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