]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
hr_HR locale: Don’t use single code points for the digraphs in LC_TIME
[thirdparty/glibc.git] / NEWS
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.
4
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
7 \f
8 Version 2.27
9
10 Major new features:
11
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.
14
15 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
16
17 * Optimized generic expf, exp2f, logf, log2f and powf.
18
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.
26
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.
31
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.
37
38 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
39 mips64, powerpc64le, s390, sparc and x86_64), the math library now
40 implements interfaces for that type, as defined by ISO/IEC TS
41 18661-3:2015. These are corresponding interfaces to those supported for
42 _Float128.
43
44 * glibc now implements the memfd_create and mlock2 functions on Linux.
45
46 Deprecated and removed features, and other changes affecting compatibility:
47
48 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
49 defined by <sys/ptrace.h>.
50
51 * libm no longer supports SVID error handling (calling a user-provided
52 matherr function on error) or the _LIB_VERSION variable to control error
53 handling. (SVID error handling and the _LIB_VERSION variable still work
54 for binaries linked against older versions of the GNU C Library.) The
55 libieee.a library is no longer provided. math.h no longer defines struct
56 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
57 TLOSS, PLOSS and HUGE.
58
59 * The libm functions pow10, pow10f and pow10l are no longer supported for
60 new programs. Programs should use the standard names exp10, exp10f and
61 exp10l for these functions instead.
62
63 * The mcontext_t type is no longer the same as struct sigcontext. On
64 platforms where it was previously the same, this changes the C++ name
65 mangling for interfaces involving this type.
66
67 * The add-ons mechanism for building additional packages at the same time as
68 glibc has been removed. The --enable-add-ons configure option is now
69 ignored.
70
71 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
72 the specified string can be parsed as a domain name.
73
74 * In the malloc_info output, the <heap> element may contain another <aspace>
75 element, "subheaps", which contains the number of sub-heaps.
76
77 * The libresolv function p_secstodate is no longer supported for new
78 programs.
79
80 Changes to build and runtime requirements:
81
82 * bison version 2.7 or later is required to generate code in the 'intl'
83 subdirectory.
84
85 Security related changes:
86
87 CVE-2009-5064: The ldd script would sometimes run the program under
88 examination directly, without preventing code execution through the
89 dynamic linker. (The glibc project disputes that this is a security
90 vulnerability; only trusted binaries must be examined using the ldd
91 script.)
92
93 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
94 suffered from a one-byte overflow during ~ operator processing (either
95 on the stack or the heap, depending on the length of the user name).
96 Reported by Tim Rühsen.
97
98 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
99 would sometimes fail to free memory allocated during ~ operator
100 processing, leading to a memory leak and, potentially, to a denial
101 of service.
102
103 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
104 without GLOB_NOESCAPE, could write past the end of a buffer while
105 unescaping user names. Reported by Tim Rühsen.
106
107 The following bugs are resolved with this release:
108
109 [The release manager will add the list generated by
110 scripts/list-fixed-bugs.py just before the release.]
111
112 \f
113 Version 2.26
114
115 Major new features:
116
117 * A per-thread cache has been added to malloc. Access to the cache requires
118 no locks and therefore significantly accelerates the fast path to allocate
119 and free small amounts of memory. Refilling an empty cache requires locking
120 the underlying arena. Performance measurements show significant gains in a
121 wide variety of user workloads. Workloads were captured using a special
122 instrumented malloc and analyzed with a malloc simulator. Contributed by
123 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
124
125 * Unicode 10.0.0 Support: Character encoding, character type info, and
126 transliteration tables are all updated to Unicode 10.0.0, using
127 generator scripts contributed by Mike FABIAN (Red Hat).
128 These updates cause user visible changes, especially the changes in
129 wcwidth for many emoji characters cause problems when emoji sequences
130 are rendered with pango, see for example:
131 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
132
133 * Collation of Hungarian has been overhauled and is now consistent with "The
134 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
135 Egmont Koblinger.
136
137 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
138
139 - The GNU C Library will now detect when /etc/resolv.conf has been
140 modified and reload the changed configuration. The new resolver option
141 “no-reload” (RES_NORELOAD) disables this behavior.
142
143 - The GNU C Library now supports an arbitrary number of search domains
144 (configured using the “search” directive in /etc/resolv.conf);
145 previously, there was a hard limit of six domains. For backward
146 compatibility, applications that directly modify the ‘_res’ global
147 object are still limited to six search domains.
148
149 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
150 Library will now randomly pick a name server from the configuration as a
151 starting point. (Previously, the second name server was always used.)
152
153 * The tunables feature is now enabled by default. This allows users to tweak
154 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
155
156 * New function reallocarray, which resizes an allocated block (like realloc)
157 to the product of two sizes, with a guaranteed clean failure upon integer
158 overflow in the multiplication. Originally from OpenBSD, contributed by
159 Dennis Wölfing and Rüdiger Sonderfeld.
160
161 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
162 These are extended versions of preadv and pwritev, respectively, taking an
163 additional flags argument. The set of supported flags depends on the
164 running kernel; full support currently requires kernel 4.7 or later.
165
166 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
167 create a new session ID for the spawned process. This feature is
168 scheduled to be added to the next major revision of POSIX; for the time
169 being, it is available under _GNU_SOURCE.
170
171 * errno.h is now safe to use from C-preprocessed assembly language on all
172 supported operating systems. In this context, it will only define the
173 Exxxx constants, as preprocessor macros expanding to integer literals.
174
175 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
176 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
177 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
178 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
179
180 To compile programs that use this feature, the compiler must support
181 128-bit floating point with the type name _Float128 (as defined by TS
182 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
183 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
184 must be defined to make the new interfaces visible.
185
186 The new functions and macros correspond to those present for other
187 floating-point types (except for a few obsolescent interfaces not
188 supported for the new type), with F128 or f128 suffixes; for example,
189 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
190 printf or scanf formats for the new type; the strfromf128 and strtof128
191 interfaces should be used instead.
192
193 Deprecated and removed features, and other changes affecting compatibility:
194
195 * The synchronization that pthread_spin_unlock performs has been changed to
196 now be equivalent to a C11 atomic store with release memory order to the
197 spin lock's memory location. Previously, several (but not all)
198 architectures used stronger synchronization (e.g., containing what is
199 often called a full barrier). This change can improve performance, but
200 may affect odd fringe uses of spin locks that depend on the previous
201 behavior (e.g., using spin locks as atomic variables to try to implement
202 Dekker's mutual exclusion algorithm).
203
204 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
205 removed.
206
207 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
208 will only be built and installed when the GNU C Library is configured with
209 --enable-obsolete-rpc. This allows alternative RPC implementations, such
210 as TIRPC or rpcsvc-proto, to be used.
211
212 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
213 libnss_compat, are deprecated, and will not be built or installed by
214 default.
215
216 The NIS(+) support library, libnsl, is also deprecated. By default, a
217 compatibility shared library will be built and installed, but not headers
218 or development libraries. Only a few NIS-related programs require this
219 library. (In particular, the GNU C Library has never required programs
220 that use 'gethostbyname' to be linked with libnsl.)
221
222 Replacement implementations based on TIRPC, which additionally support
223 IPv6, are available from <https://github.com/thkukuk/>. The configure
224 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
225 name service modules, to be built and installed.
226
227 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
228 support is enabled, the configured recursive resolver must support EDNS.
229 (Responding to EDNS-enabled queries with responses which are not
230 EDNS-enabled is fine, but FORMERR responses are not.)
231
232 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
233 servers have not supported this opcode for a long time.
234
235 * The _res_opcodes variable has been removed from libresolv. It had been
236 exported by accident.
237
238 * <string.h> no longer includes inline versions of any string functions,
239 as this kind of optimization is better done by the compiler. The macros
240 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
241
242 * The nonstandard header <xlocale.h> has been removed. Most programs should
243 use <locale.h> instead. If you have a specific need for the definition of
244 locale_t with no other declarations, please contact
245 libc-alpha@sourceware.org and explain.
246
247 * The obsolete header <sys/ultrasound.h> has been removed.
248
249 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
250
251 * The obsolete function cfree has been removed. Applications should use
252 free instead.
253
254 * The stack_t type no longer has the name struct sigaltstack. This changes
255 the C++ name mangling for interfaces involving this type.
256
257 * The ucontext_t type no longer has the name struct ucontext. This changes
258 the C++ name mangling for interfaces involving this type.
259
260 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
261 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
262 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
263 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
264 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
265 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
266 fpregset_t type no longer has the name struct fpu. This changes the C++
267 name mangling for interfaces involving those types.
268
269 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
270 synced with the kernel:
271
272 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
273 are not supported on this architecture and have been removed.
274
275 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
276 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
277 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
278
279 Programs that assume the GET/SETREGS ptrace requests are universally
280 available will now fail to build, instead of malfunctioning at runtime.
281
282 Changes to build and runtime requirements:
283
284 * Linux kernel 3.2 or later is required at runtime, on all architectures
285 supported by that kernel. (This is a change from version 2.25 only for
286 x86-32 and x86-64.)
287
288 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
289
290 * On most architectures, GCC 4.9 or later is required to build the GNU C
291 Library. On powerpc64le, GCC 6.2 or later is required.
292
293 Older GCC versions and non-GNU compilers are still supported when
294 compiling programs that use the GNU C Library. (We do not know exactly
295 how old, and some GNU extensions to C may be _de facto_ required. If you
296 are interested in helping us make this statement less vague, please
297 contact libc-alpha@sourceware.org.)
298
299 Security related changes:
300
301 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
302 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
303
304 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
305 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
306
307 * Avoid printing a backtrace from the __stack_chk_fail function since it is
308 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
309 (CVE-2010-3192).
310
311 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
312 fixed (CVE-2017-12133).
313
314 The following bugs are resolved with this release:
315
316 [984] network: Respond to changed resolv.conf in gethostbyname
317 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
318 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
319 fields
320 [12189] libc: __stack_chk_fail should not attempt a backtrace
321 (CVE-2010-3192)
322 [14096] time: Race condition on timezone/tst-timezone.out
323 [14172] localedata: az_IR: new locale
324 [14995] build: glibc fails to build if gold is the default linker, even if
325 ld.bfd is available
326 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
327 [16637] network: inet_pton function is accepting IPv6 with bad format
328 [16640] string: string/strtok.c: undefined behaviour inconsistent between
329 x86 and other generic code
330 [16875] localedata: ko_KR: fix lang_name
331 [17225] localedata: ar_SY: localized month names for May and June are
332 incorrect
333 [17297] localedata: da_DK: wrong date_fmt string
334 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
335 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
336 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
337 [19066] localedata: ar_SA abbreviated day and month names are in English
338 [19569] network: resolv: Support an arbitrary number of search domains
339 [19570] network: Implement random DNS server selection in the stub
340 resolver
341 [19838] locale: localedef fails on PA-RISC
342 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
343 order of 0D36 and 0D37
344 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
345 chillu characters
346 [20098] libc: FAIL: debug/backtrace-tst on hppa
347 [20257] network: sunrpc: clntudp_call does not enforce timeout when
348 receiving data
349 [20275] localedata: locale day/abday/mon/abmon should not have trailing
350 whitespace
351 [20313] localedata: Update locale data to Unicode 9.0
352 [20424] manual: Document how to provide a malloc replacement
353 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
354 for Peru
355 [20686] locale: Add el_GR@euro to SUPPORTED.
356 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
357 failures consistently
358 [21015] dynamic-link: Document and fix --enable-bind-now
359 [21016] nptl: pthread_cond support is broken on hppa
360 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
361 frame-pointer on i386
362 [21049] libc: segfault in longjmp_chk() due to clobbered processor
363 register
364 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
365 [21088] libc: Build fails with --enable-static-nss
366 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
367 generic c code is used
368 [21109] libc: Tunables broken on big-endian
369 [21112] math: powf has large ulp errors with base close to 1 and exponent
370 around 4000
371 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
372 (CVE-2017-12133)
373 [21120] malloc: glibc malloc is incompatible with GCC 7
374 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
375 with -lieee
376 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
377 (-0.0) when linking with -lieee
378 [21171] math: log10, log2 and lgamma return incorrect results
379 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
380 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
381 [21207] localedata: ce_RU: update weekdays from CLDR
382 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
383 [21217] localedata: Update months from CLDR-31
384 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
385 [21243] libc: support_delete_temp_file should issue warning for failed
386 remove()
387 [21244] libc: support resolv_test_start() socket fd close should be
388 checked for errors.
389 [21253] libc: localedef randomly segfaults when using -fstack-check due to
390 new posix_spawn implementation
391 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
392 leads to lower CPU frequency
393 [21259] libc: [alpha] termios.h missing IXANY for POSIX
394 [21261] libc: [sparc64] bits/setjmp.h namespace
395 [21267] network: [mips] bits/socket.h IOC* namespace
396 [21268] libc: [alpha] termios.h NL2, NL3 namespace
397 [21270] libc: mmap64 silently truncates large offset values
398 [21275] libc: posix_spawn always crashes on ia64 now
399 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
400 [21280] math: [powerpc] logbl for POWER7 return incorrect results
401 [21289] libc: Incorrect declaration for 32-bit platforms with
402 _FILE_OFFSET_BITS=64 causes build error
403 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
404 not support gethostbyname4_r
405 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
406 switching
407 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
408 arenas
409 [21340] libc: Support POSIX_SPAWN_SETSID
410 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
411 [21359] network: ns_name_pack needs additional byte in destination buffer
412 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
413 against fragmentation attacks (CVE-2017-12132)
414 [21369] network: resolv: Remove EDNS fallback
415 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
416 and _POSIX_C_SOURCE
417 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
418 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
419 [21393] stdio: Missing dup3 error check in freopen, freopen64
420 [21396] libc: Use AVX2 memcpy/memset on Skylake server
421 [21399] localedata: Bad description for U00EC in
422 localedata/charmaps/CP1254
423 [21411] malloc: realloc documentation error
424 [21426] network: sys/socket.h uio.h namespace
425 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
426 [21445] libc: signal.h bsd_signal namespace
427 [21455] network: Network headers stdint.h namespace
428 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
429 retry value
430 [21475] network: resolv: Overlong search path is truncated mid-label
431 [21511] libc: sigstack namespace
432 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
433 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
434 bits/syscall.h: No such file or directory
435 [21517] libc: struct sigaltstack namespace
436 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
437 [21533] localedata: Update locale data to Unicode 10.0
438 [21537] libc:
439 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
440 junk at end of line, first unrecognized character is `@'
441 [21538] libc: SIG_HOLD missing for XPG4
442 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
443 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
444 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
445 [21543] libc: sigevent namespace
446 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
447 size for general purpose registers in mcontext_t structure
448 [21550] libc: sigwait namespace
449 [21552] libc: XPG4 bsd_signal namespace
450 [21554] libc: sigpause namespace
451 [21560] libc: sys/wait.h signal.h namespace
452 [21561] libc: waitid namespace
453 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
454 '/usr/include/stdlib.h': Permission denied
455 [21575] libc: sys/wait.h missing struct rusage definition
456 [21584] libc: sigaltstack etc namespace
457 [21597] libc: siginterrupt namespace
458 [21607] math: hppa: FAIL: math/test-tgmath
459 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
460 issue on x86-64
461 [21622] libc: [tile] missing SA_* for POSIX.1:2008
462 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
463 stack and heap (CVE-2017-1000366)
464 [21625] libc: wait3 namespace
465 [21654] nss: Incorrect pointer alignment in NSS group merge result
466 construction
467 [21657] network: Parse interface zone id for node-local multicast
468 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
469 between_2_3
470 [21666] libc: .symver is used on common symbol
471 [21668] network: resolv: res_init cross-thread broadcast introduces race
472 conditions
473 [21687] math: tgmath.h totalorder, totalordermag return type
474 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
475 Hindi Locale
476 [21696] libc: Incorrect assumption of of __cpu_mask in
477 posix/sched_cpucount.c
478 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
479 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
480 locale
481 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
482 leading to relocation crash
483 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
484 `resolv_conf_matches (resp, conf)' failed.
485 [21710] localedata: Added Samoan language locale for Samoa
486 [21711] localedata: Pashto yesstr/nostr locale are missing
487 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
488 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
489 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
490 [21723] localedata: yesstr/nostr missing for Chinese language locale
491 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
492 locale
493 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
494 locale
495 [21728] localedata: New Locale for Tongan language
496 [21729] localedata: incorrect LC_NAME fields for hi_IN
497 [21733] localedata: yesstr and nostr are missing for zh_HK
498 [21734] localedata: Missing yesstr and nostr are for kw_GB
499 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
500 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
501 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
502 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
503 strings
504 [21744] libc: Tests failing on --enable-tunables --enable-stack-
505 protector=all
506 [21749] localedata: Wrong abbreviated day name (“abday”) for
507 ar_JO/ar_LB/ar_SY
508 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
509 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
510 [21759] localedata: missing yesstr and nostr for Tigrinya
511 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
512 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
513 [21767] localedata: Missing Bislama locales
514 [21768] localedata: Missing yesstr and nostr for aa_ET
515 [21770] localedata: Missing Field in li_NL
516 [21778] nptl: Robust mutex may deadlock
517 [21779] libc: MicroBlaze segfaults when loading libpthread
518 [21783] localedata: Fix int_select international_call_prefixes
519 [21784] localedata: Inconsistency in country_isbn
520 [21788] localedata: Missing Country Postal Abbreviations
521 [21794] localedata: Added-country_isbn-for-Italy
522 [21795] localedata: Add/Fix country_isbn for France
523 [21796] localedata: Added country_isbn for Republic of Korea
524 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
525 [21799] localedata: Added int_select international_call_prefixes
526 [21801] localedata: Added int_select international_call_prefixes
527 [21804] nptl: Double semicolon in thread-shared-types.h
528 [21807] localedata: LC_ADDRESS fix for pap_CW
529 [21808] localedata: Fix LC_ADDRESS for pap_AW
530 [21821] localedata: Added country_name in mai_IN
531 [21822] localedata: Fix LC_TIME for mai_IN
532 [21823] localedata: missing yesstr, nostr for sa_IN
533 [21825] localedata: Fix name_mrs for mag_IN
534 [21828] localedata: 2.26 changelog should mention user visible changes
535 with unicode 9.0
536 [21835] localedata: Added Maithili language locale for Nepal
537 [21838] localedata: Removed redundant data for the_NP
538 [21839] localedata: Fix LC_MONETARY for ta_LK
539 [21844] localedata: Fix Latin characters and Months Sequence.
540 [21848] localedata: Fix mai_NP Title Name
541
542 \f
543 Version 2.25
544
545 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
546 24731-2:2010, is supported to enable declarations of functions from that
547 TR. Note that not all functions from that TR are supported by the GNU C
548 Library.
549
550 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
551 18661-1:2014, is supported to enable declarations of functions and macros
552 from that TS. Note that not all features from that TS are supported by
553 the GNU C Library.
554
555 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
556 18661-4:2015, is supported to enable declarations of functions and macros
557 from that TS. Note that most features from that TS are not supported by
558 the GNU C Library.
559
560 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
561 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
562 Since the GNU C Library defaults to a much newer revision of POSIX, this
563 will only affect programs that specifically request an old conformance
564 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
565 see a change in the visible declarations, but a program compiled with
566 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
567 will not.
568
569 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
570 defined by all multithreaded code, but glibc has not required this for
571 many years.
572
573 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
574 means that in a future release, the macros “major”, “minor”, and “makedev”
575 will only be available from <sys/sysmacros.h>.
576
577 These macros are not part of POSIX nor XSI, and their names frequently
578 collide with user code; see for instance glibc bug 19239 and Red Hat bug
579 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
580 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
581 problem.
582
583 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
584 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
585 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
586
587 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
588 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
589 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
590 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
591 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
592 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
593 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
594 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
595 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
596 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
597 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
598 WINT_WIDTH.
599
600 * New <math.h> features are added from TS 18661-1:2014:
601
602 - Signaling NaN macros: SNANF, SNAN, SNANL.
603
604 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
605 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
606 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
607
608 - llogb functions: the llogb, llogbf and llogbl functions, and the
609 FP_LLOGB0 and FP_LLOGBNAN macros.
610
611 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
612 fminmagf, fminmagl.
613
614 - Comparison macros: iseqsig.
615
616 - Classification macros: iscanonical, issubnormal, iszero.
617
618 - Total order functions: totalorder, totalorderf, totalorderl,
619 totalordermag, totalordermagf, totalordermagl.
620
621 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
622
623 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
624 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
625
626 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
627 are added to libc. They convert a floating-point number into string.
628
629 * Most of glibc can now be built with the stack smashing protector enabled.
630 It is recommended to build glibc with --enable-stack-protector=strong.
631 Implemented by Nick Alcock (Oracle).
632
633 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
634 intended to be used instead of memset() to erase sensitive data after use;
635 the compiler will not optimize out calls to explicit_bzero even if they
636 are "unnecessary" (in the sense that no _correct_ program can observe the
637 effects of the memory clear).
638
639 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
640 to float instead of double. This does not affect the ABI of any libraries
641 that are part of the GNU C Library, but may affect the ABI of other
642 libraries that use this type in their interfaces.
643
644 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
645 float_t and double_t types are now defined to long double instead of float
646 and double. These options are not the default, and this does not affect
647 the ABI of any libraries that are part of the GNU C Library, but it may
648 affect the ABI of other libraries that use this type in their interfaces,
649 if they are compiled or used with those options.
650
651 * The getentropy and getrandom functions, and the <sys/random.h> header file
652 have been added.
653
654 * The buffer size for byte-oriented stdio streams is now limited to 8192
655 bytes by default. Previously, on Linux, the default buffer size on most
656 file systems was 4096 bytes (and thus remains unchanged), except on
657 network file systems, where the buffer size was unpredictable and could be
658 as large as several megabytes.
659
660 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
661 for the Linux quota interface which predates kernel version 2.4.22 has
662 been removed.
663
664 * The malloc_get_state and malloc_set_state functions have been removed.
665 Already-existing binaries that dynamically link to these functions will
666 get a hidden implementation in which malloc_get_state is a stub. As far
667 as we know, these functions are used only by GNU Emacs and this change
668 will not adversely affect already-built Emacs executables. Any undumped
669 Emacs executables, which normally exist only during an Emacs build, should
670 be rebuilt by re-running “./configure; make” in the Emacs build tree.
671
672 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
673 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
674 “no-ip6-dotint” had already been the default, and support for the
675 “ip6-dotint” option was removed from the Internet in 2006.
676
677 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
678 flag from <resolv.h> have been removed. The option relied on a
679 backwards-incompatible DNS extension which was never deployed on the
680 Internet.
681
682 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
683 RES_BLAST defined in the <resolv.h> header file have been deprecated.
684 They were already unimplemented.
685
686 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
687 _res.flags are deprecated. The flag was standardized in RFC 2133, but
688 removed again from the IETF name lookup interface specification in RFC
689 2553. Applications should use getaddrinfo instead.
690
691 * DNSSEC-related declarations and definitions have been removed from the
692 <arpa/nameser.h> header file, and libresolv will no longer attempt to
693 decode the data part of DNSSEC record types. Previous versions of glibc
694 only implemented minimal support for the previous version of DNSSEC, which
695 is incompatible with the currently deployed version.
696
697 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
698 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
699 <arpa/nameser.h> header file because the distinction between RR types and
700 meta-RR types is not officially standardized, subject to revision, and
701 thus not suitable for encoding in a macro.
702
703 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
704 and rhook members of the res_state type in <resolv.h> have been removed.
705 The glibc stub resolver did not support these hooks, but the header file
706 did not reflect that.
707
708 * For multi-arch support it is recommended to use a GCC which has
709 been built with support for GNU indirect functions. This ensures
710 that correct debugging information is generated for functions
711 selected by IFUNC resolvers. This support can either be enabled by
712 configuring GCC with '--enable-gnu-indirect-function', or by
713 enabling it by default by setting 'default_gnu_indirect_function'
714 variable for a particular architecture in the GCC source file
715 'gcc/config.gcc'.
716
717 * GDB pretty printers have been added for mutex and condition variable
718 structures in POSIX Threads. When installed and loaded in gdb these pretty
719 printers show various pthread variables in human-readable form when read
720 using the 'print' or 'display' commands in gdb.
721
722 * Tunables feature added to allow tweaking of the runtime for an application
723 program. This feature can be enabled with the '--enable-tunables' configure
724 flag. The GNU C Library manual has details on usage and README.tunables has
725 instructions on adding new tunables to the library.
726
727 * A new version of condition variables functions have been implemented in
728 the NPTL implementation of POSIX Threads to provide stronger ordering
729 guarantees.
730
731 * A new version of pthread_rwlock functions have been implemented to use a more
732 scalable algorithm primarily through not using a critical section anymore to
733 make state changes.
734
735 Security related changes:
736
737 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
738 have been created with makecontext could fail to terminate due to a
739 missing .cantunwind annotation. This has been observed to lead to a hang
740 (denial of service) in some Go applications compiled with gccgo. Reported
741 by Andreas Schwab. (CVE-2016-6323)
742
743 * The DNS stub resolver functions would crash due to a NULL pointer
744 dereference when processing a query with a valid DNS question type which
745 was used internally in the implementation. The stub resolver now uses a
746 question type which is outside the range of valid question type values.
747 (CVE-2015-5180)
748
749 The following bugs are resolved with this release:
750
751 [4099] stdio: Overly agressive caching by stream i/o functions.
752 [7065] build: Support building glibc with -fstack-protector or -fstack-
753 protector-all
754 [9842] localedata: en_CA: incorrect date format
755 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
756 before it started waiting
757 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
758 [15765] nptl: sem_open is wrongly a cancellation point
759 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
760 [16458] libc: endian.h and netinet/in.h byte order macros return values of
761 wrong type
762 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
763 library linked with pthread
764 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
765 "FIXME: Ingo" issue)
766 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
767 [17252] libc: getrandom and getentropy syscall
768 [17730] malloc: thread-local storage is sometimes improperly free()'d
769 after being __libc_memalign()'d
770 [18241] stdio: failed fseek on memstream does not set errno and can fail
771 when it shouldnt
772 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
773 not
774 [18463] nptl: pthread_cond_broadcast issue when surrounded by
775 PTHREAD_PRIO_INHERIT mutex on ARM
776 [18784] network: res_query and related function crash for special record
777 type queries (CVE-2015-5180)
778 [19380] math: strtod does not raise "inexact"
779 [19387] string: Integer overflow in memchr
780 [19390] string: Integer overflow in strncat
781 [19398] build: linknamespace tests fail with massively parallel build
782 [19402] nptl: Deadlock with robust shared mutex and asynchronous
783 termination
784 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
785 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
786 [19514] libc: [PATCH] Fix spelling errors in spelling
787 "implement"/"implementation" in several places
788 [19582] network: Deprecate RES_USE_INET6
789 [19673] manual: clog10 docs appear to be erroneous
790 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
791 causes a segmentation fault
792 [19826] libc: invalid pointer returned from __tls_get_addr with static
793 linking
794 [20016] network: resolv: Remove hooks support from the API
795 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
796 uninitialized GOT
797 [20033] math: [x86_64] vectorized math function don't call the __finite
798 versions
799 [20116] nptl: use after free in pthread_create
800 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
801 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
802 is always true.
803 [20311] nptl: please install proc_service.h
804 [20366] libc: Compilation errors in installed headers in strict-compliance
805 modes
806 [20370] malloc: malloc: Arena free list management is still racy
807 (incorrect fix in bug 19243)
808 [20386] libc: assert (X = 0) does not result in GCC warning
809 [20432] malloc: malloc: Minimize interface required for interposition
810 [20435] libc: Missing unwind info in __startcontext causes infinite loop
811 in _Unwind_Backtrace (CVE-2016-6323)
812 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
813 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
814 libpthread.a
815 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
816 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
817 [20477] network: resolv: incorrect double-checked locking related to
818 _res_hconf
819 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
820 information.
821 [20495] math: x86_64 performance degradation due to AVX/SSE transition
822 penalty
823 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
824 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
825 cause transition penalty
826 [20517] math: sparcv9 missing fdiml compat symbol
827 [20524] manual: strverscmp is inconsistent
828 [20525] libc: <sys/quota.h> should be based on kernel headers
829 [20539] math: GCC 7 -static -lm fails to link at -O3
830 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
831 mcount.oS)
832 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
833 Checking
834 [20591] network: Remove obsolete DNSSEC support
835 [20592] network: DNS resource record type classification macros in
836 <arpa/nameser.h> are incorrect
837 [20593] network: Update DNS RR type definitions
838 [20611] network: getaddrinfo accepts invalid numeric scope IDs
839 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
840 multi-arch
841 [20629] network: libresolv: Remove support for bitstring labels
842 (RES_USEBSTRING)
843 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
844 assertion: "maxidx >= 2"
845 [20660] math: [arm] Use VSQRT
846 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
847 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
848 [20689] libc: Test for FMA should also check for AVX.
849 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
850 [20715] math: iszero macro breaks existing code
851 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
852 [20728] libc: powerpc: Missing TOC stub in clone
853 [20729] build: glibc-2.24 fails to build for i486 with -Os
854 [20750] build: Build fails with default PIE enabled gcc-6.x
855 [20768] math: [alpha] sqrt fegetenv namespace
856 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
857 glibc
858 [20787] math: float_t is defined as float incorrectly on x86_64 even if
859 __FLT_EVAL_METHOD__ is 2
860 [20790] network: rpcgen buffer overrun in get_prog_declaration
861 [20822] nptl: race condition in __lll_unlock_elision on powerpc
862 [20829] libc: crypt snprintf namespace
863 [20847] libc: tst-vfork3 failure
864 [20855] math: Default bits/mathdef.h has inappropriate float_t
865 [20859] math: [sh4] FP_ILOGB0 invalid
866 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
867 U+20AC), not same as GBK
868 [20915] dynamic-link: global-dynamic TLS broken on aarch64
869 [20916] math: pow handling of sNaN arguments
870 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
871 [20919] math: Bad pow (qNaN, 0) result with -lieee
872 [20940] math: hypot sNaN handling
873 [20947] math: fmax, fmin sNaN handling
874 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
875 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
876 RES_USE_INET6
877 [20971] string: powerpc64/power7 memchr overflows internal pointer check
878 [20973] nptl: robust mutexes: Lost wake-ups
879 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
880 [20978] nis: strlen on null pointer in nss_nisplus
881 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
882 wrong condition
883 [21014] string: i686 memchr overflows internal pointer check
884 [21019] libc: [mips] n32 lseek incorrect on overflow
885 [21022] libc: [microblaze] __backtrace get_frame_size namespace
886 [21026] libc: [MIPS] readahead syscall is broken on n64
887 [21028] math: Fallback fesetexceptflag should always succeed
888 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
889 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
890 clang
891 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
892 [21061] librt: [microblaze] librt lost clock_* exports
893 [21073] libc: tunables: insecure environment variables passed to
894 subprocesses with AT_SECURE
895 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
896 \f
897 Version 2.24
898
899 * The minimum Linux kernel version that this version of the GNU C Library
900 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
901 kernel version 2.6.32 or later suffices (on architectures that already
902 required kernel versions more recent than 3.2, those requirements remain
903 unchanged). Linux 3.2 or later kernel headers are required on all
904 architectures.
905
906 * The pap_AN locale has been deleted. This has been deprecated for a long
907 time. It has been replaced by pap_AW & pap_CW, both of which have long
908 been included in previous releases.
909
910 * The readdir_r and readdir64_r functions have been deprecated. It is
911 recommended to use readdir and readdir64 instead.
912
913 * The type “union wait” has been removed. It was deprecated in the early
914 1990s and never part of POSIX. Application code should use the int type
915 instead of “union wait”.
916
917 * A new NSS action is added to facilitate large distributed system
918 administration. The action, MERGE, allows remote user stores like LDAP
919 to be merged into local user stores like /etc/groups in order to provide
920 easy to use, updated, and managed sets of merged credentials. The new
921 action can be used by configuring it in /etc/nsswitch.conf:
922 group: files [SUCCESS=merge] nis
923 Implemented by Stephen Gallagher (Red Hat).
924
925 * The deprecated __malloc_initialize_hook variable has been removed from the
926 API.
927
928 * The long unused localedef --old-style option has been removed. It hasn't
929 done anything in over 16 years. Scripts using this option can safely
930 drop it.
931
932 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
933 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
934 return the next representable value in the direction of positive infinity
935 and the nextdown functions return the next representable value in the
936 direction of negative infinity. These are currently enabled as GNU
937 extensions.
938
939 Security related changes:
940
941 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
942 could result in a stack overflow when getnetbyname was called with an
943 overly long name. (CVE-2016-3075)
944
945 * Previously, getaddrinfo copied large amounts of address data to the stack,
946 even after the fix for CVE-2013-4458 has been applied, potentially
947 resulting in a stack overflow. getaddrinfo now uses a heap allocation
948 instead. Reported by Michael Petlan. (CVE-2016-3706)
949
950 * The glob function suffered from a stack-based buffer overflow when it was
951 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
952 Reported by Alexander Cherepanov. (CVE-2016-1234)
953
954 * The Sun RPC UDP client could exhaust all available stack space when
955 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
956 alloca plugin for GCC. (CVE-2016-4429)
957
958 * The IPv6 name server management code in libresolv could result in a memory
959 leak for each thread which is created, performs a failing naming lookup,
960 and exits. Over time, this could result in a denial of service due to
961 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
962
963 The following bugs are resolved with this release:
964
965 [1170] localedata: ne_NP: update Nepali locale definition file
966 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
967 of MS-DOS.
968 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
969 [6796] math: fdim() does not set errno on overflow
970 [10354] libc: posix_spawn should use vfork() in more cases than presently
971 [11213] localedata: localedata: add copyright disclaimer to locale files
972 [12143] localedata: chr_US: new Cherokee locale
973 [12450] localedata: sgs_LT: new locale
974 [12676] localedata: ln_CD: new locale
975 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
976 CLDR data
977 [13304] math: fma, fmaf, fmal produce wrong results
978 [14259] build: --localedir arg to configure is ignored
979 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
980 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
981 [14934] localedata: es_CL: wrong first weekday chilean locale
982 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
983 romanisation
984 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
985 and +/-
986 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
987 [15368] nptl: raise() is not async-signal-safe
988 [15479] math: ceil, floor, round and trunc raise inexact exception
989 [15578] localedata: kk_KZ: various updates
990 [16003] localedata: pap_AN: punt old locale
991 [16137] localedata: iw_IL: punt old locale
992 [16190] localedata: eo: new esperanto locale
993 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
994 [16742] malloc: race condition: pthread_atfork() called before first
995 malloc() results in unexpected locking behaviour/deadlocks
996 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
997 all locales
998 [16983] localedata: postal_fmt does not allow %l and %n modifiers
999 [17565] localedata: pt_PT: wrong (work-)week start
1000 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1001 [17950] build: Build fails with -msse
1002 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1003 [18433] libc: posix_spawn does not return correctly upon failure to
1004 execute
1005 [18453] localedata: charmaps/IBM875: incorrect codes
1006 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1007 -Wsystem-headers
1008 [18896] localedata: he_IL: improvements for currency
1009 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1010 Romanian locale data
1011 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1012 symbol
1013 [19056] libc: Deprecate readdir_r
1014 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1015 language
1016 [19198] localedata: nl_NL: small improvements for Dutch locales
1017 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1018 nameservers (CVE-2016-5417)
1019 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1020 machine
1021 [19400] locale: Language missing in "iso-639.def", trivial fix in
1022 description
1023 [19431] malloc: Deadlock between fflush, getdelim, and fork
1024 [19505] libc: Incorrect file descriptor validity checks in
1025 posix_spawn_file_actions_add{open,close,dup2}
1026 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1027 when using RTLD_NEXT
1028 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1029 `intl/{gettextP,loadinfo}.h'
1030 [19534] libc: execle, execlp may use malloc
1031 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1032 [19573] network: res_nclose and __res_maybe_init disagree about name
1033 server initialization, breaking Hesiod
1034 [19575] localedata: Status of GB18030 tables
1035 [19581] localedata: sr_* date_fmt string contains additional newline
1036 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1037 Excavator core
1038 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1039 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1040 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1041 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1042 double range
1043 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1044 part zero incorrect
1045 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1046 equality tests
1047 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1048 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1049 [19633] locale: strfmon_l applies global locale to number formatting
1050 [19642] network: Memory leak in getnameinfo
1051 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1052 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1053 glibc-2.22
1054 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1055 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1056 'tst-numeric.c'
1057 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1058 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1059 low part
1060 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1061 result
1062 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1063 unaligned stack
1064 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1065 pointers and lengths in error-case.
1066 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1067 report errors on UTF-16 surrogates values.
1068 [19755] nscd: nscd assertion failure in gc
1069 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1070 [19759] libc: mempcpy shouldn't be inlined
1071 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1072 [19765] libc: s390 needs an optimized mempcpy
1073 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1074 NAME_MAX limit assumption (CVE-2016-1234)
1075 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1076 [19787] network: Missing and incorrect truncation checks in getnameinfo
1077 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1078 modes
1079 [19791] network: Assertion failure in res_query.c with un-connectable name
1080 server addresses
1081 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1082 [19822] math: libm.so install clobbers old version
1083 [19825] network: resolv: send_vc can return uninitialized data in second
1084 response to getaddrinfo
1085 [19830] network: nss_dns: should check RDATA length against buffer length
1086 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1087 confronted with A/AAAA records of invalid size
1088 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1089 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1090 [19853] stdio: Printing IBM long double in decimal with high precision is
1091 sometimes incorrect
1092 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1093 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1094 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1095 record types
1096 [19865] network: Assertion failure or memory leak in
1097 _nss_dns_getcanonname_r
1098 [19868] network: nss_dns: netent code does not skip over non-PTR records
1099 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1100 (CVE-2016-3075)
1101 [19881] string: Improve x86-64 memset
1102 [19907] string: Incorrect memcpy tests
1103 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1104 resolving symbols
1105 [19925] libc: termios.h XCASE namespace
1106 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1107 [19929] libc: limits.h NL_NMAX namespace
1108 [19931] stdio: Memory leak in vfprintf
1109 [19957] libc: clone(CLONE_VM) access invalid parent memory
1110 [19963] localedata: en_IL: New locale
1111 [19989] stdio: stdio.h cuserid namespace
1112 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1113 gethosts
1114 [19996] locale: langinfo.h nl_langinfo_l namespace
1115 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1116 [20010] network: getaddrinfo: Stack overflow in hostent translation
1117 (CVE-2016-3706)
1118 [20012] stdio: libio: fmemopen append mode failure
1119 [20014] stdio: stdio.h namespace for pre-threads POSIX
1120 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1121 [20023] libc: fcntl.h timespec namespace
1122 [20024] math: [x86_64] vectorized sincos trashes the stack
1123 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1124 [20041] time: sys/time.h timespec namespace
1125 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1126 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1127 [20051] libc: ttyslot in wrong header under wrong conditions
1128 [20054] libc: gethostname not declared for XPG4
1129 [20055] libc: termios.h missing tcgetsid for XPG4
1130 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1131 executable
1132 [20073] libc: sys/stat.h fchmod namespace
1133 [20074] libc: stdlib.h rand_r namespace
1134 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1135 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1136 XPG3
1137 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1138 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1139 (CVE-2016-4429)
1140 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1141 [20119] libc: Wrong mask for processors level type from CPUID
1142 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1143 AS not supporting AVX512
1144 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1145 sNaN argument
1146 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1147 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1148 argument
1149 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1150 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1151 [20178] libc: posix_spawn{p} should not call exit
1152 [20191] stdio: libio: vtables hardening
1153 [20195] string: FMA4 detection requires CPUID execution with register
1154 eax=0x80000001
1155 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1156 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1157 subnormals
1158 [20212] math: acos (sNaN) returns sNaN
1159 [20213] math: asin (sNaN) returns sNaN
1160 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1161 [20218] math: [i386] asinhl (sNaN) returns sNaN
1162 [20219] math: [i386] atanhl (sNaN) returns sNaN
1163 [20222] stdio: fopencookie: Mangle function pointers
1164 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1165 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1166 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1167 input
1168 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1169 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1170 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1171 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1172 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1173 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1174 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1175 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1176 [20240] math: modf (sNaN) returns sNaN
1177 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1178 [20250] math: frexp (sNaN) returns sNaN
1179 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1180 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1181 double rounding
1182 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1183 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1184 subscript is below array bounds [-Werror=array-bounds]
1185 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1186 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1187 (Only arm/linux)
1188 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1189 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1190 fallbacks
1191 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1192 "invalid" exceptions
1193 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1194 [20316] localedata: id_ID: Februari instead of Pebruari
1195 [20327] string: POWER8 strcasecmp returns incorrect result
1196 [20347] math: Failure: Test: j0_downward (0xap+0)
1197 [20348] libc: FAIL: misc/tst-preadvwritev64
1198 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1199 [20350] libc: There is no test for p{read,write}64
1200 [20357] math: Incorrect cos result for 1.5174239687223976
1201 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1202 \f
1203 Version 2.23
1204
1205 * Unicode 8.0.0 Support: Character encoding, character type info, and
1206 transliteration tables are all updated to Unicode 8.0.0, using new
1207 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1208 These updates cause user visible changes, such as the fixes for bugs
1209 89, 16061, and 18568.
1210
1211 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1212 kernel-internal CPU set size. This means that requests that change the
1213 CPU affinity which failed before (for example, an all-ones CPU mask) will
1214 now succeed. Applications that need to determine the effective CPU
1215 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1216 setting it because the kernel can adjust it (and the previous size check
1217 would not detect this in the majority of cases).
1218
1219 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1220 the following new symbols are used: fts64_children, fts64_close,
1221 fts64_open, fts64_read and fts64_set.
1222
1223 * getaddrinfo now detects certain invalid responses on an internal netlink
1224 socket. If such responses are received, an affected process will
1225 terminate with an error message of "Unexpected error <number> on netlink
1226 descriptor <number>" or "Unexpected netlink response of size <number> on
1227 descriptor <number>". The most likely cause for these errors is a
1228 multi-threaded application which erroneously closes and reuses the netlink
1229 file descriptor while it is used by getaddrinfo.
1230
1231 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1232 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1233 unnecessary serialization of memory allocation requests across threads.
1234 The defect is now corrected. Users should see a substantial increase in
1235 the concurent throughput of allocation requests for applications which
1236 trigger this bug. Affected applications typically create and
1237 destroy threads frequently. (Bug 19048 was reported and analyzed by
1238 Ericsson.)
1239
1240 * There is now a --disable-timezone-tools configure option for disabling the
1241 building and installing of the timezone related utilities (zic, zdump, and
1242 tzselect). This is useful for people who build the timezone data and code
1243 independent of the GNU C Library.
1244
1245 * The obsolete header <regexp.h> has been removed. Programs that require
1246 this header must be updated to use <regex.h> instead.
1247
1248 * The obsolete functions bdflush, create_module, get_kernel_syms,
1249 query_module and uselib are no longer available to newly linked binaries;
1250 the header <sys/kdaemon.h> has been removed. These functions and header
1251 were specific to systems using the Linux kernel and could not usefully be
1252 used with the GNU C Library on systems with version 2.6 or later of the
1253 Linux kernel.
1254
1255 * Optimized string, wcsmbs and memory functions for IBM z13.
1256 Implemented by Stefan Liebler.
1257
1258 * Newly linked programs that define a variable called signgam will no longer
1259 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1260 require signgam to be set by those functions must ensure that they use the
1261 variable provided by the GNU C Library and declared in <math.h>, without
1262 defining their own copy.
1263
1264 * The minimum GCC version that can be used to build this version of the GNU
1265 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1266 still be used to compile programs using the GNU C Library.
1267
1268 Security related changes:
1269
1270 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1271 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1272
1273 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1274 the pointer guard feature. It is always enabled. Previously,
1275 LD_POINTER_GUARD could be used to disable security hardening in binaries
1276 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1277 (CVE-2015-8777)
1278
1279 * An integer overflow in hcreate and hcreate_r could lead to an
1280 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1281
1282 * The catopen function no longer has unbounded stack usage. Reported by
1283 Max. (CVE-2015-8779)
1284
1285 * The nan, nanf and nanl functions no longer have unbounded stack usage
1286 depending on the length of the string passed as an argument to the
1287 functions. Reported by Joseph Myers. (CVE-2014-9761)
1288
1289 * A stack-based buffer overflow was found in libresolv when invoked from
1290 libnss_dns, allowing specially crafted DNS responses to seize control
1291 of execution flow in the DNS client. The buffer overflow occurs in
1292 the functions send_dg (send datagram) and send_vc (send TCP) for the
1293 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1294 family. The use of AF_UNSPEC triggers the low-level resolver code to
1295 send out two parallel queries for A and AAAA. A mismanagement of the
1296 buffers used for those queries could result in the response of a query
1297 writing beyond the alloca allocated buffer created by
1298 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1299 the overflow. Thanks to the Google Security Team and Red Hat for
1300 reporting the security impact of this issue, and Robert Holiday of
1301 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1302
1303 The following bugs are resolved with this release:
1304
1305 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1306 [887] math: Math library function "logb" and "nextafter" inconsistent
1307 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1308 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1309 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1310 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1311 use `mkstemp'
1312 [4404] localedata: German translation of "Alarm clock" is misleading
1313 [6799] math: nextafter() and nexttoward() doen't set errno on
1314 overflow/underflow errors
1315 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1316 overflow/underflow
1317 [10432] nis: _nss_nis_setnetgrent assertion failure
1318 [11460] libc: fts has no LFS support
1319 [12926] network: getaddrinfo()/make_request() may spin forever
1320 [13065] nptl: Race condition in pthread barriers
1321 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1322 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1323 are not contiguous
1324 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1325 [14912] libc: Rename non-installed bits/*.h headers
1326 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1327 [15367] math: Let gcc use __builtin_isinf
1328 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1329 [15421] math: lgamma wrongly sets signgam for ISO C
1330 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1331 FE_INVALID with argument out of range
1332 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1333 all exceptions
1334 [15786] dynamic-link: ifunc resolver functions can smash function
1335 arguments
1336 [15918] math: Unnecessary check for equality in hypotf()
1337 [16061] localedata: Review / update transliteration data
1338 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1339 should include
1340 [16141] time: strptime %z offset restriction
1341 [16171] math: drem should be alias of remainder
1342 [16296] math: fegetround is pure?
1343 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1344 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1345 GNU/Linux
1346 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1347 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1348 arguments
1349 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1350 "Exception "Inexact" set" on ppc32
1351 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1352 [16517] math: Missing underflow exception from tanf/tan/tanl
1353 [16519] math: Missing underflow exception from sinhf
1354 [16520] math: Missing underflow exception from tanhf
1355 [16521] math: Missing underflow exception from exp2
1356 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1357 rounding results
1358 [16734] stdio: fopen calls mmap to allocate its buffer
1359 [16961] math: nan function incorrect handling of bad sequences
1360 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1361 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1362 [16985] locale: localedef: confusing error message when opening output
1363 fails
1364 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1365 [17197] locale: Redundant shift character in iconv conversion output at
1366 block boundary
1367 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1368 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1369 [17250] dynamic-link: static linking breaks nss loading
1370 (getaddrinfo/getpwnam/etc...)
1371 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1372 4.7?
1373 [17441] math: isnan() should use __builtin_isnan() in GCC
1374 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1375 (related to lock elision)
1376 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1377 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1378 [17887] time: strptime should be able to parse "+01:00" style timezones
1379 [17905] libc: catopen() Multiple unbounded stack allocations
1380 (CVE-2015-8779)
1381 [18084] libc: backtrace (..., 0) dumps core on x86
1382 [18086] libc: nice() sets errno to 0 on success
1383 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1384 is too large (CVE-2015-8778)
1385 [18251] dynamic-link: SONAME missing when audit modules provides path
1386 [18265] libc: add attributes for wchar string and memory functions
1387 [18370] math: csqrt missing underflows
1388 [18421] libc: [hppa] read-only segment has dynamic relocations
1389 [18472] libc: Obsolete syscall wrappers should be compat symbols
1390 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1391 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1392 [18525] localedata: Remove locale timezone information
1393 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1394 [18568] localedata: Update locale data to Unicode 8.0
1395 [18589] locale: sort-test.sh fails at random
1396 [18595] math: ctan, ctanh missing underflows
1397 [18604] libc: assert macro-expands its argument
1398 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1399 contains a vector instruction exception.
1400 [18611] math: j1, jn missing errno setting on underflow
1401 [18618] localedata: sync Chechen locale definitions with other *_RU
1402 locales
1403 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1404 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1405 [18665] network: In send_dg, the recvfrom function is NOT always using the
1406 buffer size of a newly created buffer (CVE-2015-7547)
1407 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1408 missing break ?
1409 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1410 32bit processes
1411 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1412 [18699] math: tilegx cproj() for various complex infinities does not yield
1413 infinity
1414 [18724] libc: Harden put*ent functions against data injection
1415 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1416 [18755] build: build errors with -DNDEBUG
1417 [18757] stdio: fmemopen fails to set errno on failure
1418 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1419 be forced unloaded
1420 [18781] libc: openat64 lacks O_LARGEFILE
1421 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1422 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1423 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1424 [18790] math: [ldbl-128ibm] tanhl inaccurate
1425 [18795] libc: stpncpy fortification misses buffer lengths that are
1426 statically too large
1427 [18796] build: build fails for --disable-mathvec
1428 [18803] math: hypot missing underflows
1429 [18820] stdio: fmemopen may leak memory on failure
1430 [18823] math: csqrt spurious underflows
1431 [18824] math: fma spurious underflows
1432 [18825] math: pow missing underflows
1433 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1434 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1435 [18870] build: sem_open.c fails to compile with missing symbol
1436 FUTEX_SHARED
1437 [18872] stdio: Fix memory leak in printf_positional
1438 [18873] libc: posix_fallocate overflow check ineffective
1439 [18875] math: Excess precision leads incorrect libm
1440 [18877] libc: arm: mmap offset regression
1441 [18887] libc: memory corruption when using getmntent on blank lines
1442 [18918] localedata: hu_HU: change time to HH:MM:SS format
1443 [18921] libc: Regression: extraneous stat() and fstat() performed by
1444 opendir()
1445 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1446 binaries (CVE-2015-8777)
1447 [18951] math: tgamma missing underflows
1448 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1449 signgam
1450 [18953] localedata: lt_LT: change currency symbol to the euro
1451 [18956] math: powf inaccuracy
1452 [18961] math: [i386] exp missing underflows
1453 [18966] math: [i386] exp10 missing underflows
1454 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1455 [18969] build: multiple string test failures due to missing locale
1456 dependencies
1457 [18970] libc: Reference of pthread_setcancelstate in libc.a
1458 [18977] math: float / long double Bessel functions not in XSI POSIX
1459 [18980] math: i386 libm functions return with excess range and precision
1460 [18981] math: i386 scalb*, ldexp return with excess range and precision
1461 [18982] stdio: va_list and vprintf
1462 [18985] time: Passing out of range data to strftime() causes a segfault
1463 (CVE-2015-8776)
1464 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1465 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1466 [19012] locale: iconv_open leaks memory on error path
1467 [19016] math: clog, clog10 inaccuracy
1468 [19018] nptl: Mangle function pointers in tls_dtor_list
1469 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1470 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1471 [19048] malloc: malloc: arena free list can become cyclic, increasing
1472 contention
1473 [19049] math: [powerpc] erfc incorrect zero sign
1474 [19050] math: [powerpc] log* incorrect zero sign
1475 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1476 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1477 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1478 [19074] network: Data race in _res_hconf_reorder_addrs
1479 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1480 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1481 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1482 rounding modes
1483 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1484 ILP32
1485 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1486 [19086] manual: posix_fallocate64 documented argument order is wrong.
1487 [19088] math: lround, llround missing exceptions close to overflow
1488 threshold
1489 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1490 [19095] math: dbl-64 lrint incorrect for 64-bit long
1491 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1492 [19124] dynamic-link: ld.so failed to build with older assmebler
1493 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1494 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1495 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1496 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1497 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1498 pthread_setaffinity_np
1499 [19156] math: [ldbl-128] j0l spurious underflows
1500 [19164] nptl: tst-getcpu fails with many possible CPUs
1501 [19168] math: math/test-ildoubl and math/test-ldouble failure
1502 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1503 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1504 prelink
1505 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1506 (FE_NOMASK_ENV) do not clear SSE exceptions
1507 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1508 _int_new_arena/reused_arena
1509 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1510 [19201] math: dbl-64 remainder incorrect sign of zero result
1511 [19205] math: bits/math-finite.h conditions do not match math.h and
1512 bits/mathcalls.h
1513 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1514 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1515 for C99-based standards
1516 [19212] libc: features.h not -Wundef clean
1517 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1518 math-only
1519 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1520 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1521 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1522 disabled
1523 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1524 "inexact" exceptions
1525 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1526 arguments
1527 [19242] libc: strtol incorrect in Turkish locales
1528 [19243] malloc: reused_arena can pick an arena on the free list, leading
1529 to an assertion failure and reference count corruption
1530 [19253] time: tzset() ineffective when temporary TZ did not include DST
1531 rules
1532 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1533 [19270] math: [hppa] Shared libm missing __isnanl
1534 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1535 MAP_HUGETLB and MAP_STACK defines
1536 [19313] nptl: Wrong __cpu_mask for x32
1537 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1538 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1539 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1540 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1541 [19363] time: x32: times() return value wrongly truncates/sign extends
1542 from 32bit
1543 [19367] dynamic-link: Improve branch prediction on Silvermont
1544 [19369] network: Default domain name not reset by res_ninit when "search"
1545 / "domain" entry is removed from resolv.conf
1546 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1547 subnormals
1548 [19385] localedata: bg_BG: time separator should be colon, not comma
1549 [19408] libc: linux personality syscall wrapper may erroneously return an
1550 error on 32-bit architectures
1551 [19415] libc: dladdr returns wrong names on hppa
1552 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1553 IBM905, IBM907, and IBM909
1554 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1555 [19443] build: build failures with -DDEBUG
1556 [19451] build: Make check fails on test-double-vlen2
1557 [19462] libc: Glibc failed to build with -Os
1558 [19465] math: Wrong code with -Os
1559 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1560 -Os
1561 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1562 CPU's.
1563 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1564 [19486] math: S390: Math tests fail with "Exception Inexact set".
1565 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1566 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1567 architectures
1568 [19590] math: Fail to build shared objects that use libmvec.so functions.
1569 \f
1570 Version 2.22
1571
1572 * The following bugs are resolved with this release:
1573
1574 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1575 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1576 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1577 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1578 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1579 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1580 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1581 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1582 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1583 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1584 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1585 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1586 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1587 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1588 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1589 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1590 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1591 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1592 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1593 18657, 18676, 18694, 18696, 18887.
1594
1595 * Cache information can be queried via sysconf() function on s390 e.g. with
1596 _SC_LEVEL1_ICACHE_SIZE as argument.
1597
1598 * A buffer overflow in gethostbyname_r and related functions performing DNS
1599 requests has been fixed. If the NSS functions were called with a
1600 misaligned buffer, the buffer length change due to pointer alignment was
1601 not taken into account. This could result in application crashes or,
1602 potentially arbitrary code execution, using crafted, but syntactically
1603 valid DNS responses. (CVE-2015-1781)
1604
1605 * The time zone file parser has been made more robust against crafted time
1606 zone files, avoiding heap buffer overflows related to the processing of
1607 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1608 large time zone data files. Overly long time zone specifiers in the TZ
1609 variable no longer result in stack overflows and crashes.
1610
1611 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1612 for LD and GD on x86 and x86-64, has been implemented. You will need
1613 binutils-2.24 or later to enable this optimization.
1614
1615 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1616 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1617 Hat). These updates cause user visible changes, such as the fix for bug
1618 17998.
1619
1620 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1621 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1622 condition in some applications.
1623
1624 * Added vector math library named libmvec with the following vectorized x86_64
1625 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1626 pow, powf.
1627 The library can be disabled with --disable-mathvec. Use of the functions is
1628 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1629 Shared library libmvec.so is linked in as needed when using -lm (no need to
1630 specify -lmvec explicitly for not static builds).
1631 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1632
1633 * A new fmemopen implementation has been added with the goal of POSIX
1634 compliance. The new implementation fixes the following long-standing
1635 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1636 old implementation is still present for use be by existing binaries.
1637
1638 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1639 releases. It has been fixed to match 2.19 and older, but binaries built
1640 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1641
1642 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1643 Contributed by Roland McGrath (Google).
1644
1645 * The header <regexp.h> is deprecated, and will be removed in a future
1646 release. Use of this header will trigger a deprecation warning.
1647 Application developers should update their code to use <regex.h> instead.
1648
1649 This header was formerly part of SUS, but was deprecated in 1994 and
1650 removed from the standard in 2001. Also, the glibc implementation
1651 leaks memory. See BZ#18681 for more details.
1652 \f
1653 Version 2.21
1654
1655 * The following bugs are resolved with this release:
1656
1657 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1658 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1659 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1660 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1661 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1662 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1663 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1664 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1665 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1666 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1667 17892.
1668
1669 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1670 too little memory for the to-be-scanned arguments and overflow the
1671 allocated buffer. The implementation now correctly computes the required
1672 buffer size when using malloc, and switches to malloc from alloca as
1673 intended.
1674
1675 * A new semaphore algorithm has been implemented in generic C code for all
1676 machines. Previous custom assembly implementations of semaphore were
1677 difficult to reason about or ensure that they were safe. The new version
1678 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1679 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1680 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1681
1682 * Port to Altera Nios II has been contributed by Mentor Graphics.
1683
1684 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1685 implementations for powerpc64/powerpc64le.
1686 Implemented by Adhemerval Zanella (IBM).
1687
1688 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1689 and powerpc64le. This may improve lock scaling of existing programs on
1690 HTM capable systems. The lock elision code is only enabled with
1691 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1692 powerpc will issue a transaction abort on every syscall to avoid side
1693 effects being visible outside transactions.
1694
1695 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1696 AArch64. Contributed by ARM Ltd.
1697
1698 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1699
1700 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1701 under certain input conditions resulting in the execution of a shell for
1702 command substitution when the application did not request it. The
1703 implementation now checks WRDE_NOCMD immediately before executing the
1704 shell and returns the error WRDE_CMDSUB as expected.
1705
1706 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1707 processing format strings with a large number of format specifiers.
1708
1709 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1710 infinite loop if the DNS response contained a PTR record of an unexpected
1711 format.
1712
1713 * The minimum GCC version that can be used to build this version of the GNU
1714 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1715 still be used to compile programs using the GNU C Library.
1716
1717 * The GNU C Library is now built with -Werror by default. This can be
1718 disabled by configuring with --disable-werror.
1719
1720 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1721
1722 * The obsolete sigvec function has been removed. This was the original
1723 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1724 programs have been using instead for about 25 years. Of course, ABI
1725 compatibility for old binaries using sigvec remains intact.
1726
1727 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1728 with newer versions of bison.
1729
1730 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1731 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1732 registers overlay two consecutive single-precision registers. MIPS32R2
1733 introduced a new FPU mode (FR=1) where double-precision registers extend the
1734 corresponding single-precision registers which is incompatible with the
1735 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1736 require the use of FR=1 making a transition necessary. New o32 ABI
1737 extensions enable users to migrate over time from the original o32 ABI
1738 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1739 tracks the ABI of any loaded object and verifies that new objects are
1740 compatible. Mode transitions will also be requested as required and
1741 unsupportable objects will be rejected. The ABI checks include both soft and
1742 hard float ABIs for o32, n32 and n64.
1743
1744 GCC 5 with GNU binutils 2.25 onwards:
1745 It is strongly recommended that all o32 system libraries are built using the
1746 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1747 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1748 compiler using --with-fp-32=xx to set this by default.
1749 \f
1750 Version 2.20
1751
1752 * The following bugs are resolved with this release:
1753
1754 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1755 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1756 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1757 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1758 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1759 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1760 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1761 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1762 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1763 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1764 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1765 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1766 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1767 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1768 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1769
1770 * Reverted change of ABI data structures for s390 and s390x:
1771 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1772 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1773 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1774 are preserved pointing straight to the same implementation as the old ones.
1775 Given that, new callers will simply provide a too-big buffer to these
1776 functions. Any applications/libraries out there that embed jmp_buf or
1777 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1778 against 2.19 headers will have to rebuilt again. This is necessary in any
1779 case to revert the breakage in their ABI caused by the glibc change.
1780
1781 * Support for file description locks is added to systems running the
1782 Linux kernel. The standard file locking interfaces are extended to
1783 operate on file descriptions, not file descriptors, via the use of
1784 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1785 are associated with an open file instead of a process.
1786
1787 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1788
1789 * The minimum Linux kernel version that this version of the GNU C Library
1790 can be used with is 2.6.32.
1791
1792 * Running the testsuite no longer terminates as soon as a test fails.
1793 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1794 with PASS or FAIL lines for individual tests. A summary of the results is
1795 printed, including a list of failing lists, and "make check" exits with
1796 error status if there were any unexpected failures. "make check
1797 stop-on-test-failure=y" may be used to keep the old behavior.
1798
1799 * The am33 port, which had not worked for several years, has been removed
1800 from ports.
1801
1802 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1803 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1804 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1805 conflicted with POSIX (support for which was removed in 2.19), the
1806 interfaces those macros enabled remain available when compiling with
1807 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1808 test macros defined.
1809
1810 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1811
1812 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1813 This may improve lock scaling of existing programs on TX capable systems.
1814 The lock elision code is only built with --enable-lock-elision=yes and
1815 then requires a GCC version supporting the TX builtins. With lock elision
1816 default mutexes are elided via __builtin_tbegin, if the cpu supports
1817 transactions. By default lock elision is not enabled and the elision code
1818 is not built.
1819
1820 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1821 copy the path argument. This allowed programs to cause posix_spawn to
1822 deference a dangling pointer, or use an unexpected pathname argument if
1823 the string was modified after the posix_spawn_file_actions_addopen
1824 invocation.
1825
1826 * All supported architectures now use the main glibc sysdeps directory
1827 instead of some being in a separate "ports" directory (which was
1828 distributed separately before glibc 2.17).
1829
1830 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1831 On configurations that support it (all Linux configurations), it's now
1832 used regardless of the --enable-add-ons switch to configure. It is no
1833 longer possible to build such configurations without pthreads support.
1834
1835 * Locale names, including those obtained from environment variables (LANG
1836 and the LC_* variables), are more tightly checked for proper syntax.
1837 setlocale will now fail (with EINVAL) for locale names that are overly
1838 long, contain slashes without starting with a slash, or contain ".." path
1839 components. (CVE-2014-0475) Previously, some valid locale names were
1840 silently replaced with the "C" locale when running in AT_SECURE mode
1841 (e.g., in a SUID program). This is no longer necessary because of the
1842 additional checks.
1843
1844 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1845 with application code using Intel MPX instructions. (With all previous
1846 versions, the MPX register state could be clobbered when making calls
1847 into or out of a shared library.) Note that while the new dynamic
1848 linker is compatible with all known x86 hardware whether or not it
1849 supports Intel MPX, some x86 instruction-set emulators might fail to
1850 handle the new instruction encodings. This is known to affect Valgrind
1851 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1852 release), and might affect other tools that do instruction emulation.
1853
1854 * Support for loadable gconv transliteration modules has been removed.
1855 The support for transliteration modules has been non-functional for
1856 over a decade, and the removal is prompted by security defects. The
1857 normal gconv conversion modules are still supported. Transliteration
1858 with //TRANSLIT is still possible, and the //IGNORE specifier
1859 continues to be supported. (CVE-2014-5119)
1860
1861 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1862 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1863 resulting a denial-of-service security vulnerability in applications which
1864 use functions related to iconv. (CVE-2014-6040)
1865 \f
1866 Version 2.19
1867
1868 * The following bugs are resolved with this release:
1869
1870 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1871 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1872 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1873 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1874 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1875 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1876 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1877 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1878 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1879 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1880 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1881 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1882 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1883 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1884 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1885 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1886 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1887 16431, 16453, 16474, 16506, 16510, 16529
1888
1889 * Slovenian translations for glibc messages have been contributed by the
1890 Translation Project's Slovenian team of translators.
1891
1892 * The public headers no longer use __unused nor __block. This change is to
1893 support compiling programs that are derived from BSD sources and use
1894 __unused internally, and to support compiling with Clang's -fblock
1895 extension which uses __block.
1896
1897 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1898 large collation sequences to optimize multiple passes. This cache
1899 computation may overflow for large collation sequences and may cause a
1900 stack or buffer overflow. This is now fixed to use a slower algorithm
1901 which does not use a cache if there is an integer overflow.
1902
1903 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1904 rules for large collation sequences to optimize multiple passes and falls
1905 back to alloca if malloc fails, resulting in a possible stack overflow.
1906 The implementation now falls back to an uncached collation sequence lookup
1907 if malloc fails.
1908
1909 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1910 initialized for static applications resulting in the security feature
1911 being disabled. The pointer guard is now correctly initialized to a
1912 random value for static applications. Existing static applications need
1913 to be recompiled to take advantage of the fix (bug 15754).
1914
1915 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1916 to the d_name member of struct dirent, or omit the terminating NUL
1917 character. (Bugzilla #14699).
1918
1919 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1920 aligned_alloc functions could allocate too few bytes or corrupt the
1921 heap when passed very large allocation size values (Bugzilla #15855,
1922 #15856, #15857).
1923
1924 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1925 for AF_INET6 has been fixed (Bugzilla #16072).
1926
1927 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1928 pap_CW, quz_PE, the_NP.
1929
1930 * Substantially revised locales: gd_GB, ht_HT
1931
1932 * The LC_ADDRESS field was updated to support country_car for almost all
1933 supported locales.
1934
1935 * ISO 1427 definitions were updated.
1936
1937 * ISO 3166 definitions were updated.
1938
1939 * The localedef utility now supports --big-endian and --little-endian
1940 command-line options to generate locales for a different system from that
1941 for which the C library was built.
1942
1943 * Binary locale files now only depend on the endianness of the system for
1944 which they are generated and not on other properties of that system. As a
1945 consequence, binary files generated with new localedef may be incompatible
1946 with old versions of the GNU C Library, and binary files generated with
1947 old localedef may be incompatible with this version of the GNU C Library,
1948 in the following circumstances:
1949
1950 + Locale files may be incompatible on m68k systems.
1951
1952 + Locale archive files (but not separate files for individual locales) may
1953 be incompatible on systems where plain "char" is signed.
1954
1955 * The configure option --disable-versioning has been removed. Builds with
1956 --disable-versioning had not worked for several years.
1957
1958 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1959 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1960
1961 * SystemTap probes for malloc have been introduced.
1962
1963 * SystemTap probes for slow multiple precision fallback paths of
1964 transcendental functions have been introduced.
1965
1966 * Support for powerpc64le has been added.
1967
1968 * The soft-float powerpc port now supports e500 processors.
1969
1970 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1971
1972 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1973 set of header declarations that are enabled by default, even when other
1974 feature test macros or compiler options such as -std=c99 would otherwise
1975 disable some of those declarations.
1976
1977 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1978 conflict with POSIX. The libbsd-compat library (which was a dummy library
1979 that did nothing) has also been removed.
1980
1981 * Preliminary documentation about Multi-Thread, Async-Signal and
1982 Async-Cancel Safety has been added.
1983
1984 * Change of ABI data structures for s390 and s390x:
1985 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1986 allow for future hardware extensions. All library functions that accept or
1987 return these structures were versioned in order to provide backwards
1988 compatibility with existing code. However, not all cases can be handled
1989 with interface versioning. If an application embeds either structure into
1990 another structure and that structure is passed to another compilation unit
1991 which expects the newer larger-sized object then any access to the new
1992 fields will result in undefined behaviour. Similarly any access to fields
1993 that were moved by the enlarging of the structures will cause undefined
1994 behaviour. To fix the undefined behaviour all objects that are part of the
1995 final in-memory image must agree on the size of structures, and this may
1996 require recompilation.
1997 \f
1998 Version 2.18
1999
2000 * The following bugs are resolved with this release:
2001
2002 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2003 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2004 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2005 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2006 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2007 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2008 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2009 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2010 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2011 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2012 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2013 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2014 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2015 15755, 15759.
2016
2017 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2018 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2019 Distributions can re-enable building and using pt_chown via the new configure
2020 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2021 considerable security risks and should only be used if the distribution
2022 understands and accepts the risks.
2023
2024 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2025 #15078).
2026
2027 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2028 fixed (Bugzilla #15330).
2029
2030 * Add support for calling C++11 thread_local object destructors on thread
2031 and program exit. This needs compiler support for offloading C++11
2032 destructor calls to glibc.
2033
2034 * Improved worst case performance of libm functions with double inputs and
2035 output.
2036
2037 * Support for priority inherited mutexes in pthread condition variables on
2038 non-x86 architectures.
2039
2040 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2041
2042 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2043
2044 * Optimized string functions for ARM. Implemented by Will Newton and
2045 Richard Henderson.
2046
2047 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2048
2049 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2050 Richard Henderson.
2051
2052 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2053 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2054
2055 * Added a benchmark framework to track performance of functions in glibc.
2056
2057 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2058 It is based on draft TS 18661 and currently enabled as a GNU extension.
2059
2060 * On Linux, the clock function now uses the clock_gettime system call
2061 for improved precision, rather than old times system call.
2062
2063 * Added support for version-3 tz format files. This is needed when using
2064 the tz database release 2013e or later, and affects a few unusual cases --
2065 currently only TZ='America/Godthab' for time stamps after 2037.
2066
2067 * Added new API functions pthread_getattr_default_np and
2068 pthread_setattr_default_np to get and set the default pthread
2069 attributes of a process.
2070
2071 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2072 This may improve lock scaling of existing programs on TSX capable systems.
2073 When the --enable-lock-elision=yes parameter is specified at configure
2074 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2075 mutexes.
2076
2077 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2078 availability). Implemented by Ryan S. Arnold.
2079
2080 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2081 \f
2082 Version 2.17
2083
2084 * The following bugs are resolved with this release:
2085
2086 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2087 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2088 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2089 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2090 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2091 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2092 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2093 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2094 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2095 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2096 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2097 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2098 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2099 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2100 14879, 14889, 14893, 14898, 14914.
2101
2102 * Optimization of memcpy for MIPS.
2103
2104 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2105 EMFILE has been fixed (Bugzilla #14889).
2106
2107 * The add-on ports collection is now distributed in the "ports" subdirectory
2108 of the main GNU C Library distribution, rather than separately.
2109
2110 * Port to ARM AArch64 contributed by Linaro.
2111
2112 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2113 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2114 zEnterprise z196.
2115 Implemented by Andreas Krebbel.
2116
2117 * The new function secure_getenv allows secure access to the environment,
2118 returning NULL if running in a SUID/SGID process. This function replaces
2119 the internal function __secure_getenv.
2120
2121 * SystemTap static probes have been added into the dynamic linker.
2122 Implemented by Gary Benson.
2123
2124 * Optimizations of string functions strstr, strcasestr and memmem.
2125 Implemented by Maxim Kuvyrkov.
2126
2127 * The minimum Linux kernel version that this version of the GNU C Library
2128 can be used with is 2.6.16.
2129
2130 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2131 powerpc POWER7. Implemented by Will Schmidt.
2132
2133 * New configure option --disable-nscd builds the C library such that it
2134 never attempts to contact the Name Service Caching Daemon (nscd).
2135 New configure option --disable-build-nscd avoids building nscd itself;
2136 this is the default if --disable-nscd is used.
2137
2138 * Improved support for cross-compilation, including cross-testing and
2139 bootstrap builds without a previously built glibc.
2140
2141 * Several testsuite tests are now able to test multiple IFUNC variants of an
2142 interface, rather than just testing the one that would be chooen by
2143 default.
2144
2145 * New configure options --with-bugurl and --with-pkgversion, for
2146 distributors to use to embed their bug-reporting and package version
2147 information in --help and --version output.
2148
2149 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2150 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2151 allows creation of chroots without the procfs mounted on /proc.
2152
2153 * The `crypt' function now fails if passed salt bytes that violate the
2154 specification for those values. On Linux, the `crypt' function will
2155 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2156 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2157 when the mode is enabled.
2158
2159 * The `clock_*' suite of functions (declared in <time.h>) is now available
2160 directly in the main C library. Previously it was necessary to link with
2161 -lrt to use these functions. This change has the effect that a
2162 single-threaded program that uses a function such as `clock_gettime' (and
2163 is not linked with -lrt) will no longer implicitly load the pthreads
2164 library at runtime and so will not suffer the overheads associated with
2165 multi-thread support in other code such as the C++ runtime library.
2166
2167 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2168 sat_IN, and szl_PL.
2169 \f
2170 Version 2.16
2171
2172 * The following bugs are resolved with this release:
2173
2174 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2175 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2176 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2177 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2178 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2179 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2180 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2181 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2182 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2183 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2184 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2185 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2186 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2187 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2188 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2189 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2190 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2191 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2192 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2193 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2194 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2195 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2196 14277, 14278.
2197
2198 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2199 configuring glibc with:
2200 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2201 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2202 Implemented by H.J. Lu.
2203
2204 * ISO C11 support:
2205
2206 + define static_assert
2207
2208 + do not declare gets
2209
2210 + declare at_quick_exit and quick_exit also for ISO C11
2211
2212 + aligned_alloc. NB: The code is deliberately allows the size parameter
2213 to not be a multiple of the alignment. This is a moronic requirement
2214 in the standard but it is only a requirement on the caller, not the
2215 implementation.
2216
2217 + timespec_get added
2218
2219 + uchar.h support added
2220
2221 + CMPLX, CMPLXF, CMPLXL added
2222
2223 Implemented by Ulrich Drepper.
2224
2225 * Support for the IA-64 has been moved to ports.
2226
2227 * Remove support for anything but ELF binary format
2228
2229 * Checking versions of poll, ppoll added.
2230 Implemented by Ulrich Drepper.
2231
2232 * More generic and 64-bit performance optimizations to math functions.
2233 Implemented by Ulrich Drepper.
2234
2235 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2236 headers and functions available at compile time as they were before
2237 version 2.14. This option will be removed at some time in the future
2238 after the TI-RPC library becomes fully sufficient for the needs of
2239 existing applications.
2240
2241 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2242 Note that glibc is not expected to work with any Linux kernel version
2243 before 2.6.
2244
2245 * New header <sys/auxv.h> and function getauxval allowing easy access to
2246 the AT_* key-value pairs passed from the Linux kernel. The header also
2247 defines the HWCAP_* bits associated with the AT_HWCAP key.
2248
2249 * New locales: mag_IN
2250
2251 * New configure option --enable-systemtap builds SystemTap static probes
2252 into libc for setjmp and longjmp and into libpthread for various operations.
2253 So far the setjmp/longjmp probes and some of the libpthread probes are
2254 provided only for i*86 and x86_64.
2255 Implemented by Roland McGrath and Rayson Ho.
2256
2257 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2258
2259 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2260 and Will Schmidt.
2261
2262 * More optimized functions for SPARC. Implemented by David S. Miller.
2263
2264 * Improved support for cross-compilation, especially bootstrap builds
2265 without a previously built glibc.
2266
2267 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2268 Chris Metcalf from Tilera.
2269
2270 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2271 now supported for ARM processors.
2272
2273 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2274 as the name of the dynamic linker, to distinguish it from the
2275 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2276
2277 * Support for CRIS has been removed from ports.
2278
2279 * A new class of installed header has been documented for low-level
2280 platform-specific functionality. PowerPC added the first instance with a
2281 function to provide time base register access. Contributed by Tulio
2282 Magno Quites Machado Filho.
2283
2284 * ix86 configurations now install header files that are consistent with
2285 what x86-64 configurations install. These same header files can be used
2286 for -m32, -m64, or -mx32 builds.
2287 Contributed by H.J. Lu.
2288
2289 * Math library bug fixes. A thorough audit of all open math library bugs was
2290 conducted by Joseph Myers. Significant progress was made on many math
2291 library bugs resulting in more accurate exceptions and function results.
2292 Many thanks to all those that contributed including Andreas Jaeger for his
2293 patch review and work on the x87 trigonometric instruction issues.
2294
2295 * Timezone data is no longer installed. Timezone-related binaries and scripts
2296 will continue to be installed. Users should obtain their timezone data from
2297 their distribution provider or from the tzdata package at
2298 <ftp://munnari.oz.au/pub/>.
2299 \f
2300 Version 2.15
2301
2302 * The following bugs are resolved with this release:
2303
2304 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2305 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2306 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2307 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2308 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2309 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2310 13484, 13506, 13515, 13523, 13524, 13538, 13540
2311
2312 * New program pldd to list loaded object of a process
2313 Implemented by Ulrich Drepper.
2314
2315 * Add nss_db support back to glibc. No more dependency on Berkeley db
2316 and support for initgroups lookups.
2317 Implemented by Ulrich Drepper.
2318
2319 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2320 Contributed by HJ Lu.
2321
2322 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2323 Contributed by HJ Lu.
2324
2325 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2326 on x86-32 and x86-64.
2327 Contributed by Liubov Dmitrieva.
2328
2329 * Optimized strchr and strrchr for SSE on x86-32.
2330 Contributed by Liubov Dmitrieva.
2331
2332 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2333 for x86-64 and x86-32.
2334 Contributed by Liubov Dmitrieva.
2335
2336 * New interfaces: scandirat, scandirat64
2337 Implemented by Ulrich Drepper.
2338
2339 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2340 Implemented by Ulrich Drepper.
2341
2342 * nscd now also caches the netgroup database.
2343 Implemented by Ulrich Drepper.
2344
2345 * Integrate libm with gcc's -ffinite-math-only option.
2346 Implemented by Ulrich Drepper.
2347
2348 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2349 to math functions. Implemented by Ulrich Drepper.
2350
2351 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2352 Implemented by Ulrich Drepper.
2353
2354 * New Linux interfaces: process_vm_readv, process_vm_writev
2355
2356 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2357 Implemented by Ulrich Drepper.
2358
2359 * Optimized nearbyint and strcasecmp for PPC.
2360 Implemented by Adhemerval Zanella.
2361
2362 * New locales: bho_IN, unm_US, es_CU, ta_LK
2363 \f
2364 Version 2.14
2365
2366 * The following bugs are resolved with this release:
2367
2368 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2369 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2370 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2371 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2372 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2373 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2374 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2375 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2376 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2377 12795, 12811, 12813, 12814, 12841
2378
2379 * The RPC implementation in libc is obsoleted. Old programs keep working
2380 but new programs cannot be linked with the routines in libc anymore.
2381 Programs in need of RPC functionality must be linked against TI-RPC.
2382 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2383
2384 Visible changes of this change include (obviously) the inability to link
2385 programs using RPC functions without referencing the TI-RPC library and the
2386 removal of the RPC headers from the glibc headers.
2387 Implemented by Ulrich Drepper.
2388
2389 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2390 syncfs, setns, sendmmsg
2391
2392 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2393 yue_HK, lij_IT, mhr_RU
2394
2395 * New encodings: CP770, CP771, CP772, CP773, CP774
2396
2397 * New program sotruss to trace calls through PLTs
2398 Implemented by Ulrich Drepper.
2399
2400 * The malloc hook implementation is marked deprecated and will be removed
2401 from the default implementation in the next version. The design never
2402 worked ever since the introduction of threads. Even programs which do
2403 not create threads themselves can use multiple threads created internally.
2404 \f
2405 Version 2.13
2406
2407 * The following bugs are resolved with this release:
2408
2409 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2410 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2411 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2412 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2413 12378, 12394, 12397
2414
2415 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2416
2417 * POWER7 optimizations: memset, memcmp, strncmp
2418
2419 * New optimized string functions for x86-64: strnlen (SSE2),
2420 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2421 Implemented by Ulrich Drepper.
2422 \f
2423 Version 2.12
2424
2425 * The following bugs are resolved with this release:
2426
2427 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2428 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2429 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2430 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2431 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2432 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2433 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2434 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2435
2436 * New interfaces: pthread_getname_np, pthread_setname_np
2437
2438 * New Linux interface: recvmmsg
2439
2440 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2441
2442 * The dynamic linker now recognizes supported ABI versions from the
2443 EI_ABIVERSION field in the ELF header.
2444 Implemented by Ulrich Drepper.
2445
2446 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2447 will not be used to fill in password fields in the passwd.byname replies.
2448 Instead it is used to synthesize the shadow.byname table, should it be
2449 missing. This is a useful mode in some installations involving Solaris.
2450 Implemented by Ulrich Drepper.
2451
2452 * New locales: kok_IN, sq_MK, cv_RU
2453 \f
2454 Version 2.11
2455
2456 * The following bugs are resolved with this release:
2457
2458 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2459 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2460 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2461 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2462 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2463 10789, 10817, 10823, 10840, 10847
2464
2465 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2466 mkostemps64
2467 Implemented by Ulrich Drepper.
2468
2469 * Checking version of longjmp added that fails if an uninitialized stack
2470 frame would be created. Implemented by Ulrich Drepper.
2471
2472 * STT_GNU_IFUNC is now supported in static executables.
2473 Implemented by H.J. Lu.
2474
2475 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2476
2477 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2478 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2479 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2480 Contributed by H.J. Lu.
2481
2482 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2483 Implemented by Ulrich Drepper.
2484
2485 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2486 strstr, strcasestr.
2487 Contributed by H.J. Lu.
2488
2489 * Support for fma instruction in AVX on x86-64.
2490 Implemented by H.J. Lu and Ulrich Drepper.
2491
2492 * AVX support in x86-64 auditing support in ld.so.
2493 Implemented by H.J. Lu.
2494
2495 * STB_GNU_UNIQUE support added.
2496 Implemented by Ulrich Drepper.
2497
2498 * Implement second fallback mode for optimized DNS lookup for even more
2499 broken environments. If two requests from the same port are not
2500 handled correctly close the socket and open a new one before sending
2501 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2502 can be used to select this mode right away, instead of rediscovering the
2503 necessity is every process again.
2504 Implemented by Ulrich Drepper.
2505
2506 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2507 Implemented by Adam Tkac.
2508
2509 * Optimized iconv conversions for S390x.
2510 Implemented by Andreas Krebbel.
2511
2512 * Using condvars with PI mutexes is now more efficient due to kernel
2513 support for requeueing to PI futexes. NPTL support added for x86-64.
2514 Implemented by Ulrich Drepper.
2515
2516 * New locales: ps_AF, my_MM
2517 \f
2518 Version 2.10
2519
2520 * The following bugs are resolved with this release:
2521
2522 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2523 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2524 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2525 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2526 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2527
2528 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2529 Implemented by Ulrich Drepper.
2530
2531 * New Linux interfaces: accept4, fallocate, fallocate64.
2532 Implemented by Ulrich Drepper.
2533
2534 * Correct declarations of string function when used in C++ code. This
2535 could lead to compile errors for invalid C++ code.
2536
2537 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2538 now in POSIX.
2539
2540 * New POSIX 2008 interface: psiginfo
2541 Implemented by Ulrich Drepper.
2542
2543 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2544 Implemented by Ulrich Drepper.
2545
2546 * Support for selecting between multiple function definitions at runtime
2547 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2548
2549 * The libcrypt library can now use the hash function implementations in
2550 NSS. Implemented by Ulrich Drepper.
2551
2552 * The malloc implementation can be compiled to be less memory efficient
2553 but higher performing in multi-threaded programs.
2554 Implemented by Ulrich Drepper.
2555
2556 * New locales: nan_TW@latin, ks_IN
2557
2558 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2559 Implemented by Ulrich Drepper.
2560
2561 * Extended printf hook support. It is possible to use user-defined types
2562 and extend existing format specifiers.
2563 Implemented by Ulrich Drepper.
2564
2565 * Handling for group shadow files has been added.
2566 Implemented by Ulrich Drepper.
2567
2568 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2569 servers (the case, e.g., for some people using the built-in DNS
2570 server in ADSL modems/routers). There is a once-per-process timeout
2571 in case of a broken server. To avoid it, users can run nscd or put
2572 'options single-request' in /etc/resolv.conf.
2573 Implemented by Ulrich Drepper.
2574 \f
2575 Version 2.9
2576
2577 * The following bugs are resolved with this release:
2578
2579 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2580 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2581 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2582 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2583 6968, 6974, 6980, 6995, 7008, 7009, 7029
2584
2585 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2586 up at the same time. Implemented by Ulrich Drepper.
2587
2588 * TLS descriptors for LD and GD on x86 and x86-64.
2589 Implemented by Alexandre Oliva.
2590
2591 * getaddrinfo now handles DCCP and UDPlite.
2592 Implemented by Ulrich Drepper.
2593
2594 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2595 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2596 Implemented by Ulrich Drepper.
2597
2598 * New implementation of memmem, strstr, and strcasestr which is O(n).
2599 Implemented by Eric Blake.
2600
2601 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2602
2603 * Implement "e" option for popen to open file descriptor with the
2604 close-on-exec flag set. Implemented by Ulrich Drepper.
2605
2606 * Implement "b" mode for fmemopen. In this mode writes writes don't
2607 implicitly add a NUL byte and seeks from the end of the buffer really
2608 use the buffer end, not the string length as the basis.
2609 Implemented by Ulrich Drepper.
2610
2611 * Many functions, exported and internal, now atomically set the close-on-exec
2612 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2613
2614 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2615 Sinhala)
2616 Implemented by Pravin Satpute.
2617
2618 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2619 \f
2620 Version 2.8
2621
2622 * The following bugs are resolved with this release:
2623
2624 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2625 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2626 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2627 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2628 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2629 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2630 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2631
2632 * New locales: bo_CN, bo_IN, shs_CA.
2633
2634 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2635
2636 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2637 Implemented by Pravin Satpute.
2638
2639 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2640
2641 * nscd caches now all timeouts for DNS entries
2642 Implemented by Ulrich Drepper.
2643
2644 * nscd is more efficient and wakes up less often.
2645 Implemented by Ulrich Drepper.
2646
2647 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2648 vdprintf, and obstack_vprintf.
2649 Implemented by Jakub Jelinek.
2650
2651 * Faster memset for x86-64.
2652 Implemented by Harsha Jagasia and H.J. Lu.
2653
2654 * Faster memcpy on x86.
2655 Implemented by Ulrich Drepper.
2656
2657 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2658 Implemented by Ulrich Drepper.
2659
2660 * Faster sqrt and sqrtf implemention for some PPC variants.
2661 Implemented by Steven Munroe.
2662 \f
2663 Version 2.7
2664
2665 * The following bugs are resolved with this release:
2666
2667 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2668 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2669 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2670 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2671 5071, 5103, 5104, 5112, 5113, 5184, 5186
2672
2673 * More checking functions: fread, fread_unlocked, open*, mq_open.
2674 Implemented by Jakub Jelinek and Ulrich Drepper.
2675
2676 * Extend fortification to C++. Implemented by Jakub Jelinek.
2677
2678 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2679 by not recognizing 'a' as a modifier when those specs are requested.
2680 Implemented by Jakub Jelinek.
2681
2682 * PPC optimizations to math and string functions.
2683 Implemented by Steven Munroe.
2684
2685 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2686 options to be passed. Implemented by Ulrich Drepper.
2687
2688 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2689
2690 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2691 Implemented by Ulrich Drepper.
2692
2693 * Handle private futexes in the NPTL implementation.
2694 Implemented by Jakub Jelinek and Ulrich Drepper.
2695
2696 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2697 Implemented by Roland McGrath and Ulrich Drepper.
2698
2699 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2700
2701 * SHA-256 and SHA-512 based password encryption.
2702 Implemented by Ulrich Drepper.
2703
2704 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2705 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2706 yo_NG.
2707
2708 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2709 Implemented by Ulrich Drepper.
2710 \f
2711 Version 2.6
2712
2713 * The following bugs are resolved with this release:
2714
2715 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2716 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2717 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2718 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2719 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2720 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2721 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2722 4702, 4858
2723
2724 * New Linux interfaces: epoll_pwait, sched_getcpu.
2725
2726 * New generic interfaces: strerror_l.
2727
2728 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2729 \f
2730 Version 2.5
2731
2732 * The following bugs are resolved with this release:
2733
2734 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2735 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2736 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2737 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2738 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2739 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2740 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2741 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2742 3225, 3251, 3252, 3253, 3273
2743
2744 * For Linux, the sorting of addresses returned by getaddrinfo now also
2745 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2746 Implemented by Ulrich Drepper.
2747
2748 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2749 Implemented by Ulrich Drepper.
2750
2751 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2752
2753 * New iconv module for MIK. Contributed by Alexander Shopov.
2754
2755 * For sites with broken group and/or passwd database, the auto-propagate
2756 option of nscd can prevent creating ID lookup entries from the results
2757 of a name lookup and vice versa. This usually is no problem but some
2758 site might have problems with the default behavior.
2759 Implemented by Ulrich Drepper.
2760
2761 * Iterating over entire database in NIS can be slow. With the
2762 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2763 to trade time for memory. The entire database will be read at once.
2764 Implemented by Ulrich Drepper.
2765
2766 * The interfaces introduced in RFC 3542 have been implemented by
2767 Ulrich Drepper.
2768
2769 * Support for the new ELF hash table format was added by Ulrich Drepper.
2770
2771 * Support for priority inheritance mutexes added by Jakub Jelinek and
2772 Ulrich Drepper.
2773
2774 * Support for priority protected mutexes added by Jakub Jelinek.
2775
2776 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2777 or_IN, csb_PL, fy_NL, sr_ME.
2778 \f
2779 Version 2.4
2780
2781 * More overflow detection functions.
2782
2783 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2784 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2785
2786 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2787 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2788 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2789 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2790 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2791 by Masahide Washizawa.
2792
2793 * It is now possible to install audit libraries for the dynamic linker using
2794 LD_AUDIT. Implemented by Ulrich Drepper.
2795
2796 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2797 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2798 For a libc and libpthread that works well on Linux 2.4 kernels, we
2799 recommend using the stable 2.3 branch.
2800
2801 * The new function open_wmemstream acts like open_memstream,
2802 but uses a wchar_t wide character buffer.
2803
2804 * The new function ppoll is an improved version of poll, similar to pselect.
2805
2806 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2807 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2808 renameat, unlinkat, linkat, symlinkat, readlinkat.
2809
2810 * New Linux kernel system calls: unshare,
2811 inotify_init, inotify_add_watch, inotify_rm_watch.
2812
2813 * The euidaccess function is now also known by the alias eaccess,
2814 for compatibility with some other systems.
2815
2816 * Timezone data updated to 2006b version.
2817 \f
2818 Version 2.3.6
2819
2820 * The following bugs are resolved with this release:
2821
2822 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2823 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2824 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2825 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2826 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2827 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2828
2829 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2830
2831 * As of this release, GCC 4 can be used to compile the C Library.
2832
2833 * Timezone data updated to 2005m version.
2834 \f
2835 Version 2.3.5
2836
2837 * The following bugs are resolved with this release:
2838
2839 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2840 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2841 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2842 777, 787, 821, 822, 823, 825
2843
2844 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2845 \f
2846 Version 2.3.4
2847
2848 * Support for RFC 3678. Real implementations exist only for Linux so far.
2849 Implemented by Ulrich Drepper.
2850
2851 * nscd can now cache entries persistently. Expiring entries are reloaded.
2852 For speedups the cache can be shared in memory with client processes.
2853 Implemented by Ulrich Drepper.
2854
2855 * nscd can now perform SELinux checks.
2856 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2857
2858 * getaddrinfo queries are now cached. Canonical name lookup is performed
2859 efficiently.
2860 Implemented by Ulrich Drepper.
2861
2862 * The nothrow function attribute is used when headers are used by gcc when
2863 compiling C code. This can avoid significant amounts of exception
2864 handling data.
2865
2866 * The malloc functions perform more error checking and are stricter when
2867 it comes to reacting on errors. The default action is to terminate
2868 the process after showing an error message. Implemented by Ulrich Drepper.
2869
2870 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2871 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2872
2873 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2874 namespaces using the new function dlmopen(). This feature is of course,
2875 like most other dynamic loading functionality, not available in statically
2876 linked applications. Implemented by Ulrich Drepper.
2877
2878 * Low-overhead boundary checking variants of string and some stdio functions
2879 were added. These are to be used in conjunction with a gcc patch by
2880 Jakub Jelinek which adds calls to these functions if possible.
2881 Implemented by Jakub Jelinek and Ulrich Drepper.
2882
2883 * Old code for several operating systems and machine architectures that
2884 have not been in working condition in a long time have been removed from
2885 the main source tree maintained by the GNU C Library's maintainers.
2886 These files are now reside in the separate `ports' source module
2887 that is usable as an add-on when building the library.
2888 \f
2889 Version 2.3.3
2890
2891 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2892 interrogate the dynamic linker, compatible with the Solaris interface.
2893
2894 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2895 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2896
2897 * getifaddrs now uses the netlink interface on Linux to get its information.
2898 Implemented by Thorsten Kukuk.
2899
2900 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2901 Implemented by Ulrich Drepper.
2902
2903 * support for non-executable stacks on x86 has been added. Changes mostly
2904 by Roland McGrath.
2905
2906 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2907 and Ulrich Drepper.
2908
2909 * getaddrinfo now performs destination address selection according to
2910 RFC 3484.
2911 \f
2912 Version 2.3.2
2913
2914 * Thread-safe interfaces for many functions that access locale data
2915 were added in version 2.3, but these features were omitted from NEWS.
2916 Many functions have variants with an `_l' suffix that take a `locale_t'
2917 object as a parameter rather than consulting the current locale.
2918 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2919 create and maintain `locale_t' objects. Additionally, the new function
2920 `uselocale' sets "the current locale" (as used by functions not so
2921 parameterized) set for an individual thread. These features were added
2922 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2923
2924 * The functions getresuid, getresgid, setresuid, and setresgid, which
2925 have long been available on Linux, are now declared in <unistd.h>
2926 and are now also available on the Hurd.
2927
2928 * ELF thread-local storage support (TLS) now works on x86-64.
2929
2930 * The new dynamic string token $LIB is expanded in shared library names.
2931 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2932
2933 * Aldy Hernandez contributed complete software floating point support for
2934 PowerPC machines with no FPU.
2935
2936 * fexecve is implemented on Linux.
2937
2938 * The `btowc' function should work at least twice as fast due to
2939 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2940
2941 * With approriate thread add-ons cancelable functions are now implemented
2942 in libc.so as well. No need to call the function in libpthread. This
2943 change allowed to finally disable the incorrect and expensive handling
2944 of weak definition in ld.so.
2945
2946 * Yet more PLT entries in libc.so have been removed. We finally arrived
2947 at the bare minimum. Startup times improved appropriately.
2948
2949 * Support for the new Linux/x86 system call interface was added. The
2950 AT_SYSINFO auxiliary vector entry is recognized and handled.
2951 \f
2952 Version 2.3
2953
2954 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2955 charsets.
2956
2957 * iconv (the program and the interface) now accepts empty names (excluding
2958 options like //TRANSLIT) to mean "use charset of current locale".
2959
2960 * localedef can now transliterate characters in strings which are not in
2961 the provided charmap. The information from the input locale is used.
2962
2963 * Prelinking support was added for ELF targets. This requires additional
2964 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2965
2966 * Read-only stdio streams now use mmap to speed up operation by eliminating
2967 copying and buffer underflows. To use add 'm' to the mode string of
2968 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2969
2970 * The malloc functions were completely rewritten by Wolfram Gloger based
2971 on Doug Lea's malloc-2.7.0.c.
2972
2973 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2974 implementation of regex.
2975
2976 * Bruno Haible upgraded the iconv and locale implementation to support
2977 Unicode 3.2.
2978
2979 * Contents of the LC_* and LANG environment variables in the CEN style are
2980 not recognized anymore. It never was used. Change by Ulrich Drepper.
2981
2982 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2983 thread-local storage (TLS) ABI on some platforms.
2984 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2985
2986 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2987 EUC-JISX0213, and TSCII.
2988
2989 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2990 BSD-compatible interface for getting all network interface addresses.
2991 Implementation for IPv4 by Roland McGrath.
2992
2993 * Loading of locale data is faster due to the introduction of a locale
2994 archive. Implemented by Roland McGrath and Ulrich Drepper.
2995
2996 * Startup times are significantly reduced by not using exported functions
2997 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2998 and Ulrich Drepper.
2999
3000 * Steven Munroe contributed a port to PowerPC64/Linux.
3001 \f
3002 Version 2.2.6
3003
3004 * The Hurd now uses the GNU libio implementation of stdio rather than the
3005 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3006
3007 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3008 with the same behavior as the Linux system call of the same name.
3009 \f
3010 Version 2.2.5
3011
3012 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3013 128-bit long double format.
3014
3015 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3016 IBM1161, and IBM1162 charsets.
3017
3018 * Andreas Jaeger contributed a port to x86-64/Linux.
3019
3020 * Peter Bruin contributed a port to PowerPC/Hurd.
3021
3022 * libc functions using I/O streams now can handle wide-oriented streams
3023 as well.
3024
3025 * optimizations in the dynamic linker. Binaries created by recent binutils
3026 versions start up quicker due to reduced time spent on relocations.
3027
3028 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3029 \f
3030 Version 2.2.4
3031
3032 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3033 asinh, atanh, j0 for the 128-bit long double format.
3034
3035 * Bruno Haible updated all the code handling Unicode in some form to
3036 support Unicode 3.1.
3037
3038 * Speed of regex for single-byte locales is back to previous levels.
3039 Patch by Isamu Hasegawa.
3040
3041 * Alpha, SPARC, and IA-64 now also using floating stacks.
3042
3043 * Startup time of internationalized applications greatly improved through
3044 iconv cache. Use iconvconfig to generate the cache file.
3045 Contributed by Ulrich Drepper.
3046
3047 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3048 severe performance problems. Patch by David Mosberger.
3049
3050 * The Hurd port got a lot more functionality like AIO, various stdio
3051 extensions, etc. Mainly done by Roland McGrath.
3052
3053 * mtrace can now lookup symbols in shared libraries.
3054 \f
3055 Version 2.2.3
3056
3057 * Intel's IA-64 math library is largely integrated. It provides fast and
3058 accurate implementatations for most basic and standard math functions
3059 in float, double, and long double format.
3060
3061 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3062 and asin for the 96-bit long double format and asin, log, tan for the
3063 128-bit long double format.
3064
3065 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3066 The basic double functions exist today. Contributed by Abraham Ziv
3067 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3068 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3069
3070 * An asynchronous name lookup library was added. The interface is designed
3071 after POSIX AIO. The proposal was circulated beforehand to get comments.
3072 No negative ones came in. Implemented by Ulrich Drepper.
3073
3074 * Port to S390/64bit contributed by Martin Schwidefsky
3075 <schwidefsky@de.ibm.com>.
3076
3077 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3078 of functions for Linux/IA-64.
3079
3080 * The RPC code is now thread safe. Threads can now use the same service
3081 of different services at the same time. Patch by Eric Norum
3082 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3083
3084 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3085 family of functions for Linux/S390.
3086
3087 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3088 of functions for Linux/x86.
3089
3090 * Port to Linux/CRIS contributed by Axis Communications.
3091 \f
3092 Version 2.2.2
3093
3094 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3095 we can now check for namespace violations and missing declarations. The
3096 result is that almost all headers are now Unix-compliant (as defined in
3097 the upcoming XPG6). The negative side is that some programs might need
3098 corrections, too, if they depend on the incorrect form of the headers in
3099 previous versions which defined too many symbols and included too many
3100 other headers.
3101
3102 * regex now handles multibyte character sets correctly.
3103 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3104
3105 * iconv (the program) does now conform to the upcoming XPG6 and handles
3106 charmaps. Instead of the charset names the path of charmaps can be
3107 provided and the conversion happens based on this data.
3108 Contributed by Ulrich Drepper.
3109
3110 * The locale program now provides detailed information about the installed
3111 locales. While
3112
3113 locale -a
3114
3115 only lists the names of the supported locales
3116
3117 locale -a --verbose
3118
3119 provides details such as country, language, and codeset name.
3120 Contributed by Ulrich Drepper.
3121 \f
3122 Version 2.2.1
3123
3124 * The gencat program now parses the input file according to the charset
3125 selected by the LC_CTYPE category. This is important for stateful
3126 character sets. To make generating catalogs easier there is a way
3127 to overwrite the charset selected by the locale: before the first
3128 message or $ quote line the catalog can contain a line like
3129
3130 $ codeset=ISO-8859-2
3131
3132 to select the charset (ISO-8859-2 in this case).
3133
3134 Implemented by Shinya Hanataka and Ulrich Drepper.
3135
3136 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3137 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3138 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3139
3140 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3141 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3142
3143 * The LANGUAGE environment variable is now ignored unless the locale is
3144 changed from the default "C" locale.
3145
3146 * The usual bug fixes.
3147 \f
3148 Version 2.2
3149
3150 * Greg McGary added runtime support for bounds checking using gcc's
3151 new -fbounded-pointers option. ix86 target is complete. PowerPC
3152 is in progress.
3153
3154 * Thorsten Kukuk added secure mode support to nscd.
3155
3156 * The Berkeley db libraries have been removed.
3157
3158 Related, the nss_db module is now in a separate package since it
3159 obviously requires a database library being available.
3160
3161 * Wide character I/O streams implemented by Ulrich Drepper.
3162
3163 * Functions from the extended socket API added by Ulrich Drepper.
3164
3165 * Functions feenableexcept and fedisableexcept to control the
3166 behaviour of individual exceptions have been added by Andreas Jaeger.
3167
3168 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3169
3170 * The resolver code has been updated from bind 8.2.3-T5B which supports
3171 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3172 and Mark Kettenis.
3173
3174 This change could in some situations effect backward compatibility. Since
3175 now `_res' is a thread-local instead of a global variable, modifying it
3176 in one thread does not have any effect in other threads.
3177
3178 The resolver library was also extended to allow IPv6 as the transport
3179 protocol for the requests. This work was done by Stig Venaas.
3180
3181 * Compatibility code for K&R C compilers has been removed from the
3182 header files. A ISO C compiler is needed to use the library
3183 (conforming to either C89 or C99 standard).
3184
3185 * Complete rewrite of the localedef program to support multibyte character
3186 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3187 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3188 Implemented by Ulrich Drepper.
3189
3190 Bruno Haible significantly improved the generation and use of the data
3191 structures for the wide character tables.
3192
3193 * Plural handling in gettext implemented by Ulrich Drepper.
3194
3195 * The utmp daemon has been removed.
3196
3197 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3198
3199 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3200 and Yutaka Niibe.
3201
3202 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3203
3204 * POSIX barriers implemented by Kaz Kylheku.
3205
3206 * POSIX spawn function family implemented by Ulrich Drepper.
3207
3208 * POSIX spinlocks are now available.
3209
3210 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3211
3212 * the configure option --enable-kernel=X.Y.Z allows to strip out
3213 compatibility for kernel versions before X.Y.Z. This is currently only
3214 implemented for Linux.
3215
3216 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3217 field sin6_scope_id. This means that all programs using IPv6 should be
3218 recompiled. Don't expect binary compatibility with previous glibc
3219 versions.
3220
3221 * various conversion modules for IBM character sets contributed by
3222 Masahide Washizawa.
3223
3224 * IA-64 port by Jes Sorensen and HJ Lu.
3225 \f
3226 Version 2.1.3
3227
3228 * bug fixes
3229
3230 \f
3231 Version 2.1.2
3232
3233 * bug fixes
3234
3235 \f
3236 Version 2.1.1
3237
3238 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3239
3240 * New xdr functions are added; some rpc functions are now 64bit clean.
3241
3242 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3243
3244 * Fixed known incompatibilities with glibc 2.0.
3245
3246 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3247
3248 * Optimized a number of functions (especially the ELF dynamic loader).
3249
3250 * Update timezone data files.
3251
3252 * lots of charmaps corrections
3253
3254 * some new locale definitions and charmaps
3255
3256 \f
3257 Version 2.1
3258
3259 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3260 conform to POSIX member type requirements. Symbol versions have been
3261 adjusted as needed within the library, and for direct use by applications,
3262 but there is potential for problems if third-party libraries use
3263 struct timeval as part of their interface. This does not present
3264 a problem for X and other "essential" system libraries.
3265
3266 * An additional locale model to support C++ Standard Library locale
3267 model and probably more was implemented by Ulrich Drepper.
3268
3269 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3270 symbol level.
3271
3272 * Miles Bader provided the `argp' function family to support hierachical
3273 command line argument parsing, layered on top of getopt.
3274
3275 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3276
3277 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3278 numbers.
3279
3280 * scanf recognizes the %a and %A format for scanning floating point numbers.
3281
3282 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3283 information and interfaces for the available integer types.
3284
3285 * about 130 new math functions were added to implement the ISO C9x math
3286 library.
3287
3288 * the new header <complex.h> contains definitions of the complex math
3289 functions from ISO C 9X.
3290
3291 * the new header <tgmath.h> defines generic macros to use complex or
3292 real valued functions.
3293
3294 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3295
3296 * Andreas Jaeger provided a test suite for the math library.
3297
3298 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3299
3300 * Ulrich Drepper added character set conversion functions (iconv).
3301
3302 * Optimized string functions have been added.
3303
3304 * The localedata addon is now part of glibc.
3305
3306 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3307
3308 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3309 daemon for NSS (nscd).
3310
3311 Missing a better place here are some numbers on improvements. Under
3312 Linux 2.1.125 un-tar-ing the kernel sources takes
3313
3314 user system wall
3315
3316 using local files 12.19s 6.88s 22.91s
3317
3318 using NIS 13.92s 8.91s 26.34s
3319
3320 using NIS & nscd 10.37s 7.34s 25.30s
3321
3322 using NIS+ 27.57s 30.37s 640.46s
3323
3324 using NIS+ & nscd 10.25s 7.83s 26.51s
3325
3326 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3327
3328 Keep in mind that non-namelookup related operations dominate above times.
3329 It was just a common complain that using NIS+ unpacking the kernel is
3330 horribly slow.
3331
3332 [1] The old nscd implementation is not available anymore since it was
3333 distributed with glibc up to version 2.0.98 and thus is now replaced.
3334
3335 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3336
3337 * Mark Kettenis provided a Hesiod NSS module.
3338
3339 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3340 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3341
3342 * rcmd can now handle netgroups (Dick Streefland).
3343
3344 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3345 Bambrough.
3346
3347 * Support for the IPv6 protocol has been added to the socket API, as per the
3348 latest draft standards.
3349
3350 * Support for Linux 2.2 has been added.
3351
3352 * Interface changes relative to the latest 2.0.x release:
3353 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3354 addseverity NEW: Unix98
3355 alphasort64 NEW: LFS
3356 argp_err_exit_status NEW: argp, GNU ext
3357 argp_error NEW: argp, GNU ext
3358 argp_failure NEW: argp, GNU ext
3359 argp_help NEW: argp, GNU ext
3360 argp_parse NEW: argp, GNU ext
3361 argp_program_bug_address NEW: argp, GNU ext
3362 argp_program_version NEW: argp, GNU ext
3363 argp_program_version_hook NEW: argp, GNU ext
3364 argp_state_help NEW: argp, GNU ext
3365 argp_usage NEW: argp, GNU ext
3366 authdes_create NEW: Secure RPC
3367 authdes_getucred NEW: Secure RPC
3368 authdes_pk_create NEW: Secure RPC
3369 backtrace NEW: GNU ext.
3370 backtrace_symbols NEW: GNU ext.
3371 backtrace_symbols_fd NEW: GNU ext.
3372 cacos NEW: ISO C 9x
3373 cacosf NEW: ISO C 9x
3374 cacosh NEW: ISO C 9x
3375 cacoshf NEW: ISO C 9x
3376 cacoshl NEW: ISO C 9x
3377 cacosl NEW: ISO C 9x
3378 capget NEW: kernel
3379 capset NEW: kernel
3380 carg NEW: ISO C 9x
3381 cargf NEW: ISO C 9x
3382 cargl NEW: ISO C 9x
3383 casin NEW: ISO C 9x
3384 casinf NEW: ISO C 9x
3385 casinh NEW: ISO C 9x
3386 casinhf NEW: ISO C 9x
3387 casinhl NEW: ISO C 9x
3388 casinl NEW: ISO C 9x
3389 catan NEW: ISO C 9x
3390 catanf NEW: ISO C 9x
3391 catanh NEW: ISO C 9x
3392 catanhf NEW: ISO C 9x
3393 catanhl NEW: ISO C 9x
3394 catanl NEW: ISO C 9x
3395 cbc_crypt NEW: Secure RPC
3396 ccos NEW: ISO C 9x
3397 ccosf NEW: ISO C 9x
3398 ccosh NEW: ISO C 9x
3399 ccoshf NEW: ISO C 9x
3400 ccoshl NEW: ISO C 9x
3401 ccosl NEW: ISO C 9x
3402 cexp NEW: ISO C 9x
3403 cexpf NEW: ISO C 9x
3404 cexpl NEW: ISO C 9x
3405 cimag NEW: ISO C 9x
3406 cimagf NEW: ISO C 9x
3407 cimagl NEW: ISO C 9x
3408 clearerr_locked REMOVED
3409 clntunix_create NEW: sunrpc ext
3410 clog NEW: ISO C 9x
3411 clog10 NEW: ISO C 9x
3412 clog10f NEW: ISO C 9x
3413 clog10l NEW: ISO C 9x
3414 clogf NEW: ISO C 9x
3415 clogl NEW: ISO C 9x
3416 conj NEW: ISO C 9x
3417 conjf NEW: ISO C 9x
3418 conjl NEW: ISO C 9x
3419 cpow NEW: ISO C 9x
3420 cpowf NEW: ISO C 9x
3421 cpowl NEW: ISO C 9x
3422 cproj NEW: ISO C 9x
3423 cprojf NEW: ISO C 9x
3424 cprojl NEW: ISO C 9x
3425 creal NEW: ISO C 9x
3426 crealf NEW: ISO C 9x
3427 creall NEW: ISO C 9x
3428 creat64 NEW: LFS
3429 csin NEW: ISO C 9x
3430 csinf NEW: ISO C 9x
3431 csinh NEW: ISO C 9x
3432 csinhf NEW: ISO C 9x
3433 csinhl NEW: ISO C 9x
3434 csinl NEW: ISO C 9x
3435 csqrt NEW: ISO C 9x
3436 csqrtf NEW: ISO C 9x
3437 csqrtl NEW: ISO C 9x
3438 ctan NEW: ISO C 9x
3439 ctanf NEW: ISO C 9x
3440 ctanh NEW: ISO C 9x
3441 ctanhf NEW: ISO C 9x
3442 ctanhl NEW: ISO C 9x
3443 ctanl NEW: ISO C 9x
3444 des_setparity NEW: Secure RPC
3445 ecb_crypt NEW: Secure RPC
3446 endutxent NEW: Unix98
3447 exp10 NEW: ISO C 9x
3448 exp10f NEW: ISO C 9x
3449 exp10l NEW: ISO C 9x
3450 exp2 NEW: ISO C 9x
3451 exp2f NEW: ISO C 9x
3452 exp2l NEW: ISO C 9x
3453 fattach NEW: STREAMS
3454 fdetach NEW: STREAMS
3455 fdim NEW: ISO C 9x
3456 fdimf NEW: ISO C 9x
3457 fdiml NEW: ISO C 9x
3458 feclearexcept NEW: ISO C 9x
3459 fegetenv NEW: ISO C 9x
3460 fegetexceptflag NEW: ISO C 9x
3461 fegetround NEW: ISO C 9x
3462 feholdexcept NEW: ISO C 9x
3463 feof_locked REMOVED
3464 feraiseexcept NEW: ISO C 9x
3465 ferror_locked REMOVED
3466 fesetenv NEW: ISO C 9x
3467 fesetexceptflag NEW: ISO C 9x
3468 fesetround NEW: ISO C 9x
3469 fetestexcept NEW: ISO C 9x
3470 feupdateenv NEW: ISO C 9x
3471 fflush_locked REMOVED
3472 ffsl NEW: GNU ext.
3473 ffsll NEW: GNU ext.
3474 fgetpos64 NEW: LFS
3475 fgets_unlocked NEW: GNU ext.
3476 fileno_locked REMOVED
3477 fma NEW: ISO C 9x
3478 fmaf NEW: ISO C 9x
3479 fmal NEW: ISO C 9x
3480 fmax NEW: ISO C 9x
3481 fmaxf NEW: ISO C 9x
3482 fmaxl NEW: ISO C 9x
3483 fmin NEW: ISO C 9x
3484 fminf NEW: ISO C 9x
3485 fminl NEW: ISO C 9x
3486 fmtmsg NEW: Unix98
3487 fopen64 NEW: LFS
3488 fputc_locked REMOVED
3489 fputs_unlocked NEW: GNU ext.
3490 fread_unlocked NEW: GNU ext.
3491 freopen64 NEW: LFS
3492 fseeko NEW: Unix98
3493 fsetpos64 NEW: LFS
3494 fstatfs64 NEW: LFS
3495 fstatvfs NEW: Unix98
3496 fstatvfs64 NEW: LFS
3497 ftello NEW: Unix98
3498 ftello64 NEW: LFS
3499 ftruncate64 NEW: LFS
3500 ftw64 NEW: LFS
3501 fwrite_unlocked NEW: GNU ext.
3502 gai_strerror NEW: IPv6
3503 gamma_r REMOVED
3504 gammaf_r REMOVED
3505 gammal_r REMOVED
3506 getchar_locked REMOVED
3507 getdate NEW: Unix98
3508 getdate_err NEW: Unix98
3509 getdate_r NEW: GNU ext.
3510 getmsg NEW: STREAMS
3511 getnameinfo NEW: IPv6
3512 getnetname NEW: Secure RPC
3513 getpmsg NEW: STREAMS
3514 getpt NEW: Unix98 PTY
3515 getrlimit64 NEW: LFS
3516 getutxent NEW: Unix98
3517 getutxid NEW: Unix98
3518 getutxline NEW: Unix98
3519 glob64 NEW: GNU ext.
3520 globfree64 NEW: GNU ext.
3521 gnu_get_libc_release NEW: GNU ext.
3522 gnu_get_libc_version NEW: GNU ext.
3523 grantpt NEW: Unix98 PTY
3524 host2netname NEW: Secure RPC
3525 iconv NEW: iconv
3526 iconv_close NEW: iconv
3527 iconv_open NEW: iconv
3528 if_freenameindex NEW: IPv6
3529 if_indextoname NEW: IPv6
3530 if_nameindex NEW: IPv6
3531 if_nametoindex NEW: IPv6
3532 in6addr_any NEW: IPv6
3533 in6addr_loopback NEW: IPv6
3534 inet6_isipv4mapped NEW: IPv6
3535 isastream NEW: STREAMS
3536 iswblank NEW: GNU ext.
3537 key_decryptsession NEW: Secure RPC
3538 key_decryptsession_pk NEW: Secure RPC
3539 key_encryptsession NEW: Secure RPC
3540 key_encryptsession_pk NEW: Secure RPC
3541 key_gendes NEW: Secure RPC
3542 key_get_conv NEW: Secure RPC
3543 key_secretkey_is_set NEW: Secure RPC
3544 key_setnet NEW: Secure RPC
3545 key_setsecret NEW: Secure RPC
3546 llrint NEW: ISO C 9x
3547 llrintf NEW: ISO C 9x
3548 llrintl NEW: ISO C 9x
3549 llround NEW: ISO C 9x
3550 llroundf NEW: ISO C 9x
3551 llroundl NEW: ISO C 9x
3552 log2 NEW: ISO C 9x
3553 log2f NEW: ISO C 9x
3554 log2l NEW: ISO C 9x
3555 lrint NEW: ISO C 9x
3556 lrintf NEW: ISO C 9x
3557 lrintl NEW: ISO C 9x
3558 lround NEW: ISO C 9x
3559 lroundf NEW: ISO C 9x
3560 lroundl NEW: ISO C 9x
3561 lseek64 NEW: LFS
3562 makecontext NEW: Unix98
3563 mempcpy NEW: GNU ext.
3564 mmap64 NEW: LFS
3565 moncontrol REMOVED
3566 modify_ldt NEW: kernel
3567 nan NEW: ISO C 9x
3568 nanf NEW: ISO C 9x
3569 nanl NEW: ISO C 9x
3570 nearbyint NEW: ISO C 9x
3571 nearbyintf NEW: ISO C 9x
3572 nearbyintl NEW: ISO C 9x
3573 netname2host NEW: Secure RPC
3574 netname2user NEW: Secure RPC
3575 nexttoward NEW: ISO C 9x
3576 nexttowardf NEW: ISO C 9x
3577 nexttowardl NEW: ISO C 9x
3578 nftw NEW: Unix98
3579 nftw64 NEW: LFS
3580 open64 NEW: LFS
3581 passwd2des NEW: Secure RPC
3582 pow10 NEW: GNU ext.
3583 pow10f NEW: GNU ext.
3584 pow10l NEW: GNU ext.
3585 pread NEW: Unix98
3586 pread64 NEW: LFS
3587 printf_size NEW: GNU ext.
3588 printf_size_info NEW: GNU ext.
3589 profil_counter REMOVED
3590 pthread_mutexattr_getkind_np REPLACED
3591 pthread_mutexattr_setkind_np REPLACED
3592 ptsname NEW: Unix98 PTY
3593 ptsname_r NEW: Unix98 PTY
3594 putc_locked REMOVED
3595 putchar_locked REMOVED
3596 putgrent NEW: GNU ext.
3597 putmsg NEW: STREAMS
3598 putpmsg NEW: STREAMS
3599 pututxline NEW: Unix98
3600 pwrite NEW: Unix98
3601 pwrite64 NEW: LFS
3602 readdir64 NEW: LFS
3603 readdir64_r NEW: LFS
3604 remquo NEW: ISO C 9x
3605 remquof NEW: ISO C 9x
3606 remquol NEW: ISO C 9x
3607 round NEW: ISO C 9x
3608 roundf NEW: ISO C 9x
3609 roundl NEW: ISO C 9x
3610 rtime NEW: GNU ext.
3611 scalbln NEW: ISO C 9x
3612 scalblnf NEW: ISO C 9x
3613 scalblnl NEW: ISO C 9x
3614 scandir64 NEW: LFS
3615 sendfile NEW: kernel
3616 setcontext NEW: Unix98
3617 setrlimit64 NEW: LFS
3618 setutxent NEW: Unix98
3619 sighold NEW: Unix98
3620 sigignore NEW: Unix98
3621 sigqueue NEW: POSIX.1b
3622 sigrelse NEW: Unix98
3623 sigset NEW: POSIX.1b
3624 sigtimedwait NEW: POSIX.1b
3625 sigwaitinfo NEW: POSIX.1b
3626 sincos NEW: GNU ext.
3627 sincosf NEW: GNU ext.
3628 sincosl NEW: GNU ext.
3629 statfs64 NEW: LFS
3630 statvfs NEW: Unix98
3631 statvfs64 NEW: LFS
3632 strcasestr NEW: GNU ext.
3633 strtoimax NEW: ISO C 9x
3634 strtoumax NEW: ISO C 9x
3635 strverscmp NEW: GNU ext.
3636 svcauthdes_stats NEW: Secure RPC
3637 svcunix_create NEW: sunrpc etx
3638 svcunixfd_create NEW: sunrpc ext
3639 swapcontext NEW: Unix98
3640 tcgetsid NEW: Unix98 PTY
3641 tdestroy NEW: GNU ext.
3642 tgamma NEW: ISO C 9x
3643 tgammaf NEW: ISO C 9x
3644 tgammal NEW: ISO C 9x
3645 tmpfile64 NEW: LFS
3646 trunc NEW: ISO C 9x
3647 truncate64 NEW: LFS
3648 truncf NEW: ISO C 9x
3649 truncl NEW: ISO C 9x
3650 umount2 NEW: kernel
3651 unlockpt NEW: Unix98 PTY
3652 updwtmpx NEW: Unix98
3653 user2netname NEW: Secure RPC
3654 utmpxname NEW: Unix98
3655 versionsort NEW: GNU ext.
3656 versionsort64 NEW: GNU ext.
3657 waitid NEW: Unix98
3658 wcscasecmp NEW: GNU ext.
3659 wcsncasecmp NEW: GNU ext.
3660 wcsnlen NEW: GNU ext.
3661 wcstoimax NEW: ISO C 9x
3662 wcstoll NEW: ISO C 9x
3663 wcstoull NEW: ISO C 9x
3664 wcstoumax NEW: ISO C 9x
3665 wcswcs NEW: Unix98
3666 wordexp NEW: POSIX.2
3667 wordfree NEW: POSIX.2
3668 write_profiling REMOVED
3669 xdecrypt NEW: Secure RPC
3670 xdr_authdes_cred NEW: Secure RPC
3671 xdr_authdes_verf NEW: Secure RPC
3672 xdr_cryptkeyarg NEW: Secure RPC
3673 xdr_cryptkeyarg2 NEW: Secure RPC
3674 xdr_cryptkeyres NEW: Secure RPC
3675 xdr_getcredres NEW: Secure RPC
3676 xdr_key_netstarg NEW: Secure RPC
3677 xdr_key_netstres NEW: Secure RPC
3678 xdr_keybuf NEW: Secure RPC
3679 xdr_keystatus NEW: Secure RPC
3680 xdr_netnamestr NEW: Secure RPC
3681 xdr_sizeof NEW: Secure RPC
3682 xdr_unixcred NEW: sunrpc ext
3683 xencrypt NEW: Secure RPC
3684 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3685 \f
3686 Version 2.0.6
3687
3688 * more bug fixes
3689
3690 \f
3691 Version 2.0.5
3692
3693 * more bug fixes
3694
3695 * inet_ntoa is thread-safe
3696
3697 * updwtmp is moved from libutil to libc
3698
3699 * rewrite of cbrt function
3700
3701 * update of timezone data
3702 \f
3703 Version 2.0.4
3704
3705 * more bug fixes
3706 \f
3707 Version 2.0.3
3708
3709 * more bug fixes
3710 \f
3711 Version 2.0.2
3712
3713 * more bug fixes
3714
3715 * add atoll function
3716
3717 * fix complex problems in Berkeley DB code
3718
3719 * fix math functions
3720 \f
3721 Version 2.0.1
3722
3723 * fixed lots of header problems (especially Linux/GNU specific)
3724
3725 * dynamic loader preserves all registers
3726
3727 * Roland McGrath provided support for handling of auxiliary objects in
3728 the ELF dynamic loader.
3729
3730 * support for parallel builds is improved
3731 \f
3732 Version 2.0
3733
3734 * GNU extensions are no longer declared by default. To enable them you
3735 must define the macro `_GNU_SOURCE' in your program or compile with
3736 `-D_GNU_SOURCE'.
3737
3738 * The library has changed from using GNU ld symbol aliases to using weak
3739 symbols where available. The ELF object file format supports weak
3740 symbols; GNU ld also supports weak symbols in the a.out format. (There
3741 is also now support for other GNU ld extensions in ELF. Use the
3742 `--with-elf' option to configure to indicate you have ELF, and
3743 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3744 of many files which contained only symbol aliases, reducing the size of
3745 the source and the compiled library; many other files were renamed to
3746 less cryptic names previously occupied by the symbol alias files.
3747 There is a new header file <elf.h> for programs which operate on
3748 files in the ELF format.
3749
3750 * Converted to Autoconf version 2, so `configure' has more options.
3751 Run `configure --help' to see the details.
3752
3753 * The library can now be configured to build profiling, highly-optimized
3754 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3755 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3756 `configure' enable building these extra libraries. The shared library is
3757 built by default when using both ELF and GNU ld. When shared libraries
3758 are enabled, the new library `-ldl' is available for arbitrary run-time
3759 loading of shared objects; its interface is defined in <dlfcn.h>. The
3760 new header file <link.h> gives access to the internals of the run-time
3761 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3762 application of same name on other systems and it provides information
3763 about dynamically linked binaries.
3764
3765 * The C library now provides the run-time support code for profiling
3766 executables compiled with `-pg'. Programs can control the profiling code
3767 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3768 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3769 the support for this file format was contributed by David Mosberger-Tang.
3770
3771 * The math code has been replaced with a math library based on fdlibm from
3772 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3773 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3774 functions. The math functions now reside in a separate library, so
3775 programs using them will need to use `-lm' their linking commands.
3776
3777 * John C. Bowman contributed optimized ix87 assembler inline functions.
3778
3779 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3780 mechanism similar to that found in Solaris 2. This is now used for the
3781 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3782 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3783 file controls what services are used for each individual database. This
3784 works by loading shared libraries with names specified in `nsswitch.conf',
3785 so service modules can be changed or added at any time without even
3786 relinking any program. Currently there are the file, db, and NIS based
3787 NSS services available.
3788
3789 * The new functions `strtoq' and `strtouq' parse integer values from
3790 strings, like `strtol' and `strtoul', but they return `long long int' and
3791 `unsigned long long int' values, respectively (64-bit quantities).
3792
3793 * The new functions `strtof' and `strtold' parse floating-point values from
3794 strings, like `strtod', but they return `float' and `long double' values,
3795 respectively (on some machines `double' and `long double' are the same).
3796
3797 * Ulrich Drepper has contributed new implementations of the floating-point
3798 printing and reading code used in the `printf' family of functions and
3799 `strtod', `strtof', and `strtold'. These new functions are perfectly
3800 accurate, and much faster than the old ones.
3801
3802 * The implementation of the POSIX locale model was completely rewritten by
3803 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3804 to compile the POSIX locale definition.
3805
3806 * The former dummy implementations of the strcoll and strxfrm function are
3807 now replaced by fully functional code contributed by Ulrich Drepper. The
3808 collation information comes from the POSIX locale definitions.
3809
3810 * The new header <langinfo.h> defines an interface for accessing
3811 various locale-dependent data (using the locale chosen with `setlocale').
3812
3813 * Ulrich Drepper has contributed a new suite of functions for operation on
3814 wide-character and multibyte-character strings, in <wchar.h>;
3815 and classification and case conversion of wide characters, in <wctype.h>.
3816 These new functions are conforming to the ISO C, Amendement 1 specification.
3817
3818 * There is now a second implementation of the standard I/O library available.
3819 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3820 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3821
3822 * You can now use positional parameter specifications in format strings
3823 for the `printf' and `scanf' families of functions. For example,
3824 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3825 ``Number 6, Mr Jones''. This is mainly useful when providing different
3826 format strings for different languages, whose grammars may dictate
3827 different orderings of the values being printed. To support this
3828 feature, the interface for `register_printf_handler' has changed; see
3829 the header file <printf.h> for details.
3830
3831 * The `printf' and `scanf' families of functions now understand a new
3832 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3833 to group numbers as indicated by the locale; for `scanf' and friends, this
3834 says to accept as valid only a number with all the proper grouping
3835 separators in the right places. In the default "C" locale, numbers are
3836 not grouped; but locales for specific countries will define the usual
3837 conventions (i.e. separate thousands with `,' in the US locale).
3838
3839 * The pgrp functions have been regularized, slightly incompatibly but much
3840 less confusingly. The core functions are now `getpgid' and `setpgid',
3841 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3842 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3843 provided for compatibility. There is no longer an incompatible `getpgrp'
3844 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3845
3846 * The new header file <fts.h> and suite of functions simplify programs that
3847 operate on directory trees. This code comes from 4.4 BSD.
3848
3849 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3850 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3851 NSS scheme used in glibc.
3852
3853 * There is a new malloc debugging hook `__memalign_hook'.
3854
3855 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3856 `unsigned int' in <sys/types.h>. These are for compatibility only and
3857 their use is discouraged.
3858
3859 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3860 done differently, so that it works even without GNU ld.
3861
3862 * New function `euidaccess' checks allowed access to a file like `access',
3863 but using the effective IDs instead of the real IDs.
3864
3865 * The time zone data files have been updated for the latest and greatest
3866 local time conventions of the countries of the world.
3867
3868 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3869 see <dirent.h>.
3870
3871 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3872 for formatting floating-point numbers. They are provided only for
3873 compatibility; new programs should use `sprintf' instead. There are
3874 also equivalent function for the `long double' floating-point type and
3875 all functions also exist in a reentrant form.
3876
3877 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3878 functions for maintaining the login-record files (primarily of use to
3879 system programs such as `login'), and convenient functions for
3880 allocating and initializing a pseudo-terminal (pty) device.
3881
3882 * Ulrich Drepper has contributed new support for System V style
3883 shared memory and IPC on systems that support it.
3884
3885 * Ulrich Drepper has contributed several miscellaneous new functions found
3886 in System V: The `hsearch' family of functions provide an effective
3887 implementation of hash tables; `a64l' and `l64a' provide a very simple
3888 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3889 number generator.
3890
3891 * Ulrich Drepper has contributed new reentrant counterparts for the
3892 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3893
3894 * Ulrich Drepper has contributed new, highly-optimized versions of several
3895 string functions for the i486/Pentium family of processors.
3896
3897 * Ulrich Drepper has updated the Linux-specific code, based largely
3898 on work done in Hongjiu Lu's version of GNU libc for Linux.
3899 The GNU library now supports Linux versions 2.0.10 and later,
3900 using the ELF object file format (i[3456]86-*-linux).
3901
3902 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3903
3904 * David Mosberger-Tang and Richard Henderson have ported the C library
3905 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3906 dynamic linking support for ELF/Alpha.
3907
3908 * Richard Henderson contributed several Alpha optimized assembler function
3909 for arithmetic and string handling.
3910
3911 * Ulrich Drepper has contributed a new set of message catalog functions to
3912 support multiple languages using the <libintl.h> interface, for use with
3913 his new package GNU gettext. Translation volunteers have contributed
3914 catalogs of the library's messages in Spanish, German, and Korean.
3915
3916 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3917 program and the `catgets' function for reading the catalog files it
3918 creates. (The <libintl.h> interface is preferred; we include the
3919 <nl_types.h> interface using `catgets' only for source compatibility with
3920 programs already written to use it.)
3921
3922 * New header file <values.h> gives SVID-compatible names for <limits.h>
3923 constants.
3924
3925 * Various new macros, declarations, and small header files for compatibility
3926 with 4.4 BSD.
3927
3928 * New function `group_member' is a convenient way to check if a process has
3929 a given effective group ID.
3930
3931 * When using GCC 2.7 and later, the socket functions are now declared in a
3932 special way so that passing an argument of type `struct sockaddr_in *',
3933 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3934 `struct sockaddr *' type, does not generate a type-clash warning.
3935
3936 * New function `error' declared in header file <error.h> is a convenient
3937 function for printing error messages and optionally exiting; this is the
3938 canonical function used in GNU programs. The new functions `err', `warn',
3939 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3940 doing the same thing.
3941
3942 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3943 POSIX.2 `glob' function to do ~ and {...} expansion.
3944
3945 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3946 `clearenv' which is used in POSIX.9 is also available.
3947
3948 * New function `getsid' returns session ID number on systems that support it.
3949
3950 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3951 files <db.h> and <mpool.h> provide a rich set of functions for several
3952 types of simple databases stored in memory and in files, and <ndbm.h> is
3953 an old `ndbm'-compatible interface using the `db' functions. Link with
3954 `-ldb' to get these functions.
3955
3956 * New macro `strdupa' copies a string like `strdup', but uses local stack
3957 space from `alloca' instead of dynamic heap space from `malloc'.
3958
3959 * New function `strnlen' is like `strlen' but searches only a given maximum
3960 number of characters for the null terminator. `stpncpy', `strndup' and
3961 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3962 function.
3963
3964 * New function `statfs' in header <sys/statfs.h>.
3965
3966 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3967 provide convenient functions for operating on blocks of null-terminated
3968 strings.
3969
3970 * A new suite of functions in <utmp.h> handle all the details of reading
3971 and writing the utmp file.
3972
3973 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3974 Thorsten Kukuk.
3975
3976 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3977 completely POSIX compliant and also implemented the extended functionality
3978 to handle alternate digit representation and alternate era date formats.
3979
3980 * Ulrich Drepper provided an implementation of the `strptime' function
3981 defined in XPG4.2 which transforms a string into a `struct tm' value.
3982
3983 * Paul Eggert provided the tzselect shell script as part of the timezone
3984 code. The shell script makes it easy to select the correct timezone
3985 specification.
3986
3987 * The implementation of the malloc family of functions is completely replaced
3988 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3989 The implementation uses the mmap function (if available) and it is
3990 optimized for the use in multi threaded programs.
3991
3992 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3993 functions. This new functionality is usable by specifying a special
3994 salt string and it is compatible with implementation on *BSD systems.
3995
3996 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3997 `getsubopt' to handle second level command line options, `bsd_signal'
3998 to access BSD style `signal' functionality, the obsolete `regexp' style
3999 expression matcher.
4000
4001 * the `lchown' function is available on system which support this
4002 functionality.
4003
4004 * The implementation of the shadow password handling function was contributed
4005 by Ulrich Drepper.
4006
4007 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4008
4009 * POSIX.1g support was added. The <sys/select.h> header is available,
4010 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4011 implementation of `getaddrinfo'.
4012 \f
4013 Version 1.09
4014
4015 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4016
4017 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4018 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4019
4020 * The new function `daemon' from 4.4 BSD is useful for server programs that
4021 want to put themselves in the background.
4022
4023 * Joel Sherrill has contributed support for several standalone boards that
4024 run without an operating system.
4025
4026 * `printf', `scanf' and friends now accept a `q' type modifier for long
4027 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4028
4029 * All of the code taken from BSD (notably most of the math and networking
4030 routines) has been updated from the BSD 4.4-Lite release.
4031
4032 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4033
4034 * The new functions `getdomainname' and `setdomainname' fetch or change the
4035 YP/NIS domain name. These are system calls which exist on systems which
4036 have YP (aka NIS).
4037
4038 * The time zone data files have been updated for the latest international
4039 conventions.
4040
4041 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4042 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4043 \f
4044 Version 1.08
4045
4046 * The C library now includes support for Sun RPC, from Sun's free
4047 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4048 are included. (There is still no support for YP.)
4049
4050 * Tom Quinn has contributed a port of the C library to SGI machines running
4051 Irix 4 (mips-sgi-irix4).
4052
4053 * The new `lockf' function is a simplified interface to the locking
4054 facilities of `fcntl', included for compatibility.
4055
4056 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4057
4058 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4059
4060 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4061 compatibility.
4062
4063 * The error message printed by `assert' for a failed assertion now includes
4064 the name of the program (if using GNU ld) and the name of the calling
4065 function (with versions of GCC that support this).
4066
4067 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4068
4069 * The library now includes the <sys/mman.h> header file and memory
4070 management functions `mmap', `munmap', `mprotect', `msync', and
4071 `madvise', on systems that support those facilities.
4072
4073 * The interface for `mcheck' has changed slightly: the function called to
4074 abort the program when an allocation inconsistency is detected now takes
4075 an argument that indicates the type of failure. The new function
4076 `mprobe' lets you request a consistency check for a particular block at
4077 any time (checks are normally done only when you call `free' or `realloc'
4078 on a block).
4079
4080 * It is now possible to easily cross-compile the C library, building on one
4081 system a library to run on another machine and/or operating system. All
4082 you need to do is set the variable `HOST_CC' in `configparms' to the
4083 native compiler for programs to run on the machine you are building on (a
4084 few generator programs are used on Unix systems); set `CC' to the
4085 cross-compiler.
4086
4087 * The new function `fexecve' (only implemented on the GNU system) executes
4088 a program file given a file descriptor already open on the file.
4089 \f
4090 Version 1.07
4091
4092 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4093 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4094
4095 * You can set the variable `libprefix' in `configparms' to specify a prefix
4096 to be prepended to installed library files; this makes it easy to install
4097 the GNU C library to be linked as `-lgnuc' or whatever.
4098
4099 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4100 copies a limited number of characters from a string, and returns the
4101 address of the last character written.
4102
4103 * You no longer need to check for whether the installed `stddef.h' is
4104 compatible with the GNU C library. configure now checks for you.
4105
4106 * You can now define a per-stream `fileno' function to convert the
4107 stream's cookie into an integral file descriptor.
4108
4109 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4110 allocates zero bytes of storage, and returns a unique pointer which
4111 you can pass to `realloc' or `free'. The behavior is undefined if
4112 you dereference this pointer.
4113
4114 * The C library now runs on Sony NEWS m68k machines running either
4115 NewsOS 3 or NewsOS 4.
4116
4117 * The new `syscall' function is a system-dependent primitive function
4118 for invoking system calls. It has the canonical behavior on Unix
4119 systems, including unreliable return values for some calls (such as
4120 `pipe', `fork' and `getppid').
4121
4122 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4123 to `EAGAIN', which is the preferred name. On systems whose kernels
4124 use two distinct codes, the C library now translates EWOULDBLOCK to
4125 EAGAIN in every system call function.
4126 \f
4127 Version 1.06
4128
4129 * The GNU C Library Reference Manual is now distributed with the library.
4130 `make dvi' will produce a DVI file of the printed manual.
4131 `make info' will produce Info files that you can read on line using C-h i
4132 in Emacs or the `info' program.
4133 Please send comments on the manual to bug-glibc-manual@gnu.org.
4134
4135 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4136
4137 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4138
4139 * Jason Merrill has contributed a port to the Sequent Symmetry running
4140 Dynix version 3 (i386-sequent-dynix).
4141
4142 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4143 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4144
4145 * New function `memory_warnings' lets you arrange to get warnings when
4146 malloc is running out of memory to allocate, like Emacs gives you.
4147
4148 * The C library now contains the relocating allocator used in Emacs 19 for
4149 its editing buffers. This allocator (ralloc) minimizes allocation
4150 overhead and fragmentation by moving allocated regions around whenever it
4151 needs to. You always refer to a ralloc'd region with a "handle" (a
4152 pointer to a pointer--an object of type `void **').
4153
4154 * There is a new `printf' format: `%m' gives you the string corresponding
4155 to the error code in `errno'.
4156
4157 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4158 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4159 pass, the `a' modifier says to fill in a `char **' you pass with a
4160 malloc'd string.
4161
4162 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4163 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4164 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4165
4166 * `mkstemp' is a traditional Unix function to atomically create and open a
4167 uniquely-named temporary file.
4168 \f
4169 Version 1.05
4170
4171 * The standard location for the file that says what the local timezone is
4172 has changed again. It is now `/usr/local/etc/localtime' (or more
4173 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4174
4175 * The distribution no longer contains any files with names longer than 14
4176 characters.
4177
4178 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4179 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4180
4181 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4182 \f
4183 Version 1.04
4184
4185 * The configuration process has changed quite a bit. The `configure'
4186 script is now used just like the configuration scripts for other GNU
4187 packages. The `sysdeps' directory hierarchy is much rearranged.
4188 The file `INSTALL' explains the new scheme in detail.
4189
4190 * The header files no longer need to be processed into ANSI C and
4191 traditional C versions. There is just one set of files to install, and
4192 it will work with ANSI or old C compilers (including `gcc -traditional').
4193
4194 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4195 MIPS DECStation running Ultrix 4.
4196
4197 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4198 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4199 made itself into a shared library.
4200
4201 * Yet further improved support for the i386, running 4.3 BSD-like systems
4202 (such as Mach 3 with the Unix single-server), or System V.
4203
4204 * New function `strncasecmp' to do case-insensitive string comparison
4205 with limited length.
4206
4207 * New function `strsep' is a reentrant alternative to `strtok'.
4208
4209 * New functions `scandir' and `alphasort' for searching directories.
4210
4211 * New function `setenv' is a better interface to `putenv'.
4212
4213 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4214 function for traversing a directory tree.
4215
4216 * The GNU obstack package is now also part of the C library.
4217 The new function `open_obstack_stream' creates a stdio stream that
4218 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4219 formatted output directly to an obstack.
4220
4221 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4222 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4223
4224 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4225
4226 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4227 things to your strings.
4228
4229 * There are some new test programs: test-fseek, testmb, and testrand.
4230
4231 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4232 These ports are not finished, but are a good starting place for really
4233 supporting those systems.
4234
4235 * `/etc/localtime' is now the standard location for the file that says what
4236 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4237 This follows the general principle that `/etc' is the place for all local
4238 configuration files.
4239
4240 * The C library header files now use `extern "C"' when used by the C++
4241 compiler, so the C library should now work with C++ code.
4242
4243 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4244 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4245 in <strings.h>.)
4246
4247 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4248 function which uses the `merge sort' algorithm, and is said to be
4249 significantly faster than the old GNU `qsort' function. Merge sort is
4250 now the standard `qsort' function. The new algorithm can require a lot
4251 of temporary storage; so, the old sorting function is called when the
4252 required storage is not available.
4253
4254 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4255 provides the Unix `crypt' function, plus some other entry points.
4256
4257 * The code and header files taken from 4.4 BSD have been updated with the
4258 latest files released from Berkeley.
4259 \f
4260 ----------------------------------------------------------------------
4261 Copyright information:
4262
4263 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4264
4265 Permission is granted to anyone to make or distribute verbatim copies
4266 of this document as received, in any medium, provided that the
4267 copyright notice and this permission notice are preserved,
4268 thus giving the recipient permission to redistribute in turn.
4269
4270 Permission is granted to distribute modified versions
4271 of this document, or of portions of it,
4272 under the above conditions, provided also that they
4273 carry prominent notices stating who last changed them.
4274 \f
4275 Local variables:
4276 version-control: never
4277 fill-column: 76
4278 End: