
319 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
320 [16637] network: inet_pton function is accepting IPv6 with bad format
321 [16640] string: string/strtok.c: undefined behaviour inconsistent between
322 x86 and other generic code
323 [16875] localedata: ko_KR: fix lang_name
324 [17225] localedata: ar_SY: localized month names for May and June are
326 [17297] localedata: da_DK: wrong date_fmt string
327 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
328 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
329 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
330 [19066] localedata: ar_SA abbreviated day and month names are in English
331 [19569] network: resolv: Support an arbitrary number of search domains
332 [19570] network: Implement random DNS server selection in the stub
334 [19838] locale: localedef fails on PA-RISC
335 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
336 order of 0D36 and 0D37
337 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
339 [20098] libc: FAIL: debug/backtrace-tst on hppa
340 [20257] network: sunrpc: clntudp_call does not enforce timeout when
342 [20275] localedata: locale day/abday/mon/abmon should not have trailing
344 [20313] localedata: Update locale data to Unicode 9.0
345 [20424] manual: Document how to provide a malloc replacement
346 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
348 [20686] locale: Add el_GR@euro to SUPPORTED.
349 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
350 failures consistently
351 [21015] dynamic-link: Document and fix --enable-bind-now
352 [21016] nptl: pthread_cond support is broken on hppa
353 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
354 frame-pointer on i386
355 [21049] libc: segfault in longjmp_chk() due to clobbered processor
357 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
358 [21088] libc: Build fails with --enable-static-nss
359 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
360 generic c code is used
361 [21109] libc: Tunables broken on big-endian
362 [21112] math: powf has large ulp errors with base close to 1 and exponent
364 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
366 [21120] malloc: glibc malloc is incompatible with GCC 7
367 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
369 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
370 (-0.0) when linking with -lieee
371 [21171] math: log10, log2 and lgamma return incorrect results
372 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
373 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
374 [21207] localedata: ce_RU: update weekdays from CLDR
375 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
376 [21217] localedata: Update months from CLDR-31
377 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
378 [21243] libc: support_delete_temp_file should issue warning for failed
380 [21244] libc: support resolv_test_start() socket fd close should be
382 [21253] libc: localedef randomly segfaults when using -fstack-check due to
383 new posix_spawn implementation
384 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
385 leads to lower CPU frequency
386 [21259] libc: [alpha] termios.h missing IXANY for POSIX
387 [21261] libc: [sparc64] bits/setjmp.h namespace
388 [21267] network: [mips] bits/socket.h IOC* namespace
389 [21268] libc: [alpha] termios.h NL2, NL3 namespace
390 [21270] libc: mmap64 silently truncates large offset values
391 [21275] libc: posix_spawn always crashes on ia64 now
392 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
393 [21280] math: [powerpc] logbl for POWER7 return incorrect results
394 [21289] libc: Incorrect declaration for 32-bit platforms with
395 _FILE_OFFSET_BITS=64 causes build error
396 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
397 not support gethostbyname4_r
398 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
400 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
402 [21340] libc: Support POSIX_SPAWN_SETSID
403 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
404 [21359] network: ns_name_pack needs additional byte in destination buffer
405 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
406 against fragmentation attacks (CVE-2017-12132)
407 [21369] network: resolv: Remove EDNS fallback
408 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
410 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
411 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
412 [21393] stdio: Missing dup3 error check in freopen, freopen64
413 [21396] libc: Use AVX2 memcpy/memset on Skylake server
414 [21399] localedata: Bad description for U00EC in
415 localedata/charmaps/CP1254
416 [21411] malloc: realloc documentation error
417 [21426] network: sys/socket.h uio.h namespace
418 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
419 [21445] libc: signal.h bsd_signal namespace
420 [21455] network: Network headers stdint.h namespace
421 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
423 [21475] network: resolv: Overlong search path is truncated mid-label
424 [21511] libc: sigstack namespace
425 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
426 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
427 bits/syscall.h: No such file or directory
428 [21517] libc: struct sigaltstack namespace
429 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
430 [21533] localedata: Update locale data to Unicode 10.0
432 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
433 junk at end of line, first unrecognized character is `@'
434 [21538] libc: SIG_HOLD missing for XPG4
435 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
436 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
437 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
438 [21543] libc: sigevent namespace
439 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
440 size for general purpose registers in mcontext_t structure
441 [21550] libc: sigwait namespace
442 [21552] libc: XPG4 bsd_signal namespace
443 [21554] libc: sigpause namespace
444 [21560] libc: sys/wait.h signal.h namespace
445 [21561] libc: waitid namespace
446 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
447 '/usr/include/stdlib.h': Permission denied
448 [21575] libc: sys/wait.h missing struct rusage definition
449 [21584] libc: sigaltstack etc namespace
450 [21597] libc: siginterrupt namespace
451 [21607] math: hppa: FAIL: math/test-tgmath
452 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
454 [21622] libc: [tile] missing SA_* for POSIX.1:2008
455 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
456 stack and heap (CVE-2017-1000366)
457 [21625] libc: wait3 namespace
458 [21654] nss: Incorrect pointer alignment in NSS group merge result
460 [21657] network: Parse interface zone id for node-local multicast
461 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
463 [21666] libc: .symver is used on common symbol
464 [21668] network: resolv: res_init cross-thread broadcast introduces race
466 [21687] math: tgmath.h totalorder, totalordermag return type
467 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
469 [21696] libc: Incorrect assumption of of __cpu_mask in
470 posix/sched_cpucount.c
471 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
472 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
474 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
475 leading to relocation crash
476 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
477 `resolv_conf_matches (resp, conf)' failed.
478 [21710] localedata: Added Samoan language locale for Samoa
479 [21711] localedata: Pashto yesstr/nostr locale are missing
480 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
481 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
482 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
483 [21723] localedata: yesstr/nostr missing for Chinese language locale
484 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
486 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
488 [21728] localedata: New Locale for Tongan language
489 [21729] localedata: incorrect LC_NAME fields for hi_IN
490 [21733] localedata: yesstr and nostr are missing for zh_HK
491 [21734] localedata: Missing yesstr and nostr are for kw_GB
492 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
493 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
494 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
495 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
497 [21744] libc: Tests failing on --enable-tunables --enable-stack-
499 [21749] localedata: Wrong abbreviated day name (“abday”) for
501 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
502 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
503 [21759] localedata: missing yesstr and nostr for Tigrinya
504 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
505 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
506 [21767] localedata: Missing Bislama locales
507 [21768] localedata: Missing yesstr and nostr for aa_ET
508 [21770] localedata: Missing Field in li_NL
509 [21778] nptl: Robust mutex may deadlock
510 [21779] libc: MicroBlaze segfaults when loading libpthread
511 [21783] localedata: Fix int_select international_call_prefixes
512 [21784] localedata: Inconsistency in country_isbn
513 [21788] localedata: Missing Country Postal Abbreviations
514 [21794] localedata: Added-country_isbn-for-Italy
515 [21795] localedata: Add/Fix country_isbn for France
516 [21796] localedata: Added country_isbn for Republic of Korea
517 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
518 [21799] localedata: Added int_select international_call_prefixes
519 [21801] localedata: Added int_select international_call_prefixes
520 [21804] nptl: Double semicolon in thread-shared-types.h
521 [21807] localedata: LC_ADDRESS fix for pap_CW
522 [21808] localedata: Fix LC_ADDRESS for pap_AW
523 [21821] localedata: Added country_name in mai_IN
524 [21822] localedata: Fix LC_TIME for mai_IN
525 [21823] localedata: missing yesstr, nostr for sa_IN
526 [21825] localedata: Fix name_mrs for mag_IN
527 [21828] localedata: 2.26 changelog should mention user visible changes
529 [21835] localedata: Added Maithili language locale for Nepal
530 [21838] localedata: Removed redundant data for the_NP
531 [21839] localedata: Fix LC_MONETARY for ta_LK
532 [21844] localedata: Fix Latin characters and Months Sequence.
533 [21848] localedata: Fix mai_NP Title Name
538 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
539 24731-2:2010, is supported to enable declarations of functions from that
540 TR. Note that not all functions from that TR are supported by the GNU C
543 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
544 18661-1:2014, is supported to enable declarations of functions and macros
545 from that TS. Note that not all features from that TS are supported by
548 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
549 18661-4:2015, is supported to enable declarations of functions and macros
550 from that TS. Note that most features from that TS are not supported by
553 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
554 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
555 Since the GNU C Library defaults to a much newer revision of POSIX, this
556 will only affect programs that specifically request an old conformance
557 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
558 see a change in the visible declarations, but a program compiled with
559 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
562 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
563 defined by all multithreaded code, but glibc has not required this for
566 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
567 means that in a future release, the macros “major”, “minor”, and “makedev”
568 will only be available from <sys/sysmacros.h>.
570 These macros are not part of POSIX nor XSI, and their names frequently
571 collide with user code; see for instance glibc bug 19239 and Red Hat bug
572 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
573 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
576 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
577 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
578 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
580 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
581 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
582 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
583 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
584 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
585 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
586 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
587 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
588 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
589 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
590 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
593 * New <math.h> features are added from TS 18661-1:2014:
595 - Signaling NaN macros: SNANF, SNAN, SNANL.
597 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
598 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
599 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
601 - llogb functions: the llogb, llogbf and llogbl functions, and the
602 FP_LLOGB0 and FP_LLOGBNAN macros.
604 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
607 - Comparison macros: iseqsig.
609 - Classification macros: iscanonical, issubnormal, iszero.
611 - Total order functions: totalorder, totalorderf, totalorderl,
612 totalordermag, totalordermagf, totalordermagl.
614 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
616 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
617 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
619 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
620 are added to libc. They convert a floating-point number into string.
622 * Most of glibc can now be built with the stack smashing protector enabled.
623 It is recommended to build glibc with --enable-stack-protector=strong.
624 Implemented by Nick Alcock (Oracle).
626 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
627 intended to be used instead of memset() to erase sensitive data after use;
628 the compiler will not optimize out calls to explicit_bzero even if they
629 are "unnecessary" (in the sense that no _correct_ program can observe the
630 effects of the memory clear).
632 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
633 to float instead of double. This does not affect the ABI of any libraries
634 that are part of the GNU C Library, but may affect the ABI of other
635 libraries that use this type in their interfaces.
637 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
638 float_t and double_t types are now defined to long double instead of float
639 and double. These options are not the default, and this does not affect
640 the ABI of any libraries that are part of the GNU C Library, but it may
641 affect the ABI of other libraries that use this type in their interfaces,
642 if they are compiled or used with those options.
644 * The getentropy and getrandom functions, and the <sys/random.h> header file
647 * The buffer size for byte-oriented stdio streams is now limited to 8192
648 bytes by default. Previously, on Linux, the default buffer size on most
649 file systems was 4096 bytes (and thus remains unchanged), except on
650 network file systems, where the buffer size was unpredictable and could be
651 as large as several megabytes.
653 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
654 for the Linux quota interface which predates kernel version 2.4.22 has
657 * The malloc_get_state and malloc_set_state functions have been removed.
658 Already-existing binaries that dynamically link to these functions will
659 get a hidden implementation in which malloc_get_state is a stub. As far
660 as we know, these functions are used only by GNU Emacs and this change
661 will not adversely affect already-built Emacs executables. Any undumped
662 Emacs executables, which normally exist only during an Emacs build, should
663 be rebuilt by re-running “./configure; make” in the Emacs build tree.
665 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
666 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
667 “no-ip6-dotint” had already been the default, and support for the
668 “ip6-dotint” option was removed from the Internet in 2006.
670 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
671 flag from <resolv.h> have been removed. The option relied on a
672 backwards-incompatible DNS extension which was never deployed on the
675 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
676 RES_BLAST defined in the <resolv.h> header file have been deprecated.
677 They were already unimplemented.
679 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
680 _res.flags are deprecated. The flag was standardized in RFC 2133, but
681 removed again from the IETF name lookup interface specification in RFC
682 2553. Applications should use getaddrinfo instead.
684 * DNSSEC-related declarations and definitions have been removed from the
685 <arpa/nameser.h> header file, and libresolv will no longer attempt to
686 decode the data part of DNSSEC record types. Previous versions of glibc
687 only implemented minimal support for the previous version of DNSSEC, which
688 is incompatible with the currently deployed version.
690 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
691 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
692 <arpa/nameser.h> header file because the distinction between RR types and
693 meta-RR types is not officially standardized, subject to revision, and
694 thus not suitable for encoding in a macro.
696 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
697 and rhook members of the res_state type in <resolv.h> have been removed.
698 The glibc stub resolver did not support these hooks, but the header file
699 did not reflect that.
701 * For multi-arch support it is recommended to use a GCC which has
702 been built with support for GNU indirect functions. This ensures
703 that correct debugging information is generated for functions
704 selected by IFUNC resolvers. This support can either be enabled by
705 configuring GCC with '--enable-gnu-indirect-function', or by
706 enabling it by default by setting 'default_gnu_indirect_function'
707 variable for a particular architecture in the GCC source file
710 * GDB pretty printers have been added for mutex and condition variable
711 structures in POSIX Threads. When installed and loaded in gdb these pretty
712 printers show various pthread variables in human-readable form when read
713 using the 'print' or 'display' commands in gdb.
715 * Tunables feature added to allow tweaking of the runtime for an application
716 program. This feature can be enabled with the '--enable-tunables' configure
717 flag. The GNU C Library manual has details on usage and README.tunables has
718 instructions on adding new tunables to the library.
720 * A new version of condition variables functions have been implemented in
721 the NPTL implementation of POSIX Threads to provide stronger ordering
724 * A new version of pthread_rwlock functions have been implemented to use a more
725 scalable algorithm primarily through not using a critical section anymore to
728 Security related changes:
730 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
731 have been created with makecontext could fail to terminate due to a
732 missing .cantunwind annotation. This has been observed to lead to a hang
733 (denial of service) in some Go applications compiled with gccgo. Reported
734 by Andreas Schwab. (CVE-2016-6323)
736 * The DNS stub resolver functions would crash due to a NULL pointer
737 dereference when processing a query with a valid DNS question type which
738 was used internally in the implementation. The stub resolver now uses a
739 question type which is outside the range of valid question type values.
742 The following bugs are resolved with this release:
744 [4099] stdio: Overly agressive caching by stream i/o functions.
745 [7065] build: Support building glibc with -fstack-protector or -fstack-
747 [9842] localedata: en_CA: incorrect date format
748 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
749 before it started waiting
750 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
751 [15765] nptl: sem_open is wrongly a cancellation point
752 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
753 [16458] libc: endian.h and netinet/in.h byte order macros return values of
755 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
756 library linked with pthread
757 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
759 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
760 [17252] libc: getrandom and getentropy syscall
761 [17730] malloc: thread-local storage is sometimes improperly free()'d
762 after being __libc_memalign()'d
763 [18241] stdio: failed fseek on memstream does not set errno and can fail
765 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
767 [18463] nptl: pthread_cond_broadcast issue when surrounded by
768 PTHREAD_PRIO_INHERIT mutex on ARM
769 [18784] network: res_query and related function crash for special record
770 type queries (CVE-2015-5180)
771 [19380] math: strtod does not raise "inexact"
772 [19387] string: Integer overflow in memchr
773 [19390] string: Integer overflow in strncat
774 [19398] build: linknamespace tests fail with massively parallel build
775 [19402] nptl: Deadlock with robust shared mutex and asynchronous
777 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
778 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
779 [19514] libc: [PATCH] Fix spelling errors in spelling
780 "implement"/"implementation" in several places
781 [19582] network: Deprecate RES_USE_INET6
782 [19673] manual: clog10 docs appear to be erroneous
783 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
784 causes a segmentation fault
785 [19826] libc: invalid pointer returned from __tls_get_addr with static
787 [20016] network: resolv: Remove hooks support from the API
788 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
790 [20033] math: [x86_64] vectorized math function don't call the __finite
792 [20116] nptl: use after free in pthread_create
793 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
794 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
796 [20311] nptl: please install proc_service.h
797 [20366] libc: Compilation errors in installed headers in strict-compliance
799 [20370] malloc: malloc: Arena free list management is still racy
800 (incorrect fix in bug 19243)
801 [20386] libc: assert (X = 0) does not result in GCC warning
802 [20432] malloc: malloc: Minimize interface required for interposition
803 [20435] libc: Missing unwind info in __startcontext causes infinite loop
804 in _Unwind_Backtrace (CVE-2016-6323)
805 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
806 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
808 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
809 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
810 [20477] network: resolv: incorrect double-checked locking related to
812 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
814 [20495] math: x86_64 performance degradation due to AVX/SSE transition
816 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
817 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
818 cause transition penalty
819 [20517] math: sparcv9 missing fdiml compat symbol
820 [20524] manual: strverscmp is inconsistent
821 [20525] libc: <sys/quota.h> should be based on kernel headers
822 [20539] math: GCC 7 -static -lm fails to link at -O3
823 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
825 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
827 [20591] network: Remove obsolete DNSSEC support
828 [20592] network: DNS resource record type classification macros in
829 <arpa/nameser.h> are incorrect
830 [20593] network: Update DNS RR type definitions
831 [20611] network: getaddrinfo accepts invalid numeric scope IDs
832 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
834 [20629] network: libresolv: Remove support for bitstring labels
836 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
837 assertion: "maxidx >= 2"
838 [20660] math: [arm] Use VSQRT
839 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
840 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
841 [20689] libc: Test for FMA should also check for AVX.
842 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
843 [20715] math: iszero macro breaks existing code
844 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
845 [20728] libc: powerpc: Missing TOC stub in clone
846 [20729] build: glibc-2.24 fails to build for i486 with -Os
847 [20750] build: Build fails with default PIE enabled gcc-6.x
848 [20768] math: [alpha] sqrt fegetenv namespace
849 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
851 [20787] math: float_t is defined as float incorrectly on x86_64 even if
852 __FLT_EVAL_METHOD__ is 2
853 [20790] network: rpcgen buffer overrun in get_prog_declaration
854 [20822] nptl: race condition in __lll_unlock_elision on powerpc
855 [20829] libc: crypt snprintf namespace
856 [20847] libc: tst-vfork3 failure
857 [20855] math: Default bits/mathdef.h has inappropriate float_t
858 [20859] math: [sh4] FP_ILOGB0 invalid
859 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
860 U+20AC), not same as GBK
861 [20915] dynamic-link: global-dynamic TLS broken on aarch64
862 [20916] math: pow handling of sNaN arguments
863 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
864 [20919] math: Bad pow (qNaN, 0) result with -lieee
865 [20940] math: hypot sNaN handling
866 [20947] math: fmax, fmin sNaN handling
867 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
868 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
870 [20971] string: powerpc64/power7 memchr overflows internal pointer check
871 [20973] nptl: robust mutexes: Lost wake-ups
872 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
873 [20978] nis: strlen on null pointer in nss_nisplus
874 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
876 [21014] string: i686 memchr overflows internal pointer check
877 [21019] libc: [mips] n32 lseek incorrect on overflow
878 [21022] libc: [microblaze] __backtrace get_frame_size namespace
879 [21026] libc: [MIPS] readahead syscall is broken on n64
880 [21028] math: Fallback fesetexceptflag should always succeed
881 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
882 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
884 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
885 [21061] librt: [microblaze] librt lost clock_* exports
886 [21073] libc: tunables: insecure environment variables passed to
887 subprocesses with AT_SECURE
888 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
892 * The minimum Linux kernel version that this version of the GNU C Library
893 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
894 kernel version 2.6.32 or later suffices (on architectures that already
895 required kernel versions more recent than 3.2, those requirements remain
896 unchanged). Linux 3.2 or later kernel headers are required on all
899 * The pap_AN locale has been deleted. This has been deprecated for a long
900 time. It has been replaced by pap_AW & pap_CW, both of which have long
901 been included in previous releases.
903 * The readdir_r and readdir64_r functions have been deprecated. It is
904 recommended to use readdir and readdir64 instead.
906 * The type “union wait” has been removed. It was deprecated in the early
907 1990s and never part of POSIX. Application code should use the int type
908 instead of “union wait”.
910 * A new NSS action is added to facilitate large distributed system
911 administration. The action, MERGE, allows remote user stores like LDAP
912 to be merged into local user stores like /etc/groups in order to provide
913 easy to use, updated, and managed sets of merged credentials. The new
914 action can be used by configuring it in /etc/nsswitch.conf:
915 group: files [SUCCESS=merge] nis
916 Implemented by Stephen Gallagher (Red Hat).
918 * The deprecated __malloc_initialize_hook variable has been removed from the
921 * The long unused localedef --old-style option has been removed. It hasn't
922 done anything in over 16 years. Scripts using this option can safely
925 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
926 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
927 return the next representable value in the direction of positive infinity
928 and the nextdown functions return the next representable value in the
929 direction of negative infinity. These are currently enabled as GNU
932 Security related changes:
934 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
935 could result in a stack overflow when getnetbyname was called with an
936 overly long name. (CVE-2016-3075)
938 * Previously, getaddrinfo copied large amounts of address data to the stack,
939 even after the fix for CVE-2013-4458 has been applied, potentially
940 resulting in a stack overflow. getaddrinfo now uses a heap allocation
941 instead. Reported by Michael Petlan. (CVE-2016-3706)
943 * The glob function suffered from a stack-based buffer overflow when it was
944 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
945 Reported by Alexander Cherepanov. (CVE-2016-1234)
947 * The Sun RPC UDP client could exhaust all available stack space when
948 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
949 alloca plugin for GCC. (CVE-2016-4429)
951 * The IPv6 name server management code in libresolv could result in a memory
952 leak for each thread which is created, performs a failing naming lookup,
953 and exits. Over time, this could result in a denial of service due to
954 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
956 The following bugs are resolved with this release:
958 [1170] localedata: ne_NP: update Nepali locale definition file
959 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
961 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
962 [6796] math: fdim() does not set errno on overflow
963 [10354] libc: posix_spawn should use vfork() in more cases than presently
964 [11213] localedata: localedata: add copyright disclaimer to locale files
965 [12143] localedata: chr_US: new Cherokee locale
966 [12450] localedata: sgs_LT: new locale
967 [12676] localedata: ln_CD: new locale
968 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
970 [13304] math: fma, fmaf, fmal produce wrong results
971 [14259] build: --localedir arg to configure is ignored
972 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
973 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
974 [14934] localedata: es_CL: wrong first weekday chilean locale
975 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
977 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
979 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
980 [15368] nptl: raise() is not async-signal-safe
981 [15479] math: ceil, floor, round and trunc raise inexact exception
982 [15578] localedata: kk_KZ: various updates
983 [16003] localedata: pap_AN: punt old locale
984 [16137] localedata: iw_IL: punt old locale
985 [16190] localedata: eo: new esperanto locale
986 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
987 [16742] malloc: race condition: pthread_atfork() called before first
988 malloc() results in unexpected locking behaviour/deadlocks
989 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
991 [16983] localedata: postal_fmt does not allow %l and %n modifiers
992 [17565] localedata: pt_PT: wrong (work-)week start
993 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
994 [17950] build: Build fails with -msse
995 [18205] localedata: be_BY*: wrong first_weekday and first_workday
996 [18433] libc: posix_spawn does not return correctly upon failure to
998 [18453] localedata: charmaps/IBM875: incorrect codes
999 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1001 [18896] localedata: he_IL: improvements for currency
1002 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1003 Romanian locale data
1004 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1006 [19056] libc: Deprecate readdir_r
1007 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1009 [19198] localedata: nl_NL: small improvements for Dutch locales
1010 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1011 nameservers (CVE-2016-5417)
1012 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1014 [19400] locale: Language missing in "iso-639.def", trivial fix in
1016 [19431] malloc: Deadlock between fflush, getdelim, and fork
1017 [19505] libc: Incorrect file descriptor validity checks in
1018 posix_spawn_file_actions_add{open,close,dup2}
1019 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1020 when using RTLD_NEXT
1021 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1022 `intl/{gettextP,loadinfo}.h'
1023 [19534] libc: execle, execlp may use malloc
1024 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1025 [19573] network: res_nclose and __res_maybe_init disagree about name
1026 server initialization, breaking Hesiod
1027 [19575] localedata: Status of GB18030 tables
1028 [19581] localedata: sr_* date_fmt string contains additional newline
1029 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1031 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1032 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1033 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1034 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1036 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1038 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1040 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1041 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1042 [19633] locale: strfmon_l applies global locale to number formatting
1043 [19642] network: Memory leak in getnameinfo
1044 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1045 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1047 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1048 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1050 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1051 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1053 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1055 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1057 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1058 pointers and lengths in error-case.
1059 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1060 report errors on UTF-16 surrogates values.
1061 [19755] nscd: nscd assertion failure in gc
1062 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1063 [19759] libc: mempcpy shouldn't be inlined
1064 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1065 [19765] libc: s390 needs an optimized mempcpy
1066 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1067 NAME_MAX limit assumption (CVE-2016-1234)
1068 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1069 [19787] network: Missing and incorrect truncation checks in getnameinfo
1070 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1072 [19791] network: Assertion failure in res_query.c with un-connectable name
1074 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1075 [19822] math: libm.so install clobbers old version
1076 [19825] network: resolv: send_vc can return uninitialized data in second
1077 response to getaddrinfo
1078 [19830] network: nss_dns: should check RDATA length against buffer length
1079 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1080 confronted with A/AAAA records of invalid size
1081 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1082 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1083 [19853] stdio: Printing IBM long double in decimal with high precision is
1085 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1086 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1087 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1089 [19865] network: Assertion failure or memory leak in
1090 _nss_dns_getcanonname_r
1091 [19868] network: nss_dns: netent code does not skip over non-PTR records
1092 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1094 [19881] string: Improve x86-64 memset
1095 [19907] string: Incorrect memcpy tests
1096 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1098 [19925] libc: termios.h XCASE namespace
1099 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1100 [19929] libc: limits.h NL_NMAX namespace
1101 [19931] stdio: Memory leak in vfprintf
1102 [19957] libc: clone(CLONE_VM) access invalid parent memory
1103 [19963] localedata: en_IL: New locale
1104 [19989] stdio: stdio.h cuserid namespace
1105 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1107 [19996] locale: langinfo.h nl_langinfo_l namespace
1108 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1109 [20010] network: getaddrinfo: Stack overflow in hostent translation
1111 [20012] stdio: libio: fmemopen append mode failure
1112 [20014] stdio: stdio.h namespace for pre-threads POSIX
1113 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1114 [20023] libc: fcntl.h timespec namespace
1115 [20024] math: [x86_64] vectorized sincos trashes the stack
1116 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1117 [20041] time: sys/time.h timespec namespace
1118 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1119 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1120 [20051] libc: ttyslot in wrong header under wrong conditions
1121 [20054] libc: gethostname not declared for XPG4
1122 [20055] libc: termios.h missing tcgetsid for XPG4
1123 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1125 [20073] libc: sys/stat.h fchmod namespace
1126 [20074] libc: stdlib.h rand_r namespace
1127 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1128 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1130 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1131 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1133 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1134 [20119] libc: Wrong mask for processors level type from CPUID
1135 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1136 AS not supporting AVX512
1137 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1139 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1140 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1142 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1143 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1144 [20178] libc: posix_spawn{p} should not call exit
1145 [20191] stdio: libio: vtables hardening
1146 [20195] string: FMA4 detection requires CPUID execution with register
1148 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1149 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1151 [20212] math: acos (sNaN) returns sNaN
1152 [20213] math: asin (sNaN) returns sNaN
1153 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1154 [20218] math: [i386] asinhl (sNaN) returns sNaN
1155 [20219] math: [i386] atanhl (sNaN) returns sNaN
1156 [20222] stdio: fopencookie: Mangle function pointers
1157 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1158 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1159 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1161 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1162 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1163 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1164 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1165 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1166 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1167 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1168 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1169 [20240] math: modf (sNaN) returns sNaN
1170 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1171 [20250] math: frexp (sNaN) returns sNaN
1172 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1173 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1175 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1176 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1177 subscript is below array bounds [-Werror=array-bounds]
1178 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1179 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1181 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1182 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1184 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1185 "invalid" exceptions
1186 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1187 [20316] localedata: id_ID: Februari instead of Pebruari
1188 [20327] string: POWER8 strcasecmp returns incorrect result
1189 [20347] math: Failure: Test: j0_downward (0xap+0)
1190 [20348] libc: FAIL: misc/tst-preadvwritev64
1191 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1192 [20350] libc: There is no test for p{read,write}64
1193 [20357] math: Incorrect cos result for 1.5174239687223976
1194 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1198 * Unicode 8.0.0 Support: Character encoding, character type info, and
1199 transliteration tables are all updated to Unicode 8.0.0, using new
1200 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1201 These updates cause user visible changes, such as the fixes for bugs
1202 89, 16061, and 18568.
1204 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1205 kernel-internal CPU set size. This means that requests that change the
1206 CPU affinity which failed before (for example, an all-ones CPU mask) will
1207 now succeed. Applications that need to determine the effective CPU
1208 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1209 setting it because the kernel can adjust it (and the previous size check
1210 would not detect this in the majority of cases).
1212 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1213 the following new symbols are used: fts64_children, fts64_close,
1214 fts64_open, fts64_read and fts64_set.
1216 * getaddrinfo now detects certain invalid responses on an internal netlink
1217 socket. If such responses are received, an affected process will
1218 terminate with an error message of "Unexpected error <number> on netlink
1219 descriptor <number>" or "Unexpected netlink response of size <number> on
1220 descriptor <number>". The most likely cause for these errors is a
1221 multi-threaded application which erroneously closes and reuses the netlink
1222 file descriptor while it is used by getaddrinfo.
1224 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1225 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1226 unnecessary serialization of memory allocation requests across threads.
1227 The defect is now corrected. Users should see a substantial increase in
1228 the concurent throughput of allocation requests for applications which
1229 trigger this bug. Affected applications typically create and
1230 destroy threads frequently. (Bug 19048 was reported and analyzed by
1233 * There is now a --disable-timezone-tools configure option for disabling the
1234 building and installing of the timezone related utilities (zic, zdump, and
1235 tzselect). This is useful for people who build the timezone data and code
1236 independent of the GNU C Library.
1238 * The obsolete header <regexp.h> has been removed. Programs that require
1239 this header must be updated to use <regex.h> instead.
1241 * The obsolete functions bdflush, create_module, get_kernel_syms,
1242 query_module and uselib are no longer available to newly linked binaries;
1243 the header <sys/kdaemon.h> has been removed. These functions and header
1244 were specific to systems using the Linux kernel and could not usefully be
1245 used with the GNU C Library on systems with version 2.6 or later of the
1248 * Optimized string, wcsmbs and memory functions for IBM z13.
1249 Implemented by Stefan Liebler.
1251 * Newly linked programs that define a variable called signgam will no longer
1252 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1253 require signgam to be set by those functions must ensure that they use the
1254 variable provided by the GNU C Library and declared in <math.h>, without
1255 defining their own copy.
1257 * The minimum GCC version that can be used to build this version of the GNU
1258 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1259 still be used to compile programs using the GNU C Library.
1261 Security related changes:
1263 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1264 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1266 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1267 the pointer guard feature. It is always enabled. Previously,
1268 LD_POINTER_GUARD could be used to disable security hardening in binaries
1269 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1272 * An integer overflow in hcreate and hcreate_r could lead to an
1273 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1275 * The catopen function no longer has unbounded stack usage. Reported by
1276 Max. (CVE-2015-8779)
1278 * The nan, nanf and nanl functions no longer have unbounded stack usage
1279 depending on the length of the string passed as an argument to the
1280 functions. Reported by Joseph Myers. (CVE-2014-9761)
1282 * A stack-based buffer overflow was found in libresolv when invoked from
1283 libnss_dns, allowing specially crafted DNS responses to seize control
1284 of execution flow in the DNS client. The buffer overflow occurs in
1285 the functions send_dg (send datagram) and send_vc (send TCP) for the
1286 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1287 family. The use of AF_UNSPEC triggers the low-level resolver code to
1288 send out two parallel queries for A and AAAA. A mismanagement of the
1289 buffers used for those queries could result in the response of a query
1290 writing beyond the alloca allocated buffer created by
1291 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1292 the overflow. Thanks to the Google Security Team and Red Hat for
1293 reporting the security impact of this issue, and Robert Holiday of
1294 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1296 The following bugs are resolved with this release:
1298 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1299 [887] math: Math library function "logb" and "nextafter" inconsistent
1300 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1301 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1302 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1303 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1305 [4404] localedata: German translation of "Alarm clock" is misleading
1306 [6799] math: nextafter() and nexttoward() doen't set errno on
1307 overflow/underflow errors
1308 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1310 [10432] nis: _nss_nis_setnetgrent assertion failure
1311 [11460] libc: fts has no LFS support
1312 [12926] network: getaddrinfo()/make_request() may spin forever
1313 [13065] nptl: Race condition in pthread barriers
1314 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1315 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1317 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1318 [14912] libc: Rename non-installed bits/*.h headers
1319 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1320 [15367] math: Let gcc use __builtin_isinf
1321 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1322 [15421] math: lgamma wrongly sets signgam for ISO C
1323 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1324 FE_INVALID with argument out of range
1325 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1327 [15786] dynamic-link: ifunc resolver functions can smash function
1329 [15918] math: Unnecessary check for equality in hypotf()
1330 [16061] localedata: Review / update transliteration data
1331 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1333 [16141] time: strptime %z offset restriction
1334 [16171] math: drem should be alias of remainder
1335 [16296] math: fegetround is pure?
1336 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1337 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1339 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1340 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1342 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1343 "Exception "Inexact" set" on ppc32
1344 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1345 [16517] math: Missing underflow exception from tanf/tan/tanl
1346 [16519] math: Missing underflow exception from sinhf
1347 [16520] math: Missing underflow exception from tanhf
1348 [16521] math: Missing underflow exception from exp2
1349 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1351 [16734] stdio: fopen calls mmap to allocate its buffer
1352 [16961] math: nan function incorrect handling of bad sequences
1353 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1354 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1355 [16985] locale: localedef: confusing error message when opening output
1357 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1358 [17197] locale: Redundant shift character in iconv conversion output at
1360 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1361 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1362 [17250] dynamic-link: static linking breaks nss loading
1363 (getaddrinfo/getpwnam/etc...)
1364 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1366 [17441] math: isnan() should use __builtin_isnan() in GCC
1367 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1368 (related to lock elision)
1369 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1370 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1371 [17887] time: strptime should be able to parse "+01:00" style timezones
1372 [17905] libc: catopen() Multiple unbounded stack allocations
1374 [18084] libc: backtrace (..., 0) dumps core on x86
1375 [18086] libc: nice() sets errno to 0 on success
1376 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1377 is too large (CVE-2015-8778)
1378 [18251] dynamic-link: SONAME missing when audit modules provides path
1379 [18265] libc: add attributes for wchar string and memory functions
1380 [18370] math: csqrt missing underflows
1381 [18421] libc: [hppa] read-only segment has dynamic relocations
1382 [18472] libc: Obsolete syscall wrappers should be compat symbols
1383 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1384 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1385 [18525] localedata: Remove locale timezone information
1386 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1387 [18568] localedata: Update locale data to Unicode 8.0
1388 [18589] locale: sort-test.sh fails at random
1389 [18595] math: ctan, ctanh missing underflows
1390 [18604] libc: assert macro-expands its argument
1391 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1392 contains a vector instruction exception.
1393 [18611] math: j1, jn missing errno setting on underflow
1394 [18618] localedata: sync Chechen locale definitions with other *_RU
1396 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1397 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1398 [18665] network: In send_dg, the recvfrom function is NOT always using the
1399 buffer size of a newly created buffer (CVE-2015-7547)
1400 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1402 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1404 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1405 [18699] math: tilegx cproj() for various complex infinities does not yield
1407 [18724] libc: Harden put*ent functions against data injection
1408 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1409 [18755] build: build errors with -DNDEBUG
1410 [18757] stdio: fmemopen fails to set errno on failure
1411 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1413 [18781] libc: openat64 lacks O_LARGEFILE
1414 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1415 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1416 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1417 [18790] math: [ldbl-128ibm] tanhl inaccurate
1418 [18795] libc: stpncpy fortification misses buffer lengths that are
1419 statically too large
1420 [18796] build: build fails for --disable-mathvec
1421 [18803] math: hypot missing underflows
1422 [18820] stdio: fmemopen may leak memory on failure
1423 [18823] math: csqrt spurious underflows
1424 [18824] math: fma spurious underflows
1425 [18825] math: pow missing underflows
1426 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1427 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1428 [18870] build: sem_open.c fails to compile with missing symbol
1430 [18872] stdio: Fix memory leak in printf_positional
1431 [18873] libc: posix_fallocate overflow check ineffective
1432 [18875] math: Excess precision leads incorrect libm
1433 [18877] libc: arm: mmap offset regression
1434 [18887] libc: memory corruption when using getmntent on blank lines
1435 [18918] localedata: hu_HU: change time to HH:MM:SS format
1436 [18921] libc: Regression: extraneous stat() and fstat() performed by
1438 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1439 binaries (CVE-2015-8777)
1440 [18951] math: tgamma missing underflows
1441 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1443 [18953] localedata: lt_LT: change currency symbol to the euro
1444 [18956] math: powf inaccuracy
1445 [18961] math: [i386] exp missing underflows
1446 [18966] math: [i386] exp10 missing underflows
1447 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1448 [18969] build: multiple string test failures due to missing locale
1450 [18970] libc: Reference of pthread_setcancelstate in libc.a
1451 [18977] math: float / long double Bessel functions not in XSI POSIX
1452 [18980] math: i386 libm functions return with excess range and precision
1453 [18981] math: i386 scalb*, ldexp return with excess range and precision
1454 [18982] stdio: va_list and vprintf
1455 [18985] time: Passing out of range data to strftime() causes a segfault
1457 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1458 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1459 [19012] locale: iconv_open leaks memory on error path
1460 [19016] math: clog, clog10 inaccuracy
1461 [19018] nptl: Mangle function pointers in tls_dtor_list
1462 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1463 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1464 [19048] malloc: malloc: arena free list can become cyclic, increasing
1466 [19049] math: [powerpc] erfc incorrect zero sign
1467 [19050] math: [powerpc] log* incorrect zero sign
1468 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1469 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1470 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1471 [19074] network: Data race in _res_hconf_reorder_addrs
1472 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1473 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1474 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1476 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1478 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1479 [19086] manual: posix_fallocate64 documented argument order is wrong.
1480 [19088] math: lround, llround missing exceptions close to overflow
1482 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1483 [19095] math: dbl-64 lrint incorrect for 64-bit long
1484 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1485 [19124] dynamic-link: ld.so failed to build with older assmebler
1486 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1487 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1488 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1489 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1490 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1491 pthread_setaffinity_np
1492 [19156] math: [ldbl-128] j0l spurious underflows
1493 [19164] nptl: tst-getcpu fails with many possible CPUs
1494 [19168] math: math/test-ildoubl and math/test-ldouble failure
1495 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1496 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1498 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1499 (FE_NOMASK_ENV) do not clear SSE exceptions
1500 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1501 _int_new_arena/reused_arena
1502 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1503 [19201] math: dbl-64 remainder incorrect sign of zero result
1504 [19205] math: bits/math-finite.h conditions do not match math.h and
1506 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1507 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1508 for C99-based standards
1509 [19212] libc: features.h not -Wundef clean
1510 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1512 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1513 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1514 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1516 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1517 "inexact" exceptions
1518 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1520 [19242] libc: strtol incorrect in Turkish locales
1521 [19243] malloc: reused_arena can pick an arena on the free list, leading
1522 to an assertion failure and reference count corruption
1523 [19253] time: tzset() ineffective when temporary TZ did not include DST
1525 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1526 [19270] math: [hppa] Shared libm missing __isnanl
1527 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1528 MAP_HUGETLB and MAP_STACK defines
1529 [19313] nptl: Wrong __cpu_mask for x32
1530 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1531 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1532 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1533 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1534 [19363] time: x32: times() return value wrongly truncates/sign extends
1536 [19367] dynamic-link: Improve branch prediction on Silvermont
1537 [19369] network: Default domain name not reset by res_ninit when "search"
1538 / "domain" entry is removed from resolv.conf
1539 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1541 [19385] localedata: bg_BG: time separator should be colon, not comma
1542 [19408] libc: linux personality syscall wrapper may erroneously return an
1543 error on 32-bit architectures
1544 [19415] libc: dladdr returns wrong names on hppa
1545 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1546 IBM905, IBM907, and IBM909
1547 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1548 [19443] build: build failures with -DDEBUG
1549 [19451] build: Make check fails on test-double-vlen2
1550 [19462] libc: Glibc failed to build with -Os
1551 [19465] math: Wrong code with -Os
1552 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1554 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1556 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1557 [19486] math: S390: Math tests fail with "Exception Inexact set".
1558 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1559 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1561 [19590] math: Fail to build shared objects that use libmvec.so functions.
1565 * The following bugs are resolved with this release:
1567 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1568 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1569 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1570 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1571 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1572 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1573 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1574 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1575 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1576 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1577 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1578 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1579 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1580 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1581 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1582 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1583 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1584 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1585 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1586 18657, 18676, 18694, 18696, 18887.
1588 * Cache information can be queried via sysconf() function on s390 e.g. with
1589 _SC_LEVEL1_ICACHE_SIZE as argument.
1591 * A buffer overflow in gethostbyname_r and related functions performing DNS
1592 requests has been fixed. If the NSS functions were called with a
1593 misaligned buffer, the buffer length change due to pointer alignment was
1594 not taken into account. This could result in application crashes or,
1595 potentially arbitrary code execution, using crafted, but syntactically
1596 valid DNS responses. (CVE-2015-1781)
1598 * The time zone file parser has been made more robust against crafted time
1599 zone files, avoiding heap buffer overflows related to the processing of
1600 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1601 large time zone data files. Overly long time zone specifiers in the TZ
1602 variable no longer result in stack overflows and crashes.
1604 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1605 for LD and GD on x86 and x86-64, has been implemented. You will need
1606 binutils-2.24 or later to enable this optimization.
1608 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1609 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1610 Hat). These updates cause user visible changes, such as the fix for bug
1613 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1614 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1615 condition in some applications.
1617 * Added vector math library named libmvec with the following vectorized x86_64
1618 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1620 The library can be disabled with --disable-mathvec. Use of the functions is
1621 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1622 Shared library libmvec.so is linked in as needed when using -lm (no need to
1623 specify -lmvec explicitly for not static builds).
1624 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1626 * A new fmemopen implementation has been added with the goal of POSIX
1627 compliance. The new implementation fixes the following long-standing
1628 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1629 old implementation is still present for use be by existing binaries.
1631 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1632 releases. It has been fixed to match 2.19 and older, but binaries built
1633 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1635 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1636 Contributed by Roland McGrath (Google).
1638 * The header <regexp.h> is deprecated, and will be removed in a future
1639 release. Use of this header will trigger a deprecation warning.
1640 Application developers should update their code to use <regex.h> instead.
1642 This header was formerly part of SUS, but was deprecated in 1994 and
1643 removed from the standard in 2001. Also, the glibc implementation
1644 leaks memory. See BZ#18681 for more details.
1648 * The following bugs are resolved with this release:
1650 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1651 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1652 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1653 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1654 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1655 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1656 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1657 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1658 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1659 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1662 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1663 too little memory for the to-be-scanned arguments and overflow the
1664 allocated buffer. The implementation now correctly computes the required
1665 buffer size when using malloc, and switches to malloc from alloca as
1668 * A new semaphore algorithm has been implemented in generic C code for all
1669 machines. Previous custom assembly implementations of semaphore were
1670 difficult to reason about or ensure that they were safe. The new version
1671 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1672 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1673 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1675 * Port to Altera Nios II has been contributed by Mentor Graphics.
1677 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1678 implementations for powerpc64/powerpc64le.
1679 Implemented by Adhemerval Zanella (IBM).
1681 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1682 and powerpc64le. This may improve lock scaling of existing programs on
1683 HTM capable systems. The lock elision code is only enabled with
1684 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1685 powerpc will issue a transaction abort on every syscall to avoid side
1686 effects being visible outside transactions.
1688 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1689 AArch64. Contributed by ARM Ltd.
1691 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1693 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1694 under certain input conditions resulting in the execution of a shell for
1695 command substitution when the application did not request it. The
1696 implementation now checks WRDE_NOCMD immediately before executing the
1697 shell and returns the error WRDE_CMDSUB as expected.
1699 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1700 processing format strings with a large number of format specifiers.
1702 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1703 infinite loop if the DNS response contained a PTR record of an unexpected
1706 * The minimum GCC version that can be used to build this version of the GNU
1707 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1708 still be used to compile programs using the GNU C Library.
1710 * The GNU C Library is now built with -Werror by default. This can be
1711 disabled by configuring with --disable-werror.
1713 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1715 * The obsolete sigvec function has been removed. This was the original
1716 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1717 programs have been using instead for about 25 years. Of course, ABI
1718 compatibility for old binaries using sigvec remains intact.
1720 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1721 with newer versions of bison.
1723 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1724 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1725 registers overlay two consecutive single-precision registers. MIPS32R2
1726 introduced a new FPU mode (FR=1) where double-precision registers extend the
1727 corresponding single-precision registers which is incompatible with the
1728 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1729 require the use of FR=1 making a transition necessary. New o32 ABI
1730 extensions enable users to migrate over time from the original o32 ABI
1731 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1732 tracks the ABI of any loaded object and verifies that new objects are
1733 compatible. Mode transitions will also be requested as required and
1734 unsupportable objects will be rejected. The ABI checks include both soft and
1735 hard float ABIs for o32, n32 and n64.
1737 GCC 5 with GNU binutils 2.25 onwards:
1738 It is strongly recommended that all o32 system libraries are built using the
1739 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1740 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1741 compiler using --with-fp-32=xx to set this by default.
1745 * The following bugs are resolved with this release:
1747 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1748 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1749 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1750 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1751 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1752 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1753 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1754 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1755 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1756 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1757 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1758 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1759 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1760 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1761 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1763 * Reverted change of ABI data structures for s390 and s390x:
1764 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1765 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1766 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1767 are preserved pointing straight to the same implementation as the old ones.
1768 Given that, new callers will simply provide a too-big buffer to these
1769 functions. Any applications/libraries out there that embed jmp_buf or
1770 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1771 against 2.19 headers will have to rebuilt again. This is necessary in any
1772 case to revert the breakage in their ABI caused by the glibc change.
1774 * Support for file description locks is added to systems running the
1775 Linux kernel. The standard file locking interfaces are extended to
1776 operate on file descriptions, not file descriptors, via the use of
1777 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1778 are associated with an open file instead of a process.
1780 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1782 * The minimum Linux kernel version that this version of the GNU C Library
1783 can be used with is 2.6.32.
1785 * Running the testsuite no longer terminates as soon as a test fails.
1786 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1787 with PASS or FAIL lines for individual tests. A summary of the results is
1788 printed, including a list of failing lists, and "make check" exits with
1789 error status if there were any unexpected failures. "make check
1790 stop-on-test-failure=y" may be used to keep the old behavior.
1792 * The am33 port, which had not worked for several years, has been removed
1795 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1796 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1797 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1798 conflicted with POSIX (support for which was removed in 2.19), the
1799 interfaces those macros enabled remain available when compiling with
1800 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1801 test macros defined.
1803 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1805 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1806 This may improve lock scaling of existing programs on TX capable systems.
1807 The lock elision code is only built with --enable-lock-elision=yes and
1808 then requires a GCC version supporting the TX builtins. With lock elision
1809 default mutexes are elided via __builtin_tbegin, if the cpu supports
1810 transactions. By default lock elision is not enabled and the elision code
1813 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1814 copy the path argument. This allowed programs to cause posix_spawn to
1815 deference a dangling pointer, or use an unexpected pathname argument if
1816 the string was modified after the posix_spawn_file_actions_addopen
1819 * All supported architectures now use the main glibc sysdeps directory
1820 instead of some being in a separate "ports" directory (which was
1821 distributed separately before glibc 2.17).
1823 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1824 On configurations that support it (all Linux configurations), it's now
1825 used regardless of the --enable-add-ons switch to configure. It is no
1826 longer possible to build such configurations without pthreads support.
1828 * Locale names, including those obtained from environment variables (LANG
1829 and the LC_* variables), are more tightly checked for proper syntax.
1830 setlocale will now fail (with EINVAL) for locale names that are overly
1831 long, contain slashes without starting with a slash, or contain ".." path
1832 components. (CVE-2014-0475) Previously, some valid locale names were
1833 silently replaced with the "C" locale when running in AT_SECURE mode
1834 (e.g., in a SUID program). This is no longer necessary because of the
1837 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1838 with application code using Intel MPX instructions. (With all previous
1839 versions, the MPX register state could be clobbered when making calls
1840 into or out of a shared library.) Note that while the new dynamic
1841 linker is compatible with all known x86 hardware whether or not it
1842 supports Intel MPX, some x86 instruction-set emulators might fail to
1843 handle the new instruction encodings. This is known to affect Valgrind
1844 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1845 release), and might affect other tools that do instruction emulation.
1847 * Support for loadable gconv transliteration modules has been removed.
1848 The support for transliteration modules has been non-functional for
1849 over a decade, and the removal is prompted by security defects. The
1850 normal gconv conversion modules are still supported. Transliteration
1851 with //TRANSLIT is still possible, and the //IGNORE specifier
1852 continues to be supported. (CVE-2014-5119)
1854 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1855 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1856 resulting a denial-of-service security vulnerability in applications which
1857 use functions related to iconv. (CVE-2014-6040)
1861 * The following bugs are resolved with this release:
1863 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1864 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1865 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1866 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1867 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1868 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1869 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1870 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1871 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1872 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1873 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1874 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1875 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1876 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1877 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1878 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1879 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1880 16431, 16453, 16474, 16506, 16510, 16529
1882 * Slovenian translations for glibc messages have been contributed by the
1883 Translation Project's Slovenian team of translators.
1885 * The public headers no longer use __unused nor __block. This change is to
1886 support compiling programs that are derived from BSD sources and use
1887 __unused internally, and to support compiling with Clang's -fblock
1888 extension which uses __block.
1890 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1891 large collation sequences to optimize multiple passes. This cache
1892 computation may overflow for large collation sequences and may cause a
1893 stack or buffer overflow. This is now fixed to use a slower algorithm
1894 which does not use a cache if there is an integer overflow.
1896 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1897 rules for large collation sequences to optimize multiple passes and falls
1898 back to alloca if malloc fails, resulting in a possible stack overflow.
1899 The implementation now falls back to an uncached collation sequence lookup
1902 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1903 initialized for static applications resulting in the security feature
1904 being disabled. The pointer guard is now correctly initialized to a
1905 random value for static applications. Existing static applications need
1906 to be recompiled to take advantage of the fix (bug 15754).
1908 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1909 to the d_name member of struct dirent, or omit the terminating NUL
1910 character. (Bugzilla #14699).
1912 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1913 aligned_alloc functions could allocate too few bytes or corrupt the
1914 heap when passed very large allocation size values (Bugzilla #15855,
1917 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1918 for AF_INET6 has been fixed (Bugzilla #16072).
1920 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1921 pap_CW, quz_PE, the_NP.
1923 * Substantially revised locales: gd_GB, ht_HT
1925 * The LC_ADDRESS field was updated to support country_car for almost all
1928 * ISO 1427 definitions were updated.
1930 * ISO 3166 definitions were updated.
1932 * The localedef utility now supports --big-endian and --little-endian
1933 command-line options to generate locales for a different system from that
1934 for which the C library was built.
1936 * Binary locale files now only depend on the endianness of the system for
1937 which they are generated and not on other properties of that system. As a
1938 consequence, binary files generated with new localedef may be incompatible
1939 with old versions of the GNU C Library, and binary files generated with
1940 old localedef may be incompatible with this version of the GNU C Library,
1941 in the following circumstances:
1943 + Locale files may be incompatible on m68k systems.
1945 + Locale archive files (but not separate files for individual locales) may
1946 be incompatible on systems where plain "char" is signed.
1948 * The configure option --disable-versioning has been removed. Builds with
1949 --disable-versioning had not worked for several years.
1951 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1952 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1954 * SystemTap probes for malloc have been introduced.
1956 * SystemTap probes for slow multiple precision fallback paths of
1957 transcendental functions have been introduced.
1959 * Support for powerpc64le has been added.
1961 * The soft-float powerpc port now supports e500 processors.
1963 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1965 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1966 set of header declarations that are enabled by default, even when other
1967 feature test macros or compiler options such as -std=c99 would otherwise
1968 disable some of those declarations.
1970 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1971 conflict with POSIX. The libbsd-compat library (which was a dummy library
1972 that did nothing) has also been removed.
1974 * Preliminary documentation about Multi-Thread, Async-Signal and
1975 Async-Cancel Safety has been added.
1977 * Change of ABI data structures for s390 and s390x:
1978 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1979 allow for future hardware extensions. All library functions that accept or
1980 return these structures were versioned in order to provide backwards
1981 compatibility with existing code. However, not all cases can be handled
1982 with interface versioning. If an application embeds either structure into
1983 another structure and that structure is passed to another compilation unit
1984 which expects the newer larger-sized object then any access to the new
1985 fields will result in undefined behaviour. Similarly any access to fields
1986 that were moved by the enlarging of the structures will cause undefined
1987 behaviour. To fix the undefined behaviour all objects that are part of the
1988 final in-memory image must agree on the size of structures, and this may
1989 require recompilation.
1993 * The following bugs are resolved with this release:
1995 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1996 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1997 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1998 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1999 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2000 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2001 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2002 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2003 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2004 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2005 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2006 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2007 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2010 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2011 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2012 Distributions can re-enable building and using pt_chown via the new configure
2013 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2014 considerable security risks and should only be used if the distribution
2015 understands and accepts the risks.
2017 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2020 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2021 fixed (Bugzilla #15330).
2023 * Add support for calling C++11 thread_local object destructors on thread
2024 and program exit. This needs compiler support for offloading C++11
2025 destructor calls to glibc.
2027 * Improved worst case performance of libm functions with double inputs and
2030 * Support for priority inherited mutexes in pthread condition variables on
2031 non-x86 architectures.
2033 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2035 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2037 * Optimized string functions for ARM. Implemented by Will Newton and
2040 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2042 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2045 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2046 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2048 * Added a benchmark framework to track performance of functions in glibc.
2050 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2051 It is based on draft TS 18661 and currently enabled as a GNU extension.
2053 * On Linux, the clock function now uses the clock_gettime system call
2054 for improved precision, rather than old times system call.
2056 * Added support for version-3 tz format files. This is needed when using
2057 the tz database release 2013e or later, and affects a few unusual cases --
2058 currently only TZ='America/Godthab' for time stamps after 2037.
2060 * Added new API functions pthread_getattr_default_np and
2061 pthread_setattr_default_np to get and set the default pthread
2062 attributes of a process.
2064 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2065 This may improve lock scaling of existing programs on TSX capable systems.
2066 When the --enable-lock-elision=yes parameter is specified at configure
2067 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2070 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2071 availability). Implemented by Ryan S. Arnold.
2073 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2077 * The following bugs are resolved with this release:
2079 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2080 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2081 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2082 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2083 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2084 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2085 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2086 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2087 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2088 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2089 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2090 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2091 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2092 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2093 14879, 14889, 14893, 14898, 14914.
2095 * Optimization of memcpy for MIPS.
2097 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2098 EMFILE has been fixed (Bugzilla #14889).
2100 * The add-on ports collection is now distributed in the "ports" subdirectory
2101 of the main GNU C Library distribution, rather than separately.
2103 * Port to ARM AArch64 contributed by Linaro.
2105 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2106 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2108 Implemented by Andreas Krebbel.
2110 * The new function secure_getenv allows secure access to the environment,
2111 returning NULL if running in a SUID/SGID process. This function replaces
2112 the internal function __secure_getenv.
2114 * SystemTap static probes have been added into the dynamic linker.
2115 Implemented by Gary Benson.
2117 * Optimizations of string functions strstr, strcasestr and memmem.
2118 Implemented by Maxim Kuvyrkov.
2120 * The minimum Linux kernel version that this version of the GNU C Library
2121 can be used with is 2.6.16.
2123 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2124 powerpc POWER7. Implemented by Will Schmidt.
2126 * New configure option --disable-nscd builds the C library such that it
2127 never attempts to contact the Name Service Caching Daemon (nscd).
2128 New configure option --disable-build-nscd avoids building nscd itself;
2129 this is the default if --disable-nscd is used.
2131 * Improved support for cross-compilation, including cross-testing and
2132 bootstrap builds without a previously built glibc.
2134 * Several testsuite tests are now able to test multiple IFUNC variants of an
2135 interface, rather than just testing the one that would be chooen by
2138 * New configure options --with-bugurl and --with-pkgversion, for
2139 distributors to use to embed their bug-reporting and package version
2140 information in --help and --version output.
2142 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2143 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2144 allows creation of chroots without the procfs mounted on /proc.
2146 * The `crypt' function now fails if passed salt bytes that violate the
2147 specification for those values. On Linux, the `crypt' function will
2148 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2149 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2150 when the mode is enabled.
2152 * The `clock_*' suite of functions (declared in <time.h>) is now available
2153 directly in the main C library. Previously it was necessary to link with
2154 -lrt to use these functions. This change has the effect that a
2155 single-threaded program that uses a function such as `clock_gettime' (and
2156 is not linked with -lrt) will no longer implicitly load the pthreads
2157 library at runtime and so will not suffer the overheads associated with
2158 multi-thread support in other code such as the C++ runtime library.
2160 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2165 * The following bugs are resolved with this release:
2167 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2168 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2169 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2170 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2171 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2172 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2173 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2174 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2175 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2176 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2177 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2178 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2179 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2180 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2181 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2182 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2183 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2184 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2185 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2186 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2187 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2188 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2191 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2192 configuring glibc with:
2193 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2194 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2195 Implemented by H.J. Lu.
2199 + define static_assert
2201 + do not declare gets
2203 + declare at_quick_exit and quick_exit also for ISO C11
2205 + aligned_alloc. NB: The code is deliberately allows the size parameter
2206 to not be a multiple of the alignment. This is a moronic requirement
2207 in the standard but it is only a requirement on the caller, not the
2210 + timespec_get added
2212 + uchar.h support added
2214 + CMPLX, CMPLXF, CMPLXL added
2216 Implemented by Ulrich Drepper.
2218 * Support for the IA-64 has been moved to ports.
2220 * Remove support for anything but ELF binary format
2222 * Checking versions of poll, ppoll added.
2223 Implemented by Ulrich Drepper.
2225 * More generic and 64-bit performance optimizations to math functions.
2226 Implemented by Ulrich Drepper.
2228 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2229 headers and functions available at compile time as they were before
2230 version 2.14. This option will be removed at some time in the future
2231 after the TI-RPC library becomes fully sufficient for the needs of
2232 existing applications.
2234 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2235 Note that glibc is not expected to work with any Linux kernel version
2238 * New header <sys/auxv.h> and function getauxval allowing easy access to
2239 the AT_* key-value pairs passed from the Linux kernel. The header also
2240 defines the HWCAP_* bits associated with the AT_HWCAP key.
2242 * New locales: mag_IN
2244 * New configure option --enable-systemtap builds SystemTap static probes
2245 into libc for setjmp and longjmp and into libpthread for various operations.
2246 So far the setjmp/longjmp probes and some of the libpthread probes are
2247 provided only for i*86 and x86_64.
2248 Implemented by Roland McGrath and Rayson Ho.
2250 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2252 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2255 * More optimized functions for SPARC. Implemented by David S. Miller.
2257 * Improved support for cross-compilation, especially bootstrap builds
2258 without a previously built glibc.
2260 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2261 Chris Metcalf from Tilera.
2263 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2264 now supported for ARM processors.
2266 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2267 as the name of the dynamic linker, to distinguish it from the
2268 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2270 * Support for CRIS has been removed from ports.
2272 * A new class of installed header has been documented for low-level
2273 platform-specific functionality. PowerPC added the first instance with a
2274 function to provide time base register access. Contributed by Tulio
2275 Magno Quites Machado Filho.
2277 * ix86 configurations now install header files that are consistent with
2278 what x86-64 configurations install. These same header files can be used
2279 for -m32, -m64, or -mx32 builds.
2280 Contributed by H.J. Lu.
2282 * Math library bug fixes. A thorough audit of all open math library bugs was
2283 conducted by Joseph Myers. Significant progress was made on many math
2284 library bugs resulting in more accurate exceptions and function results.
2285 Many thanks to all those that contributed including Andreas Jaeger for his
2286 patch review and work on the x87 trigonometric instruction issues.
2288 * Timezone data is no longer installed. Timezone-related binaries and scripts
2289 will continue to be installed. Users should obtain their timezone data from
2290 their distribution provider or from the tzdata package at
2291 <ftp://munnari.oz.au/pub/>.
2295 * The following bugs are resolved with this release:
2297 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2298 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2299 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2300 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2301 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2302 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2303 13484, 13506, 13515, 13523, 13524, 13538, 13540
2305 * New program pldd to list loaded object of a process
2306 Implemented by Ulrich Drepper.
2308 * Add nss_db support back to glibc. No more dependency on Berkeley db
2309 and support for initgroups lookups.
2310 Implemented by Ulrich Drepper.
2312 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2313 Contributed by HJ Lu.
2315 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2316 Contributed by HJ Lu.
2318 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2319 on x86-32 and x86-64.
2320 Contributed by Liubov Dmitrieva.
2322 * Optimized strchr and strrchr for SSE on x86-32.
2323 Contributed by Liubov Dmitrieva.
2325 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2326 for x86-64 and x86-32.
2327 Contributed by Liubov Dmitrieva.
2329 * New interfaces: scandirat, scandirat64
2330 Implemented by Ulrich Drepper.
2332 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2333 Implemented by Ulrich Drepper.
2335 * nscd now also caches the netgroup database.
2336 Implemented by Ulrich Drepper.
2338 * Integrate libm with gcc's -ffinite-math-only option.
2339 Implemented by Ulrich Drepper.
2341 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2342 to math functions. Implemented by Ulrich Drepper.
2344 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2345 Implemented by Ulrich Drepper.
2347 * New Linux interfaces: process_vm_readv, process_vm_writev
2349 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2350 Implemented by Ulrich Drepper.
2352 * Optimized nearbyint and strcasecmp for PPC.
2353 Implemented by Adhemerval Zanella.
2355 * New locales: bho_IN, unm_US, es_CU, ta_LK
2359 * The following bugs are resolved with this release:
2361 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2362 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2363 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2364 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2365 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2366 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2367 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2368 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2369 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2370 12795, 12811, 12813, 12814, 12841
2372 * The RPC implementation in libc is obsoleted. Old programs keep working
2373 but new programs cannot be linked with the routines in libc anymore.
2374 Programs in need of RPC functionality must be linked against TI-RPC.
2375 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2377 Visible changes of this change include (obviously) the inability to link
2378 programs using RPC functions without referencing the TI-RPC library and the
2379 removal of the RPC headers from the glibc headers.
2380 Implemented by Ulrich Drepper.
2382 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2383 syncfs, setns, sendmmsg
2385 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2386 yue_HK, lij_IT, mhr_RU
2388 * New encodings: CP770, CP771, CP772, CP773, CP774
2390 * New program sotruss to trace calls through PLTs
2391 Implemented by Ulrich Drepper.
2393 * The malloc hook implementation is marked deprecated and will be removed
2394 from the default implementation in the next version. The design never
2395 worked ever since the introduction of threads. Even programs which do
2396 not create threads themselves can use multiple threads created internally.
2400 * The following bugs are resolved with this release:
2402 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2403 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2404 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2405 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2408 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2410 * POWER7 optimizations: memset, memcmp, strncmp
2412 * New optimized string functions for x86-64: strnlen (SSE2),
2413 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2414 Implemented by Ulrich Drepper.
2418 * The following bugs are resolved with this release:
2420 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2421 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2422 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2423 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2424 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2425 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2426 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2427 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2429 * New interfaces: pthread_getname_np, pthread_setname_np
2431 * New Linux interface: recvmmsg
2433 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2435 * The dynamic linker now recognizes supported ABI versions from the
2436 EI_ABIVERSION field in the ELF header.
2437 Implemented by Ulrich Drepper.
2439 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2440 will not be used to fill in password fields in the passwd.byname replies.
2441 Instead it is used to synthesize the shadow.byname table, should it be
2442 missing. This is a useful mode in some installations involving Solaris.
2443 Implemented by Ulrich Drepper.
2445 * New locales: kok_IN, sq_MK, cv_RU
2449 * The following bugs are resolved with this release:
2451 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2452 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2453 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2454 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2455 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2456 10789, 10817, 10823, 10840, 10847
2458 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2460 Implemented by Ulrich Drepper.
2462 * Checking version of longjmp added that fails if an uninitialized stack
2463 frame would be created. Implemented by Ulrich Drepper.
2465 * STT_GNU_IFUNC is now supported in static executables.
2466 Implemented by H.J. Lu.
2468 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2470 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2471 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2472 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2473 Contributed by H.J. Lu.
2475 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2476 Implemented by Ulrich Drepper.
2478 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2480 Contributed by H.J. Lu.
2482 * Support for fma instruction in AVX on x86-64.
2483 Implemented by H.J. Lu and Ulrich Drepper.
2485 * AVX support in x86-64 auditing support in ld.so.
2486 Implemented by H.J. Lu.
2488 * STB_GNU_UNIQUE support added.
2489 Implemented by Ulrich Drepper.
2491 * Implement second fallback mode for optimized DNS lookup for even more
2492 broken environments. If two requests from the same port are not
2493 handled correctly close the socket and open a new one before sending
2494 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2495 can be used to select this mode right away, instead of rediscovering the
2496 necessity is every process again.
2497 Implemented by Ulrich Drepper.
2499 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2500 Implemented by Adam Tkac.
2502 * Optimized iconv conversions for S390x.
2503 Implemented by Andreas Krebbel.
2505 * Using condvars with PI mutexes is now more efficient due to kernel
2506 support for requeueing to PI futexes. NPTL support added for x86-64.
2507 Implemented by Ulrich Drepper.
2509 * New locales: ps_AF, my_MM
2513 * The following bugs are resolved with this release:
2515 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2516 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2517 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2518 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2519 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2521 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2522 Implemented by Ulrich Drepper.
2524 * New Linux interfaces: accept4, fallocate, fallocate64.
2525 Implemented by Ulrich Drepper.
2527 * Correct declarations of string function when used in C++ code. This
2528 could lead to compile errors for invalid C++ code.
2530 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2533 * New POSIX 2008 interface: psiginfo
2534 Implemented by Ulrich Drepper.
2536 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2537 Implemented by Ulrich Drepper.
2539 * Support for selecting between multiple function definitions at runtime
2540 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2542 * The libcrypt library can now use the hash function implementations in
2543 NSS. Implemented by Ulrich Drepper.
2545 * The malloc implementation can be compiled to be less memory efficient
2546 but higher performing in multi-threaded programs.
2547 Implemented by Ulrich Drepper.
2549 * New locales: nan_TW@latin, ks_IN
2551 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2552 Implemented by Ulrich Drepper.
2554 * Extended printf hook support. It is possible to use user-defined types
2555 and extend existing format specifiers.
2556 Implemented by Ulrich Drepper.
2558 * Handling for group shadow files has been added.
2559 Implemented by Ulrich Drepper.
2561 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2562 servers (the case, e.g., for some people using the built-in DNS
2563 server in ADSL modems/routers). There is a once-per-process timeout
2564 in case of a broken server. To avoid it, users can run nscd or put
2565 'options single-request' in /etc/resolv.conf.
2566 Implemented by Ulrich Drepper.
2570 * The following bugs are resolved with this release:
2572 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2573 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2574 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2575 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2576 6968, 6974, 6980, 6995, 7008, 7009, 7029
2578 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2579 up at the same time. Implemented by Ulrich Drepper.
2581 * TLS descriptors for LD and GD on x86 and x86-64.
2582 Implemented by Alexandre Oliva.
2584 * getaddrinfo now handles DCCP and UDPlite.
2585 Implemented by Ulrich Drepper.
2587 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2588 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2589 Implemented by Ulrich Drepper.
2591 * New implementation of memmem, strstr, and strcasestr which is O(n).
2592 Implemented by Eric Blake.
2594 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2596 * Implement "e" option for popen to open file descriptor with the
2597 close-on-exec flag set. Implemented by Ulrich Drepper.
2599 * Implement "b" mode for fmemopen. In this mode writes writes don't
2600 implicitly add a NUL byte and seeks from the end of the buffer really
2601 use the buffer end, not the string length as the basis.
2602 Implemented by Ulrich Drepper.
2604 * Many functions, exported and internal, now atomically set the close-on-exec
2605 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2607 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2609 Implemented by Pravin Satpute.
2611 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2615 * The following bugs are resolved with this release:
2617 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2618 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2619 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2620 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2621 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2622 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2623 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2625 * New locales: bo_CN, bo_IN, shs_CA.
2627 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2629 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2630 Implemented by Pravin Satpute.
2632 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2634 * nscd caches now all timeouts for DNS entries
2635 Implemented by Ulrich Drepper.
2637 * nscd is more efficient and wakes up less often.
2638 Implemented by Ulrich Drepper.
2640 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2641 vdprintf, and obstack_vprintf.
2642 Implemented by Jakub Jelinek.
2644 * Faster memset for x86-64.
2645 Implemented by Harsha Jagasia and H.J. Lu.
2647 * Faster memcpy on x86.
2648 Implemented by Ulrich Drepper.
2650 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2651 Implemented by Ulrich Drepper.
2653 * Faster sqrt and sqrtf implemention for some PPC variants.
2654 Implemented by Steven Munroe.
2658 * The following bugs are resolved with this release:
2660 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2661 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2662 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2663 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2664 5071, 5103, 5104, 5112, 5113, 5184, 5186
2666 * More checking functions: fread, fread_unlocked, open*, mq_open.
2667 Implemented by Jakub Jelinek and Ulrich Drepper.
2669 * Extend fortification to C++. Implemented by Jakub Jelinek.
2671 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2672 by not recognizing 'a' as a modifier when those specs are requested.
2673 Implemented by Jakub Jelinek.
2675 * PPC optimizations to math and string functions.
2676 Implemented by Steven Munroe.
2678 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2679 options to be passed. Implemented by Ulrich Drepper.
2681 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2683 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2684 Implemented by Ulrich Drepper.
2686 * Handle private futexes in the NPTL implementation.
2687 Implemented by Jakub Jelinek and Ulrich Drepper.
2689 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2690 Implemented by Roland McGrath and Ulrich Drepper.
2692 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2694 * SHA-256 and SHA-512 based password encryption.
2695 Implemented by Ulrich Drepper.
2697 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2698 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2701 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2702 Implemented by Ulrich Drepper.
2706 * The following bugs are resolved with this release:
2708 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2709 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2710 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2711 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2712 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2713 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2714 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2717 * New Linux interfaces: epoll_pwait, sched_getcpu.
2719 * New generic interfaces: strerror_l.
2721 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2725 * The following bugs are resolved with this release:
2727 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2728 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2729 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2730 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2731 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2732 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2733 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2734 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2735 3225, 3251, 3252, 3253, 3273
2737 * For Linux, the sorting of addresses returned by getaddrinfo now also
2738 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2739 Implemented by Ulrich Drepper.
2741 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2742 Implemented by Ulrich Drepper.
2744 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2746 * New iconv module for MIK. Contributed by Alexander Shopov.
2748 * For sites with broken group and/or passwd database, the auto-propagate
2749 option of nscd can prevent creating ID lookup entries from the results
2750 of a name lookup and vice versa. This usually is no problem but some
2751 site might have problems with the default behavior.
2752 Implemented by Ulrich Drepper.
2754 * Iterating over entire database in NIS can be slow. With the
2755 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2756 to trade time for memory. The entire database will be read at once.
2757 Implemented by Ulrich Drepper.
2759 * The interfaces introduced in RFC 3542 have been implemented by
2762 * Support for the new ELF hash table format was added by Ulrich Drepper.
2764 * Support for priority inheritance mutexes added by Jakub Jelinek and
2767 * Support for priority protected mutexes added by Jakub Jelinek.
2769 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2770 or_IN, csb_PL, fy_NL, sr_ME.
2774 * More overflow detection functions.
2776 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2777 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2779 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2780 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2781 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2782 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2783 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2784 by Masahide Washizawa.
2786 * It is now possible to install audit libraries for the dynamic linker using
2787 LD_AUDIT. Implemented by Ulrich Drepper.
2789 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2790 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2791 For a libc and libpthread that works well on Linux 2.4 kernels, we
2792 recommend using the stable 2.3 branch.
2794 * The new function open_wmemstream acts like open_memstream,
2795 but uses a wchar_t wide character buffer.
2797 * The new function ppoll is an improved version of poll, similar to pselect.
2799 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2800 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2801 renameat, unlinkat, linkat, symlinkat, readlinkat.
2803 * New Linux kernel system calls: unshare,
2804 inotify_init, inotify_add_watch, inotify_rm_watch.
2806 * The euidaccess function is now also known by the alias eaccess,
2807 for compatibility with some other systems.
2809 * Timezone data updated to 2006b version.
2813 * The following bugs are resolved with this release:
2815 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2816 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2817 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2818 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2819 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2820 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2822 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2824 * As of this release, GCC 4 can be used to compile the C Library.
2826 * Timezone data updated to 2005m version.
2830 * The following bugs are resolved with this release:
2832 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2833 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2834 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2835 777, 787, 821, 822, 823, 825
2837 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2841 * Support for RFC 3678. Real implementations exist only for Linux so far.
2842 Implemented by Ulrich Drepper.
2844 * nscd can now cache entries persistently. Expiring entries are reloaded.
2845 For speedups the cache can be shared in memory with client processes.
2846 Implemented by Ulrich Drepper.
2848 * nscd can now perform SELinux checks.
2849 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2851 * getaddrinfo queries are now cached. Canonical name lookup is performed
2853 Implemented by Ulrich Drepper.
2855 * The nothrow function attribute is used when headers are used by gcc when
2856 compiling C code. This can avoid significant amounts of exception
2859 * The malloc functions perform more error checking and are stricter when
2860 it comes to reacting on errors. The default action is to terminate
2861 the process after showing an error message. Implemented by Ulrich Drepper.
2863 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2864 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2866 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2867 namespaces using the new function dlmopen(). This feature is of course,
2868 like most other dynamic loading functionality, not available in statically
2869 linked applications. Implemented by Ulrich Drepper.
2871 * Low-overhead boundary checking variants of string and some stdio functions
2872 were added. These are to be used in conjunction with a gcc patch by
2873 Jakub Jelinek which adds calls to these functions if possible.
2874 Implemented by Jakub Jelinek and Ulrich Drepper.
2876 * Old code for several operating systems and machine architectures that
2877 have not been in working condition in a long time have been removed from
2878 the main source tree maintained by the GNU C Library's maintainers.
2879 These files are now reside in the separate `ports' source module
2880 that is usable as an add-on when building the library.
2884 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2885 interrogate the dynamic linker, compatible with the Solaris interface.
2887 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2888 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2890 * getifaddrs now uses the netlink interface on Linux to get its information.
2891 Implemented by Thorsten Kukuk.
2893 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2894 Implemented by Ulrich Drepper.
2896 * support for non-executable stacks on x86 has been added. Changes mostly
2899 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2902 * getaddrinfo now performs destination address selection according to
2907 * Thread-safe interfaces for many functions that access locale data
2908 were added in version 2.3, but these features were omitted from NEWS.
2909 Many functions have variants with an `_l' suffix that take a `locale_t'
2910 object as a parameter rather than consulting the current locale.
2911 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2912 create and maintain `locale_t' objects. Additionally, the new function
2913 `uselocale' sets "the current locale" (as used by functions not so
2914 parameterized) set for an individual thread. These features were added
2915 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2917 * The functions getresuid, getresgid, setresuid, and setresgid, which
2918 have long been available on Linux, are now declared in <unistd.h>
2919 and are now also available on the Hurd.
2921 * ELF thread-local storage support (TLS) now works on x86-64.
2923 * The new dynamic string token $LIB is expanded in shared library names.
2924 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2926 * Aldy Hernandez contributed complete software floating point support for
2927 PowerPC machines with no FPU.
2929 * fexecve is implemented on Linux.
2931 * The `btowc' function should work at least twice as fast due to
2932 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2934 * With approriate thread add-ons cancelable functions are now implemented
2935 in libc.so as well. No need to call the function in libpthread. This
2936 change allowed to finally disable the incorrect and expensive handling
2937 of weak definition in ld.so.
2939 * Yet more PLT entries in libc.so have been removed. We finally arrived
2940 at the bare minimum. Startup times improved appropriately.
2942 * Support for the new Linux/x86 system call interface was added. The
2943 AT_SYSINFO auxiliary vector entry is recognized and handled.
2947 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2950 * iconv (the program and the interface) now accepts empty names (excluding
2951 options like //TRANSLIT) to mean "use charset of current locale".
2953 * localedef can now transliterate characters in strings which are not in
2954 the provided charmap. The information from the input locale is used.
2956 * Prelinking support was added for ELF targets. This requires additional
2957 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2959 * Read-only stdio streams now use mmap to speed up operation by eliminating
2960 copying and buffer underflows. To use add 'm' to the mode string of
2961 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2963 * The malloc functions were completely rewritten by Wolfram Gloger based
2964 on Doug Lea's malloc-2.7.0.c.
2966 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2967 implementation of regex.
2969 * Bruno Haible upgraded the iconv and locale implementation to support
2972 * Contents of the LC_* and LANG environment variables in the CEN style are
2973 not recognized anymore. It never was used. Change by Ulrich Drepper.
2975 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2976 thread-local storage (TLS) ABI on some platforms.
2977 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2979 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2980 EUC-JISX0213, and TSCII.
2982 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2983 BSD-compatible interface for getting all network interface addresses.
2984 Implementation for IPv4 by Roland McGrath.
2986 * Loading of locale data is faster due to the introduction of a locale
2987 archive. Implemented by Roland McGrath and Ulrich Drepper.
2989 * Startup times are significantly reduced by not using exported functions
2990 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2993 * Steven Munroe contributed a port to PowerPC64/Linux.
2997 * The Hurd now uses the GNU libio implementation of stdio rather than the
2998 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3000 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3001 with the same behavior as the Linux system call of the same name.
3005 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3006 128-bit long double format.
3008 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3009 IBM1161, and IBM1162 charsets.
3011 * Andreas Jaeger contributed a port to x86-64/Linux.
3013 * Peter Bruin contributed a port to PowerPC/Hurd.
3015 * libc functions using I/O streams now can handle wide-oriented streams
3018 * optimizations in the dynamic linker. Binaries created by recent binutils
3019 versions start up quicker due to reduced time spent on relocations.
3021 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3025 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3026 asinh, atanh, j0 for the 128-bit long double format.
3028 * Bruno Haible updated all the code handling Unicode in some form to
3029 support Unicode 3.1.
3031 * Speed of regex for single-byte locales is back to previous levels.
3032 Patch by Isamu Hasegawa.
3034 * Alpha, SPARC, and IA-64 now also using floating stacks.
3036 * Startup time of internationalized applications greatly improved through
3037 iconv cache. Use iconvconfig to generate the cache file.
3038 Contributed by Ulrich Drepper.
3040 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3041 severe performance problems. Patch by David Mosberger.
3043 * The Hurd port got a lot more functionality like AIO, various stdio
3044 extensions, etc. Mainly done by Roland McGrath.
3046 * mtrace can now lookup symbols in shared libraries.
3050 * Intel's IA-64 math library is largely integrated. It provides fast and
3051 accurate implementatations for most basic and standard math functions
3052 in float, double, and long double format.
3054 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3055 and asin for the 96-bit long double format and asin, log, tan for the
3056 128-bit long double format.
3058 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3059 The basic double functions exist today. Contributed by Abraham Ziv
3060 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3061 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3063 * An asynchronous name lookup library was added. The interface is designed
3064 after POSIX AIO. The proposal was circulated beforehand to get comments.
3065 No negative ones came in. Implemented by Ulrich Drepper.
3067 * Port to S390/64bit contributed by Martin Schwidefsky
3068 <schwidefsky@de.ibm.com>.
3070 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3071 of functions for Linux/IA-64.
3073 * The RPC code is now thread safe. Threads can now use the same service
3074 of different services at the same time. Patch by Eric Norum
3075 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3077 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3078 family of functions for Linux/S390.
3080 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3081 of functions for Linux/x86.
3083 * Port to Linux/CRIS contributed by Axis Communications.
3087 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3088 we can now check for namespace violations and missing declarations. The
3089 result is that almost all headers are now Unix-compliant (as defined in
3090 the upcoming XPG6). The negative side is that some programs might need
3091 corrections, too, if they depend on the incorrect form of the headers in
3092 previous versions which defined too many symbols and included too many
3095 * regex now handles multibyte character sets correctly.
3096 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3098 * iconv (the program) does now conform to the upcoming XPG6 and handles
3099 charmaps. Instead of the charset names the path of charmaps can be
3100 provided and the conversion happens based on this data.
3101 Contributed by Ulrich Drepper.
3103 * The locale program now provides detailed information about the installed
3108 only lists the names of the supported locales
3112 provides details such as country, language, and codeset name.
3113 Contributed by Ulrich Drepper.
3117 * The gencat program now parses the input file according to the charset
3118 selected by the LC_CTYPE category. This is important for stateful
3119 character sets. To make generating catalogs easier there is a way
3120 to overwrite the charset selected by the locale: before the first
3121 message or $ quote line the catalog can contain a line like
3123 $ codeset=ISO-8859-2
3125 to select the charset (ISO-8859-2 in this case).
3127 Implemented by Shinya Hanataka and Ulrich Drepper.
3129 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3130 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3131 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3133 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3134 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3136 * The LANGUAGE environment variable is now ignored unless the locale is
3137 changed from the default "C" locale.
3139 * The usual bug fixes.
3143 * Greg McGary added runtime support for bounds checking using gcc's
3144 new -fbounded-pointers option. ix86 target is complete. PowerPC
3147 * Thorsten Kukuk added secure mode support to nscd.
3149 * The Berkeley db libraries have been removed.
3151 Related, the nss_db module is now in a separate package since it
3152 obviously requires a database library being available.
3154 * Wide character I/O streams implemented by Ulrich Drepper.
3156 * Functions from the extended socket API added by Ulrich Drepper.
3158 * Functions feenableexcept and fedisableexcept to control the
3159 behaviour of individual exceptions have been added by Andreas Jaeger.
3161 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3163 * The resolver code has been updated from bind 8.2.3-T5B which supports
3164 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3167 This change could in some situations effect backward compatibility. Since
3168 now `_res' is a thread-local instead of a global variable, modifying it
3169 in one thread does not have any effect in other threads.
3171 The resolver library was also extended to allow IPv6 as the transport
3172 protocol for the requests. This work was done by Stig Venaas.
3174 * Compatibility code for K&R C compilers has been removed from the
3175 header files. A ISO C compiler is needed to use the library
3176 (conforming to either C89 or C99 standard).
3178 * Complete rewrite of the localedef program to support multibyte character
3179 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3180 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3181 Implemented by Ulrich Drepper.
3183 Bruno Haible significantly improved the generation and use of the data
3184 structures for the wide character tables.
3186 * Plural handling in gettext implemented by Ulrich Drepper.
3188 * The utmp daemon has been removed.
3190 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3192 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3195 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3197 * POSIX barriers implemented by Kaz Kylheku.
3199 * POSIX spawn function family implemented by Ulrich Drepper.
3201 * POSIX spinlocks are now available.
3203 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3205 * the configure option --enable-kernel=X.Y.Z allows to strip out
3206 compatibility for kernel versions before X.Y.Z. This is currently only
3207 implemented for Linux.
3209 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3210 field sin6_scope_id. This means that all programs using IPv6 should be
3211 recompiled. Don't expect binary compatibility with previous glibc
3214 * various conversion modules for IBM character sets contributed by
3217 * IA-64 port by Jes Sorensen and HJ Lu.
3231 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3233 * New xdr functions are added; some rpc functions are now 64bit clean.
3235 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3237 * Fixed known incompatibilities with glibc 2.0.
3239 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3241 * Optimized a number of functions (especially the ELF dynamic loader).
3243 * Update timezone data files.
3245 * lots of charmaps corrections
3247 * some new locale definitions and charmaps
3252 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3253 conform to POSIX member type requirements. Symbol versions have been
3254 adjusted as needed within the library, and for direct use by applications,
3255 but there is potential for problems if third-party libraries use
3256 struct timeval as part of their interface. This does not present
3257 a problem for X and other "essential" system libraries.
3259 * An additional locale model to support C++ Standard Library locale
3260 model and probably more was implemented by Ulrich Drepper.
3262 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3265 * Miles Bader provided the `argp' function family to support hierachical
3266 command line argument parsing, layered on top of getopt.
3268 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3270 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3273 * scanf recognizes the %a and %A format for scanning floating point numbers.
3275 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3276 information and interfaces for the available integer types.
3278 * about 130 new math functions were added to implement the ISO C9x math
3281 * the new header <complex.h> contains definitions of the complex math
3282 functions from ISO C 9X.
3284 * the new header <tgmath.h> defines generic macros to use complex or
3285 real valued functions.
3287 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3289 * Andreas Jaeger provided a test suite for the math library.
3291 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3293 * Ulrich Drepper added character set conversion functions (iconv).
3295 * Optimized string functions have been added.
3297 * The localedata addon is now part of glibc.
3299 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3301 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3302 daemon for NSS (nscd).
3304 Missing a better place here are some numbers on improvements. Under
3305 Linux 2.1.125 un-tar-ing the kernel sources takes
3309 using local files 12.19s 6.88s 22.91s
3311 using NIS 13.92s 8.91s 26.34s
3313 using NIS & nscd 10.37s 7.34s 25.30s
3315 using NIS+ 27.57s 30.37s 640.46s
3317 using NIS+ & nscd 10.25s 7.83s 26.51s
3319 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3321 Keep in mind that non-namelookup related operations dominate above times.
3322 It was just a common complain that using NIS+ unpacking the kernel is
3325 [1] The old nscd implementation is not available anymore since it was
3326 distributed with glibc up to version 2.0.98 and thus is now replaced.
3328 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3330 * Mark Kettenis provided a Hesiod NSS module.
3332 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3333 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3335 * rcmd can now handle netgroups (Dick Streefland).
3337 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3340 * Support for the IPv6 protocol has been added to the socket API, as per the
3341 latest draft standards.
3343 * Support for Linux 2.2 has been added.
3345 * Interface changes relative to the latest 2.0.x release:
3346 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3347 addseverity NEW: Unix98
3348 alphasort64 NEW: LFS
3349 argp_err_exit_status NEW: argp, GNU ext
3350 argp_error NEW: argp, GNU ext
3351 argp_failure NEW: argp, GNU ext
3352 argp_help NEW: argp, GNU ext
3353 argp_parse NEW: argp, GNU ext
3354 argp_program_bug_address NEW: argp, GNU ext
3355 argp_program_version NEW: argp, GNU ext
3356 argp_program_version_hook NEW: argp, GNU ext
3357 argp_state_help NEW: argp, GNU ext
3358 argp_usage NEW: argp, GNU ext
3359 authdes_create NEW: Secure RPC
3360 authdes_getucred NEW: Secure RPC
3361 authdes_pk_create NEW: Secure RPC
3362 backtrace NEW: GNU ext.
3363 backtrace_symbols NEW: GNU ext.
3364 backtrace_symbols_fd NEW: GNU ext.
3366 cacosf NEW: ISO C 9x
3367 cacosh NEW: ISO C 9x
3368 cacoshf NEW: ISO C 9x
3369 cacoshl NEW: ISO C 9x
3370 cacosl NEW: ISO C 9x
3377 casinf NEW: ISO C 9x
3378 casinh NEW: ISO C 9x
3379 casinhf NEW: ISO C 9x
3380 casinhl NEW: ISO C 9x
3381 casinl NEW: ISO C 9x
3383 catanf NEW: ISO C 9x
3384 catanh NEW: ISO C 9x
3385 catanhf NEW: ISO C 9x
3386 catanhl NEW: ISO C 9x
3387 catanl NEW: ISO C 9x
3388 cbc_crypt NEW: Secure RPC
3392 ccoshf NEW: ISO C 9x
3393 ccoshl NEW: ISO C 9x
3399 cimagf NEW: ISO C 9x
3400 cimagl NEW: ISO C 9x
3401 clearerr_locked REMOVED
3402 clntunix_create NEW: sunrpc ext
3404 clog10 NEW: ISO C 9x
3405 clog10f NEW: ISO C 9x
3406 clog10l NEW: ISO C 9x
3416 cprojf NEW: ISO C 9x
3417 cprojl NEW: ISO C 9x
3419 crealf NEW: ISO C 9x
3420 creall NEW: ISO C 9x
3425 csinhf NEW: ISO C 9x
3426 csinhl NEW: ISO C 9x
3429 csqrtf NEW: ISO C 9x
3430 csqrtl NEW: ISO C 9x
3434 ctanhf NEW: ISO C 9x
3435 ctanhl NEW: ISO C 9x
3437 des_setparity NEW: Secure RPC
3438 ecb_crypt NEW: Secure RPC
3439 endutxent NEW: Unix98
3441 exp10f NEW: ISO C 9x
3442 exp10l NEW: ISO C 9x
3446 fattach NEW: STREAMS
3447 fdetach NEW: STREAMS
3451 feclearexcept NEW: ISO C 9x
3452 fegetenv NEW: ISO C 9x
3453 fegetexceptflag NEW: ISO C 9x
3454 fegetround NEW: ISO C 9x
3455 feholdexcept NEW: ISO C 9x
3457 feraiseexcept NEW: ISO C 9x
3458 ferror_locked REMOVED
3459 fesetenv NEW: ISO C 9x
3460 fesetexceptflag NEW: ISO C 9x
3461 fesetround NEW: ISO C 9x
3462 fetestexcept NEW: ISO C 9x
3463 feupdateenv NEW: ISO C 9x
3464 fflush_locked REMOVED
3468 fgets_unlocked NEW: GNU ext.
3469 fileno_locked REMOVED
3481 fputc_locked REMOVED
3482 fputs_unlocked NEW: GNU ext.
3483 fread_unlocked NEW: GNU ext.
3488 fstatvfs NEW: Unix98
3492 ftruncate64 NEW: LFS
3494 fwrite_unlocked NEW: GNU ext.
3495 gai_strerror NEW: IPv6
3499 getchar_locked REMOVED
3501 getdate_err NEW: Unix98
3502 getdate_r NEW: GNU ext.
3504 getnameinfo NEW: IPv6
3505 getnetname NEW: Secure RPC
3506 getpmsg NEW: STREAMS
3507 getpt NEW: Unix98 PTY
3508 getrlimit64 NEW: LFS
3509 getutxent NEW: Unix98
3510 getutxid NEW: Unix98
3511 getutxline NEW: Unix98
3512 glob64 NEW: GNU ext.
3513 globfree64 NEW: GNU ext.
3514 gnu_get_libc_release NEW: GNU ext.
3515 gnu_get_libc_version NEW: GNU ext.
3516 grantpt NEW: Unix98 PTY
3517 host2netname NEW: Secure RPC
3519 iconv_close NEW: iconv
3520 iconv_open NEW: iconv
3521 if_freenameindex NEW: IPv6
3522 if_indextoname NEW: IPv6
3523 if_nameindex NEW: IPv6
3524 if_nametoindex NEW: IPv6
3525 in6addr_any NEW: IPv6
3526 in6addr_loopback NEW: IPv6
3527 inet6_isipv4mapped NEW: IPv6
3528 isastream NEW: STREAMS
3529 iswblank NEW: GNU ext.
3530 key_decryptsession NEW: Secure RPC
3531 key_decryptsession_pk NEW: Secure RPC
3532 key_encryptsession NEW: Secure RPC
3533 key_encryptsession_pk NEW: Secure RPC
3534 key_gendes NEW: Secure RPC
3535 key_get_conv NEW: Secure RPC
3536 key_secretkey_is_set NEW: Secure RPC
3537 key_setnet NEW: Secure RPC
3538 key_setsecret NEW: Secure RPC
3539 llrint NEW: ISO C 9x
3540 llrintf NEW: ISO C 9x
3541 llrintl NEW: ISO C 9x
3542 llround NEW: ISO C 9x
3543 llroundf NEW: ISO C 9x
3544 llroundl NEW: ISO C 9x
3549 lrintf NEW: ISO C 9x
3550 lrintl NEW: ISO C 9x
3551 lround NEW: ISO C 9x
3552 lroundf NEW: ISO C 9x
3553 lroundl NEW: ISO C 9x
3555 makecontext NEW: Unix98
3556 mempcpy NEW: GNU ext.
3559 modify_ldt NEW: kernel
3563 nearbyint NEW: ISO C 9x
3564 nearbyintf NEW: ISO C 9x
3565 nearbyintl NEW: ISO C 9x
3566 netname2host NEW: Secure RPC
3567 netname2user NEW: Secure RPC
3568 nexttoward NEW: ISO C 9x
3569 nexttowardf NEW: ISO C 9x
3570 nexttowardl NEW: ISO C 9x
3574 passwd2des NEW: Secure RPC
3576 pow10f NEW: GNU ext.
3577 pow10l NEW: GNU ext.
3580 printf_size NEW: GNU ext.
3581 printf_size_info NEW: GNU ext.
3582 profil_counter REMOVED
3583 pthread_mutexattr_getkind_np REPLACED
3584 pthread_mutexattr_setkind_np REPLACED
3585 ptsname NEW: Unix98 PTY
3586 ptsname_r NEW: Unix98 PTY
3588 putchar_locked REMOVED
3589 putgrent NEW: GNU ext.
3591 putpmsg NEW: STREAMS
3592 pututxline NEW: Unix98
3596 readdir64_r NEW: LFS
3597 remquo NEW: ISO C 9x
3598 remquof NEW: ISO C 9x
3599 remquol NEW: ISO C 9x
3601 roundf NEW: ISO C 9x
3602 roundl NEW: ISO C 9x
3604 scalbln NEW: ISO C 9x
3605 scalblnf NEW: ISO C 9x
3606 scalblnl NEW: ISO C 9x
3608 sendfile NEW: kernel
3609 setcontext NEW: Unix98
3610 setrlimit64 NEW: LFS
3611 setutxent NEW: Unix98
3613 sigignore NEW: Unix98
3614 sigqueue NEW: POSIX.1b
3615 sigrelse NEW: Unix98
3616 sigset NEW: POSIX.1b
3617 sigtimedwait NEW: POSIX.1b
3618 sigwaitinfo NEW: POSIX.1b
3619 sincos NEW: GNU ext.
3620 sincosf NEW: GNU ext.
3621 sincosl NEW: GNU ext.
3625 strcasestr NEW: GNU ext.
3626 strtoimax NEW: ISO C 9x
3627 strtoumax NEW: ISO C 9x
3628 strverscmp NEW: GNU ext.
3629 svcauthdes_stats NEW: Secure RPC
3630 svcunix_create NEW: sunrpc etx
3631 svcunixfd_create NEW: sunrpc ext
3632 swapcontext NEW: Unix98
3633 tcgetsid NEW: Unix98 PTY
3634 tdestroy NEW: GNU ext.
3635 tgamma NEW: ISO C 9x
3636 tgammaf NEW: ISO C 9x
3637 tgammal NEW: ISO C 9x
3641 truncf NEW: ISO C 9x
3642 truncl NEW: ISO C 9x
3644 unlockpt NEW: Unix98 PTY
3645 updwtmpx NEW: Unix98
3646 user2netname NEW: Secure RPC
3647 utmpxname NEW: Unix98
3648 versionsort NEW: GNU ext.
3649 versionsort64 NEW: GNU ext.
3651 wcscasecmp NEW: GNU ext.
3652 wcsncasecmp NEW: GNU ext.
3653 wcsnlen NEW: GNU ext.
3654 wcstoimax NEW: ISO C 9x
3655 wcstoll NEW: ISO C 9x
3656 wcstoull NEW: ISO C 9x
3657 wcstoumax NEW: ISO C 9x
3659 wordexp NEW: POSIX.2
3660 wordfree NEW: POSIX.2
3661 write_profiling REMOVED
3662 xdecrypt NEW: Secure RPC
3663 xdr_authdes_cred NEW: Secure RPC
3664 xdr_authdes_verf NEW: Secure RPC
3665 xdr_cryptkeyarg NEW: Secure RPC
3666 xdr_cryptkeyarg2 NEW: Secure RPC
3667 xdr_cryptkeyres NEW: Secure RPC
3668 xdr_getcredres NEW: Secure RPC
3669 xdr_key_netstarg NEW: Secure RPC
3670 xdr_key_netstres NEW: Secure RPC
3671 xdr_keybuf NEW: Secure RPC
3672 xdr_keystatus NEW: Secure RPC
3673 xdr_netnamestr NEW: Secure RPC
3674 xdr_sizeof NEW: Secure RPC
3675 xdr_unixcred NEW: sunrpc ext
3676 xencrypt NEW: Secure RPC
3677 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3688 * inet_ntoa is thread-safe
3690 * updwtmp is moved from libutil to libc
3692 * rewrite of cbrt function
3694 * update of timezone data
3708 * add atoll function
3710 * fix complex problems in Berkeley DB code
3712 * fix math functions
3716 * fixed lots of header problems (especially Linux/GNU specific)
3718 * dynamic loader preserves all registers
3720 * Roland McGrath provided support for handling of auxiliary objects in
3721 the ELF dynamic loader.
3723 * support for parallel builds is improved
3727 * GNU extensions are no longer declared by default. To enable them you
3728 must define the macro `_GNU_SOURCE' in your program or compile with
3731 * The library has changed from using GNU ld symbol aliases to using weak
3732 symbols where available. The ELF object file format supports weak
3733 symbols; GNU ld also supports weak symbols in the a.out format. (There
3734 is also now support for other GNU ld extensions in ELF. Use the
3735 `--with-elf' option to configure to indicate you have ELF, and
3736 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3737 of many files which contained only symbol aliases, reducing the size of
3738 the source and the compiled library; many other files were renamed to
3739 less cryptic names previously occupied by the symbol alias files.
3740 There is a new header file <elf.h> for programs which operate on
3741 files in the ELF format.
3743 * Converted to Autoconf version 2, so `configure' has more options.
3744 Run `configure --help' to see the details.
3746 * The library can now be configured to build profiling, highly-optimized
3747 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3748 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3749 `configure' enable building these extra libraries. The shared library is
3750 built by default when using both ELF and GNU ld. When shared libraries
3751 are enabled, the new library `-ldl' is available for arbitrary run-time
3752 loading of shared objects; its interface is defined in <dlfcn.h>. The
3753 new header file <link.h> gives access to the internals of the run-time
3754 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3755 application of same name on other systems and it provides information
3756 about dynamically linked binaries.
3758 * The C library now provides the run-time support code for profiling
3759 executables compiled with `-pg'. Programs can control the profiling code
3760 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3761 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3762 the support for this file format was contributed by David Mosberger-Tang.
3764 * The math code has been replaced with a math library based on fdlibm from
3765 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3766 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3767 functions. The math functions now reside in a separate library, so
3768 programs using them will need to use `-lm' their linking commands.
3770 * John C. Bowman contributed optimized ix87 assembler inline functions.
3772 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3773 mechanism similar to that found in Solaris 2. This is now used for the
3774 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3775 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3776 file controls what services are used for each individual database. This
3777 works by loading shared libraries with names specified in `nsswitch.conf',
3778 so service modules can be changed or added at any time without even
3779 relinking any program. Currently there are the file, db, and NIS based
3780 NSS services available.
3782 * The new functions `strtoq' and `strtouq' parse integer values from
3783 strings, like `strtol' and `strtoul', but they return `long long int' and
3784 `unsigned long long int' values, respectively (64-bit quantities).
3786 * The new functions `strtof' and `strtold' parse floating-point values from
3787 strings, like `strtod', but they return `float' and `long double' values,
3788 respectively (on some machines `double' and `long double' are the same).
3790 * Ulrich Drepper has contributed new implementations of the floating-point
3791 printing and reading code used in the `printf' family of functions and
3792 `strtod', `strtof', and `strtold'. These new functions are perfectly
3793 accurate, and much faster than the old ones.
3795 * The implementation of the POSIX locale model was completely rewritten by
3796 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3797 to compile the POSIX locale definition.
3799 * The former dummy implementations of the strcoll and strxfrm function are
3800 now replaced by fully functional code contributed by Ulrich Drepper. The
3801 collation information comes from the POSIX locale definitions.
3803 * The new header <langinfo.h> defines an interface for accessing
3804 various locale-dependent data (using the locale chosen with `setlocale').
3806 * Ulrich Drepper has contributed a new suite of functions for operation on
3807 wide-character and multibyte-character strings, in <wchar.h>;
3808 and classification and case conversion of wide characters, in <wctype.h>.
3809 These new functions are conforming to the ISO C, Amendement 1 specification.
3811 * There is now a second implementation of the standard I/O library available.
3812 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3813 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3815 * You can now use positional parameter specifications in format strings
3816 for the `printf' and `scanf' families of functions. For example,
3817 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3818 ``Number 6, Mr Jones''. This is mainly useful when providing different
3819 format strings for different languages, whose grammars may dictate
3820 different orderings of the values being printed. To support this
3821 feature, the interface for `register_printf_handler' has changed; see
3822 the header file <printf.h> for details.
3824 * The `printf' and `scanf' families of functions now understand a new
3825 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3826 to group numbers as indicated by the locale; for `scanf' and friends, this
3827 says to accept as valid only a number with all the proper grouping
3828 separators in the right places. In the default "C" locale, numbers are
3829 not grouped; but locales for specific countries will define the usual
3830 conventions (i.e. separate thousands with `,' in the US locale).
3832 * The pgrp functions have been regularized, slightly incompatibly but much
3833 less confusingly. The core functions are now `getpgid' and `setpgid',
3834 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3835 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3836 provided for compatibility. There is no longer an incompatible `getpgrp'
3837 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3839 * The new header file <fts.h> and suite of functions simplify programs that
3840 operate on directory trees. This code comes from 4.4 BSD.
3842 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3843 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3844 NSS scheme used in glibc.
3846 * There is a new malloc debugging hook `__memalign_hook'.
3848 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3849 `unsigned int' in <sys/types.h>. These are for compatibility only and
3850 their use is discouraged.
3852 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3853 done differently, so that it works even without GNU ld.
3855 * New function `euidaccess' checks allowed access to a file like `access',
3856 but using the effective IDs instead of the real IDs.
3858 * The time zone data files have been updated for the latest and greatest
3859 local time conventions of the countries of the world.
3861 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3864 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3865 for formatting floating-point numbers. They are provided only for
3866 compatibility; new programs should use `sprintf' instead. There are
3867 also equivalent function for the `long double' floating-point type and
3868 all functions also exist in a reentrant form.
3870 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3871 functions for maintaining the login-record files (primarily of use to
3872 system programs such as `login'), and convenient functions for
3873 allocating and initializing a pseudo-terminal (pty) device.
3875 * Ulrich Drepper has contributed new support for System V style
3876 shared memory and IPC on systems that support it.
3878 * Ulrich Drepper has contributed several miscellaneous new functions found
3879 in System V: The `hsearch' family of functions provide an effective
3880 implementation of hash tables; `a64l' and `l64a' provide a very simple
3881 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3884 * Ulrich Drepper has contributed new reentrant counterparts for the
3885 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3887 * Ulrich Drepper has contributed new, highly-optimized versions of several
3888 string functions for the i486/Pentium family of processors.
3890 * Ulrich Drepper has updated the Linux-specific code, based largely
3891 on work done in Hongjiu Lu's version of GNU libc for Linux.
3892 The GNU library now supports Linux versions 2.0.10 and later,
3893 using the ELF object file format (i[3456]86-*-linux).
3895 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3897 * David Mosberger-Tang and Richard Henderson have ported the C library
3898 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3899 dynamic linking support for ELF/Alpha.
3901 * Richard Henderson contributed several Alpha optimized assembler function
3902 for arithmetic and string handling.
3904 * Ulrich Drepper has contributed a new set of message catalog functions to
3905 support multiple languages using the <libintl.h> interface, for use with
3906 his new package GNU gettext. Translation volunteers have contributed
3907 catalogs of the library's messages in Spanish, German, and Korean.
3909 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3910 program and the `catgets' function for reading the catalog files it
3911 creates. (The <libintl.h> interface is preferred; we include the
3912 <nl_types.h> interface using `catgets' only for source compatibility with
3913 programs already written to use it.)
3915 * New header file <values.h> gives SVID-compatible names for <limits.h>
3918 * Various new macros, declarations, and small header files for compatibility
3921 * New function `group_member' is a convenient way to check if a process has
3922 a given effective group ID.
3924 * When using GCC 2.7 and later, the socket functions are now declared in a
3925 special way so that passing an argument of type `struct sockaddr_in *',
3926 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3927 `struct sockaddr *' type, does not generate a type-clash warning.
3929 * New function `error' declared in header file <error.h> is a convenient
3930 function for printing error messages and optionally exiting; this is the
3931 canonical function used in GNU programs. The new functions `err', `warn',
3932 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3933 doing the same thing.
3935 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3936 POSIX.2 `glob' function to do ~ and {...} expansion.
3938 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3939 `clearenv' which is used in POSIX.9 is also available.
3941 * New function `getsid' returns session ID number on systems that support it.
3943 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3944 files <db.h> and <mpool.h> provide a rich set of functions for several
3945 types of simple databases stored in memory and in files, and <ndbm.h> is
3946 an old `ndbm'-compatible interface using the `db' functions. Link with
3947 `-ldb' to get these functions.
3949 * New macro `strdupa' copies a string like `strdup', but uses local stack
3950 space from `alloca' instead of dynamic heap space from `malloc'.
3952 * New function `strnlen' is like `strlen' but searches only a given maximum
3953 number of characters for the null terminator. `stpncpy', `strndup' and
3954 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3957 * New function `statfs' in header <sys/statfs.h>.
3959 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3960 provide convenient functions for operating on blocks of null-terminated
3963 * A new suite of functions in <utmp.h> handle all the details of reading
3964 and writing the utmp file.
3966 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3969 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3970 completely POSIX compliant and also implemented the extended functionality
3971 to handle alternate digit representation and alternate era date formats.
3973 * Ulrich Drepper provided an implementation of the `strptime' function
3974 defined in XPG4.2 which transforms a string into a `struct tm' value.
3976 * Paul Eggert provided the tzselect shell script as part of the timezone
3977 code. The shell script makes it easy to select the correct timezone
3980 * The implementation of the malloc family of functions is completely replaced
3981 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3982 The implementation uses the mmap function (if available) and it is
3983 optimized for the use in multi threaded programs.
3985 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3986 functions. This new functionality is usable by specifying a special
3987 salt string and it is compatible with implementation on *BSD systems.
3989 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3990 `getsubopt' to handle second level command line options, `bsd_signal'
3991 to access BSD style `signal' functionality, the obsolete `regexp' style
3994 * the `lchown' function is available on system which support this
3997 * The implementation of the shadow password handling function was contributed
4000 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4002 * POSIX.1g support was added. The <sys/select.h> header is available,
4003 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4004 implementation of `getaddrinfo'.
4008 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4010 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4011 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4013 * The new function `daemon' from 4.4 BSD is useful for server programs that
4014 want to put themselves in the background.
4016 * Joel Sherrill has contributed support for several standalone boards that
4017 run without an operating system.
4019 * `printf', `scanf' and friends now accept a `q' type modifier for long
4020 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4022 * All of the code taken from BSD (notably most of the math and networking
4023 routines) has been updated from the BSD 4.4-Lite release.
4025 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4027 * The new functions `getdomainname' and `setdomainname' fetch or change the
4028 YP/NIS domain name. These are system calls which exist on systems which
4031 * The time zone data files have been updated for the latest international
4034 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4035 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4039 * The C library now includes support for Sun RPC, from Sun's free
4040 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4041 are included. (There is still no support for YP.)
4043 * Tom Quinn has contributed a port of the C library to SGI machines running
4044 Irix 4 (mips-sgi-irix4).
4046 * The new `lockf' function is a simplified interface to the locking
4047 facilities of `fcntl', included for compatibility.
4049 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4051 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4053 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4056 * The error message printed by `assert' for a failed assertion now includes
4057 the name of the program (if using GNU ld) and the name of the calling
4058 function (with versions of GCC that support this).
4060 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4062 * The library now includes the <sys/mman.h> header file and memory
4063 management functions `mmap', `munmap', `mprotect', `msync', and
4064 `madvise', on systems that support those facilities.
4066 * The interface for `mcheck' has changed slightly: the function called to
4067 abort the program when an allocation inconsistency is detected now takes
4068 an argument that indicates the type of failure. The new function
4069 `mprobe' lets you request a consistency check for a particular block at
4070 any time (checks are normally done only when you call `free' or `realloc'
4073 * It is now possible to easily cross-compile the C library, building on one
4074 system a library to run on another machine and/or operating system. All
4075 you need to do is set the variable `HOST_CC' in `configparms' to the
4076 native compiler for programs to run on the machine you are building on (a
4077 few generator programs are used on Unix systems); set `CC' to the
4080 * The new function `fexecve' (only implemented on the GNU system) executes
4081 a program file given a file descriptor already open on the file.
4085 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4086 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4088 * You can set the variable `libprefix' in `configparms' to specify a prefix
4089 to be prepended to installed library files; this makes it easy to install
4090 the GNU C library to be linked as `-lgnuc' or whatever.
4092 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4093 copies a limited number of characters from a string, and returns the
4094 address of the last character written.
4096 * You no longer need to check for whether the installed `stddef.h' is
4097 compatible with the GNU C library. configure now checks for you.
4099 * You can now define a per-stream `fileno' function to convert the
4100 stream's cookie into an integral file descriptor.
4102 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4103 allocates zero bytes of storage, and returns a unique pointer which
4104 you can pass to `realloc' or `free'. The behavior is undefined if
4105 you dereference this pointer.
4107 * The C library now runs on Sony NEWS m68k machines running either
4108 NewsOS 3 or NewsOS 4.
4110 * The new `syscall' function is a system-dependent primitive function
4111 for invoking system calls. It has the canonical behavior on Unix
4112 systems, including unreliable return values for some calls (such as
4113 `pipe', `fork' and `getppid').
4115 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4116 to `EAGAIN', which is the preferred name. On systems whose kernels
4117 use two distinct codes, the C library now translates EWOULDBLOCK to
4118 EAGAIN in every system call function.
4122 * The GNU C Library Reference Manual is now distributed with the library.
4123 `make dvi' will produce a DVI file of the printed manual.
4124 `make info' will produce Info files that you can read on line using C-h i
4125 in Emacs or the `info' program.
4126 Please send comments on the manual to bug-glibc-manual@gnu.org.
4128 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4130 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4132 * Jason Merrill has contributed a port to the Sequent Symmetry running
4133 Dynix version 3 (i386-sequent-dynix).
4135 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4136 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4138 * New function `memory_warnings' lets you arrange to get warnings when
4139 malloc is running out of memory to allocate, like Emacs gives you.
4141 * The C library now contains the relocating allocator used in Emacs 19 for
4142 its editing buffers. This allocator (ralloc) minimizes allocation
4143 overhead and fragmentation by moving allocated regions around whenever it
4144 needs to. You always refer to a ralloc'd region with a "handle" (a
4145 pointer to a pointer--an object of type `void **').
4147 * There is a new `printf' format: `%m' gives you the string corresponding
4148 to the error code in `errno'.
4150 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4151 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4152 pass, the `a' modifier says to fill in a `char **' you pass with a
4155 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4156 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4157 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4159 * `mkstemp' is a traditional Unix function to atomically create and open a
4160 uniquely-named temporary file.
4164 * The standard location for the file that says what the local timezone is
4165 has changed again. It is now `/usr/local/etc/localtime' (or more
4166 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4168 * The distribution no longer contains any files with names longer than 14
4171 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4172 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4174 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4178 * The configuration process has changed quite a bit. The `configure'
4179 script is now used just like the configuration scripts for other GNU
4180 packages. The `sysdeps' directory hierarchy is much rearranged.
4181 The file `INSTALL' explains the new scheme in detail.
4183 * The header files no longer need to be processed into ANSI C and
4184 traditional C versions. There is just one set of files to install, and
4185 it will work with ANSI or old C compilers (including `gcc -traditional').
4187 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4188 MIPS DECStation running Ultrix 4.
4190 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4191 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4192 made itself into a shared library.
4194 * Yet further improved support for the i386, running 4.3 BSD-like systems
4195 (such as Mach 3 with the Unix single-server), or System V.
4197 * New function `strncasecmp' to do case-insensitive string comparison
4198 with limited length.
4200 * New function `strsep' is a reentrant alternative to `strtok'.
4202 * New functions `scandir' and `alphasort' for searching directories.
4204 * New function `setenv' is a better interface to `putenv'.
4206 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4207 function for traversing a directory tree.
4209 * The GNU obstack package is now also part of the C library.
4210 The new function `open_obstack_stream' creates a stdio stream that
4211 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4212 formatted output directly to an obstack.
4214 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4215 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4217 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4219 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4220 things to your strings.
4222 * There are some new test programs: test-fseek, testmb, and testrand.
4224 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4225 These ports are not finished, but are a good starting place for really
4226 supporting those systems.
4228 * `/etc/localtime' is now the standard location for the file that says what
4229 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4230 This follows the general principle that `/etc' is the place for all local
4231 configuration files.
4233 * The C library header files now use `extern "C"' when used by the C++
4234 compiler, so the C library should now work with C++ code.
4236 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4237 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4240 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4241 function which uses the `merge sort' algorithm, and is said to be
4242 significantly faster than the old GNU `qsort' function. Merge sort is
4243 now the standard `qsort' function. The new algorithm can require a lot
4244 of temporary storage; so, the old sorting function is called when the
4245 required storage is not available.
4247 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4248 provides the Unix `crypt' function, plus some other entry points.
4250 * The code and header files taken from 4.4 BSD have been updated with the
4251 latest files released from Berkeley.
4253 ----------------------------------------------------------------------
4254 Copyright information:
4256 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4258 Permission is granted to anyone to make or distribute verbatim copies
4259 of this document as received, in any medium, provided that the
4260 copyright notice and this permission notice are preserved,
4261 thus giving the recipient permission to redistribute in turn.
4263 Permission is granted to distribute modified versions
4264 of this document, or of portions of it,
4265 under the above conditions, provided also that they
4266 carry prominent notices stating who last changed them.
4269 version-control: never