]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
Fix mcontext_t sigcontext namespace (bug 21457).
[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 <http://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 * In order to support faster and safer process termination the malloc API
16 family of functions will no longer print a failure address and stack
17 backtrace after detecting heap corruption. The goal is to minimize the
18 amount of work done after corruption is detected and to avoid potential
19 security issues in continued process execution. Reducing shutdown time
20 leads to lower overall process restart latency, so there is benefit both
21 from a security and performance perspective.
22
23 Deprecated and removed features, and other changes affecting compatibility:
24
25 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
26 defined by <sys/ptrace.h>.
27
28 * libm no longer supports SVID error handling (calling a user-provided
29 matherr function on error) or the _LIB_VERSION variable to control error
30 handling. (SVID error handling and the _LIB_VERSION variable still work
31 for binaries linked against older versions of the GNU C Library.) The
32 libieee.a library is no longer provided. math.h no longer defines struct
33 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
34 TLOSS, PLOSS and HUGE.
35
36 * The mcontext_t type is no longer the same as struct sigcontext. On
37 platforms where it was previously the same, this changes the C++ name
38 mangling for interfaces involving this type.
39
40 Changes to build and runtime requirements:
41
42 [Add changes to build and runtime requirements here]
43
44 Security related changes:
45
46 CVE-2009-5064: The ldd script would sometimes run the program under
47 examination directly, without preventing code execution through the
48 dynamic linker. (The glibc project disputes that this is a security
49 vulnerability; only trusted binaries must be examined using the ldd
50 script.)
51
52 The following bugs are resolved with this release:
53
54 [The release manager will add the list generated by
55 scripts/list-fixed-bugs.py just before the release.]
56
57 \f
58 Version 2.26
59
60 Major new features:
61
62 * A per-thread cache has been added to malloc. Access to the cache requires
63 no locks and therefore significantly accelerates the fast path to allocate
64 and free small amounts of memory. Refilling an empty cache requires locking
65 the underlying arena. Performance measurements show significant gains in a
66 wide variety of user workloads. Workloads were captured using a special
67 instrumented malloc and analyzed with a malloc simulator. Contributed by
68 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
69
70 * Unicode 10.0.0 Support: Character encoding, character type info, and
71 transliteration tables are all updated to Unicode 10.0.0, using
72 generator scripts contributed by Mike FABIAN (Red Hat).
73 These updates cause user visible changes, especially the changes in
74 wcwidth for many emoji characters cause problems when emoji sequences
75 are rendered with pango, see for example:
76 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
77
78 * Collation of Hungarian has been overhauled and is now consistent with "The
79 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
80 Egmont Koblinger.
81
82 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
83
84 - The GNU C Library will now detect when /etc/resolv.conf has been
85 modified and reload the changed configuration. The new resolver option
86 “no-reload” (RES_NORELOAD) disables this behavior.
87
88 - The GNU C Library now supports an arbitrary number of search domains
89 (configured using the “search” directive in /etc/resolv.conf);
90 previously, there was a hard limit of six domains. For backward
91 compatibility, applications that directly modify the ‘_res’ global
92 object are still limited to six search domains.
93
94 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
95 Library will now randomly pick a name server from the configuration as a
96 starting point. (Previously, the second name server was always used.)
97
98 * The tunables feature is now enabled by default. This allows users to tweak
99 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
100
101 * New function reallocarray, which resizes an allocated block (like realloc)
102 to the product of two sizes, with a guaranteed clean failure upon integer
103 overflow in the multiplication. Originally from OpenBSD, contributed by
104 Dennis Wölfing and Rüdiger Sonderfeld.
105
106 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
107 These are extended versions of preadv and pwritev, respectively, taking an
108 additional flags argument. The set of supported flags depends on the
109 running kernel; full support currently requires kernel 4.7 or later.
110
111 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
112 create a new session ID for the spawned process. This feature is
113 scheduled to be added to the next major revision of POSIX; for the time
114 being, it is available under _GNU_SOURCE.
115
116 * errno.h is now safe to use from C-preprocessed assembly language on all
117 supported operating systems. In this context, it will only define the
118 Exxxx constants, as preprocessor macros expanding to integer literals.
119
120 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
121 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
122 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
123 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
124
125 To compile programs that use this feature, the compiler must support
126 128-bit floating point with the type name _Float128 (as defined by TS
127 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
128 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
129 must be defined to make the new interfaces visible.
130
131 The new functions and macros correspond to those present for other
132 floating-point types (except for a few obsolescent interfaces not
133 supported for the new type), with F128 or f128 suffixes; for example,
134 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
135 printf or scanf formats for the new type; the strfromf128 and strtof128
136 interfaces should be used instead.
137
138 Deprecated and removed features, and other changes affecting compatibility:
139
140 * The synchronization that pthread_spin_unlock performs has been changed to
141 now be equivalent to a C11 atomic store with release memory order to the
142 spin lock's memory location. Previously, several (but not all)
143 architectures used stronger synchronization (e.g., containing what is
144 often called a full barrier). This change can improve performance, but
145 may affect odd fringe uses of spin locks that depend on the previous
146 behavior (e.g., using spin locks as atomic variables to try to implement
147 Dekker's mutual exclusion algorithm).
148
149 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
150 removed.
151
152 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
153 will only be built and installed when the GNU C Library is configured with
154 --enable-obsolete-rpc. This allows alternative RPC implementations, such
155 as TIRPC or rpcsvc-proto, to be used.
156
157 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
158 libnss_compat, are deprecated, and will not be built or installed by
159 default.
160
161 The NIS(+) support library, libnsl, is also deprecated. By default, a
162 compatibility shared library will be built and installed, but not headers
163 or development libraries. Only a few NIS-related programs require this
164 library. (In particular, the GNU C Library has never required programs
165 that use 'gethostbyname' to be linked with libnsl.)
166
167 Replacement implementations based on TIRPC, which additionally support
168 IPv6, are available from <https://github.com/thkukuk/>. The configure
169 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
170 name service modules, to be built and installed.
171
172 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
173 support is enabled, the configured recursive resolver must support EDNS.
174 (Responding to EDNS-enabled queries with responses which are not
175 EDNS-enabled is fine, but FORMERR responses are not.)
176
177 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
178 servers have not supported this opcode for a long time.
179
180 * The _res_opcodes variable has been removed from libresolv. It had been
181 exported by accident.
182
183 * <string.h> no longer includes inline versions of any string functions,
184 as this kind of optimization is better done by the compiler. The macros
185 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
186
187 * The nonstandard header <xlocale.h> has been removed. Most programs should
188 use <locale.h> instead. If you have a specific need for the definition of
189 locale_t with no other declarations, please contact
190 libc-alpha@sourceware.org and explain.
191
192 * The obsolete header <sys/ultrasound.h> has been removed.
193
194 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
195
196 * The obsolete function cfree has been removed. Applications should use
197 free instead.
198
199 * The stack_t type no longer has the name struct sigaltstack. This changes
200 the C++ name mangling for interfaces involving this type.
201
202 * The ucontext_t type no longer has the name struct ucontext. This changes
203 the C++ name mangling for interfaces involving this type.
204
205 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
206 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
207 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
208 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
209 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
210 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
211 fpregset_t type no longer has the name struct fpu. This changes the C++
212 name mangling for interfaces involving those types.
213
214 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
215 synced with the kernel:
216
217 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
218 are not supported on this architecture and have been removed.
219
220 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
221 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
222 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
223
224 Programs that assume the GET/SETREGS ptrace requests are universally
225 available will now fail to build, instead of malfunctioning at runtime.
226
227 Changes to build and runtime requirements:
228
229 * Linux kernel 3.2 or later is required at runtime, on all architectures
230 supported by that kernel. (This is a change from version 2.25 only for
231 x86-32 and x86-64.)
232
233 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
234
235 * On most architectures, GCC 4.9 or later is required to build the GNU C
236 Library. On powerpc64le, GCC 6.2 or later is required.
237
238 Older GCC versions and non-GNU compilers are still supported when
239 compiling programs that use the GNU C Library. (We do not know exactly
240 how old, and some GNU extensions to C may be _de facto_ required. If you
241 are interested in helping us make this statement less vague, please
242 contact libc-alpha@sourceware.org.)
243
244 Security related changes:
245
246 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
247 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
248
249 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
250 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
251
252 * Avoid printing a backtrace from the __stack_chk_fail function since it is
253 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
254 (CVE-2010-3192).
255
256 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
257 fixed (CVE-2017-12133).
258
259 The following bugs are resolved with this release:
260
261 [984] network: Respond to changed resolv.conf in gethostbyname
262 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
263 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
264 fields
265 [12189] libc: __stack_chk_fail should not attempt a backtrace
266 (CVE-2010-3192)
267 [14096] time: Race condition on timezone/tst-timezone.out
268 [14172] localedata: az_IR: new locale
269 [14995] build: glibc fails to build if gold is the default linker, even if
270 ld.bfd is available
271 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
272 [16637] network: inet_pton function is accepting IPv6 with bad format
273 [16640] string: string/strtok.c: undefined behaviour inconsistent between
274 x86 and other generic code
275 [16875] localedata: ko_KR: fix lang_name
276 [17225] localedata: ar_SY: localized month names for May and June are
277 incorrect
278 [17297] localedata: da_DK: wrong date_fmt string
279 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
280 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
281 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
282 [19066] localedata: ar_SA abbreviated day and month names are in English
283 [19569] network: resolv: Support an arbitrary number of search domains
284 [19570] network: Implement random DNS server selection in the stub
285 resolver
286 [19838] locale: localedef fails on PA-RISC
287 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
288 order of 0D36 and 0D37
289 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
290 chillu characters
291 [20098] libc: FAIL: debug/backtrace-tst on hppa
292 [20257] network: sunrpc: clntudp_call does not enforce timeout when
293 receiving data
294 [20275] localedata: locale day/abday/mon/abmon should not have trailing
295 whitespace
296 [20313] localedata: Update locale data to Unicode 9.0
297 [20424] manual: Document how to provide a malloc replacement
298 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
299 for Peru
300 [20686] locale: Add el_GR@euro to SUPPORTED.
301 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
302 failures consistently
303 [21015] dynamic-link: Document and fix --enable-bind-now
304 [21016] nptl: pthread_cond support is broken on hppa
305 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
306 frame-pointer on i386
307 [21049] libc: segfault in longjmp_chk() due to clobbered processor
308 register
309 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
310 [21088] libc: Build fails with --enable-static-nss
311 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
312 generic c code is used
313 [21109] libc: Tunables broken on big-endian
314 [21112] math: powf has large ulp errors with base close to 1 and exponent
315 around 4000
316 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
317 (CVE-2017-12133)
318 [21120] malloc: glibc malloc is incompatible with GCC 7
319 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
320 with -lieee
321 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
322 (-0.0) when linking with -lieee
323 [21171] math: log10, log2 and lgamma return incorrect results
324 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
325 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
326 [21207] localedata: ce_RU: update weekdays from CLDR
327 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
328 [21217] localedata: Update months from CLDR-31
329 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
330 [21243] libc: support_delete_temp_file should issue warning for failed
331 remove()
332 [21244] libc: support resolv_test_start() socket fd close should be
333 checked for errors.
334 [21253] libc: localedef randomly segfaults when using -fstack-check due to
335 new posix_spawn implementation
336 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
337 leads to lower CPU frequency
338 [21259] libc: [alpha] termios.h missing IXANY for POSIX
339 [21261] libc: [sparc64] bits/setjmp.h namespace
340 [21267] network: [mips] bits/socket.h IOC* namespace
341 [21268] libc: [alpha] termios.h NL2, NL3 namespace
342 [21270] libc: mmap64 silently truncates large offset values
343 [21275] libc: posix_spawn always crashes on ia64 now
344 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
345 [21280] math: [powerpc] logbl for POWER7 return incorrect results
346 [21289] libc: Incorrect declaration for 32-bit platforms with
347 _FILE_OFFSET_BITS=64 causes build error
348 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
349 not support gethostbyname4_r
350 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
351 switching
352 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
353 arenas
354 [21340] libc: Support POSIX_SPAWN_SETSID
355 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
356 [21359] network: ns_name_pack needs additional byte in destination buffer
357 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
358 against fragmentation attacks (CVE-2017-12132)
359 [21369] network: resolv: Remove EDNS fallback
360 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
361 and _POSIX_C_SOURCE
362 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
363 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
364 [21393] stdio: Missing dup3 error check in freopen, freopen64
365 [21396] libc: Use AVX2 memcpy/memset on Skylake server
366 [21399] localedata: Bad description for U00EC in
367 localedata/charmaps/CP1254
368 [21411] malloc: realloc documentation error
369 [21426] network: sys/socket.h uio.h namespace
370 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
371 [21445] libc: signal.h bsd_signal namespace
372 [21455] network: Network headers stdint.h namespace
373 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
374 retry value
375 [21475] network: resolv: Overlong search path is truncated mid-label
376 [21511] libc: sigstack namespace
377 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
378 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
379 bits/syscall.h: No such file or directory
380 [21517] libc: struct sigaltstack namespace
381 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
382 [21533] localedata: Update locale data to Unicode 10.0
383 [21537] libc:
384 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
385 junk at end of line, first unrecognized character is `@'
386 [21538] libc: SIG_HOLD missing for XPG4
387 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
388 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
389 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
390 [21543] libc: sigevent namespace
391 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
392 size for general purpose registers in mcontext_t structure
393 [21550] libc: sigwait namespace
394 [21552] libc: XPG4 bsd_signal namespace
395 [21554] libc: sigpause namespace
396 [21560] libc: sys/wait.h signal.h namespace
397 [21561] libc: waitid namespace
398 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
399 '/usr/include/stdlib.h': Permission denied
400 [21575] libc: sys/wait.h missing struct rusage definition
401 [21584] libc: sigaltstack etc namespace
402 [21597] libc: siginterrupt namespace
403 [21607] math: hppa: FAIL: math/test-tgmath
404 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
405 issue on x86-64
406 [21622] libc: [tile] missing SA_* for POSIX.1:2008
407 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
408 stack and heap (CVE-2017-1000366)
409 [21625] libc: wait3 namespace
410 [21654] nss: Incorrect pointer alignment in NSS group merge result
411 construction
412 [21657] network: Parse interface zone id for node-local multicast
413 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
414 between_2_3
415 [21666] libc: .symver is used on common symbol
416 [21668] network: resolv: res_init cross-thread broadcast introduces race
417 conditions
418 [21687] math: tgmath.h totalorder, totalordermag return type
419 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
420 Hindi Locale
421 [21696] libc: Incorrect assumption of of __cpu_mask in
422 posix/sched_cpucount.c
423 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
424 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
425 locale
426 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
427 leading to relocation crash
428 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
429 `resolv_conf_matches (resp, conf)' failed.
430 [21710] localedata: Added Samoan language locale for Samoa
431 [21711] localedata: Pashto yesstr/nostr locale are missing
432 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
433 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
434 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
435 [21723] localedata: yesstr/nostr missing for Chinese language locale
436 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
437 locale
438 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
439 locale
440 [21728] localedata: New Locale for Tongan language
441 [21729] localedata: incorrect LC_NAME fields for hi_IN
442 [21733] localedata: yesstr and nostr are missing for zh_HK
443 [21734] localedata: Missing yesstr and nostr are for kw_GB
444 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
445 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
446 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
447 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
448 strings
449 [21744] libc: Tests failing on --enable-tunables --enable-stack-
450 protector=all
451 [21749] localedata: Wrong abbreviated day name (“abday”) for
452 ar_JO/ar_LB/ar_SY
453 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
454 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
455 [21759] localedata: missing yesstr and nostr for Tigrinya
456 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
457 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
458 [21767] localedata: Missing Bislama locales
459 [21768] localedata: Missing yesstr and nostr for aa_ET
460 [21770] localedata: Missing Field in li_NL
461 [21778] nptl: Robust mutex may deadlock
462 [21779] libc: MicroBlaze segfaults when loading libpthread
463 [21783] localedata: Fix int_select international_call_prefixes
464 [21784] localedata: Inconsistency in country_isbn
465 [21788] localedata: Missing Country Postal Abbreviations
466 [21794] localedata: Added-country_isbn-for-Italy
467 [21795] localedata: Add/Fix country_isbn for France
468 [21796] localedata: Added country_isbn for Republic of Korea
469 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
470 [21799] localedata: Added int_select international_call_prefixes
471 [21801] localedata: Added int_select international_call_prefixes
472 [21804] nptl: Double semicolon in thread-shared-types.h
473 [21807] localedata: LC_ADDRESS fix for pap_CW
474 [21808] localedata: Fix LC_ADDRESS for pap_AW
475 [21821] localedata: Added country_name in mai_IN
476 [21822] localedata: Fix LC_TIME for mai_IN
477 [21823] localedata: missing yesstr, nostr for sa_IN
478 [21825] localedata: Fix name_mrs for mag_IN
479 [21828] localedata: 2.26 changelog should mention user visible changes
480 with unicode 9.0
481 [21835] localedata: Added Maithili language locale for Nepal
482 [21838] localedata: Removed redundant data for the_NP
483 [21839] localedata: Fix LC_MONETARY for ta_LK
484 [21844] localedata: Fix Latin characters and Months Sequence.
485 [21848] localedata: Fix mai_NP Title Name
486
487 \f
488 Version 2.25
489
490 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
491 24731-2:2010, is supported to enable declarations of functions from that
492 TR. Note that not all functions from that TR are supported by the GNU C
493 Library.
494
495 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
496 18661-1:2014, is supported to enable declarations of functions and macros
497 from that TS. Note that not all features from that TS are supported by
498 the GNU C Library.
499
500 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
501 18661-4:2015, is supported to enable declarations of functions and macros
502 from that TS. Note that most features from that TS are not supported by
503 the GNU C Library.
504
505 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
506 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
507 Since the GNU C Library defaults to a much newer revision of POSIX, this
508 will only affect programs that specifically request an old conformance
509 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
510 see a change in the visible declarations, but a program compiled with
511 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
512 will not.
513
514 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
515 defined by all multithreaded code, but glibc has not required this for
516 many years.
517
518 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
519 means that in a future release, the macros “major”, “minor”, and “makedev”
520 will only be available from <sys/sysmacros.h>.
521
522 These macros are not part of POSIX nor XSI, and their names frequently
523 collide with user code; see for instance glibc bug 19239 and Red Hat bug
524 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
525 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
526 problem.
527
528 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
529 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
530 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
531
532 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
533 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
534 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
535 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
536 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
537 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
538 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
539 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
540 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
541 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
542 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
543 WINT_WIDTH.
544
545 * New <math.h> features are added from TS 18661-1:2014:
546
547 - Signaling NaN macros: SNANF, SNAN, SNANL.
548
549 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
550 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
551 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
552
553 - llogb functions: the llogb, llogbf and llogbl functions, and the
554 FP_LLOGB0 and FP_LLOGBNAN macros.
555
556 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
557 fminmagf, fminmagl.
558
559 - Comparison macros: iseqsig.
560
561 - Classification macros: iscanonical, issubnormal, iszero.
562
563 - Total order functions: totalorder, totalorderf, totalorderl,
564 totalordermag, totalordermagf, totalordermagl.
565
566 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
567
568 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
569 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
570
571 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
572 are added to libc. They convert a floating-point number into string.
573
574 * Most of glibc can now be built with the stack smashing protector enabled.
575 It is recommended to build glibc with --enable-stack-protector=strong.
576 Implemented by Nick Alcock (Oracle).
577
578 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
579 intended to be used instead of memset() to erase sensitive data after use;
580 the compiler will not optimize out calls to explicit_bzero even if they
581 are "unnecessary" (in the sense that no _correct_ program can observe the
582 effects of the memory clear).
583
584 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
585 to float instead of double. This does not affect the ABI of any libraries
586 that are part of the GNU C Library, but may affect the ABI of other
587 libraries that use this type in their interfaces.
588
589 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
590 float_t and double_t types are now defined to long double instead of float
591 and double. These options are not the default, and this does not affect
592 the ABI of any libraries that are part of the GNU C Library, but it may
593 affect the ABI of other libraries that use this type in their interfaces,
594 if they are compiled or used with those options.
595
596 * The getentropy and getrandom functions, and the <sys/random.h> header file
597 have been added.
598
599 * The buffer size for byte-oriented stdio streams is now limited to 8192
600 bytes by default. Previously, on Linux, the default buffer size on most
601 file systems was 4096 bytes (and thus remains unchanged), except on
602 network file systems, where the buffer size was unpredictable and could be
603 as large as several megabytes.
604
605 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
606 for the Linux quota interface which predates kernel version 2.4.22 has
607 been removed.
608
609 * The malloc_get_state and malloc_set_state functions have been removed.
610 Already-existing binaries that dynamically link to these functions will
611 get a hidden implementation in which malloc_get_state is a stub. As far
612 as we know, these functions are used only by GNU Emacs and this change
613 will not adversely affect already-built Emacs executables. Any undumped
614 Emacs executables, which normally exist only during an Emacs build, should
615 be rebuilt by re-running “./configure; make” in the Emacs build tree.
616
617 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
618 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
619 “no-ip6-dotint” had already been the default, and support for the
620 “ip6-dotint” option was removed from the Internet in 2006.
621
622 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
623 flag from <resolv.h> have been removed. The option relied on a
624 backwards-incompatible DNS extension which was never deployed on the
625 Internet.
626
627 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
628 RES_BLAST defined in the <resolv.h> header file have been deprecated.
629 They were already unimplemented.
630
631 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
632 _res.flags are deprecated. The flag was standardized in RFC 2133, but
633 removed again from the IETF name lookup interface specification in RFC
634 2553. Applications should use getaddrinfo instead.
635
636 * DNSSEC-related declarations and definitions have been removed from the
637 <arpa/nameser.h> header file, and libresolv will no longer attempt to
638 decode the data part of DNSSEC record types. Previous versions of glibc
639 only implemented minimal support for the previous version of DNSSEC, which
640 is incompatible with the currently deployed version.
641
642 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
643 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
644 <arpa/nameser.h> header file because the distinction between RR types and
645 meta-RR types is not officially standardized, subject to revision, and
646 thus not suitable for encoding in a macro.
647
648 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
649 and rhook members of the res_state type in <resolv.h> have been removed.
650 The glibc stub resolver did not support these hooks, but the header file
651 did not reflect that.
652
653 * For multi-arch support it is recommended to use a GCC which has
654 been built with support for GNU indirect functions. This ensures
655 that correct debugging information is generated for functions
656 selected by IFUNC resolvers. This support can either be enabled by
657 configuring GCC with '--enable-gnu-indirect-function', or by
658 enabling it by default by setting 'default_gnu_indirect_function'
659 variable for a particular architecture in the GCC source file
660 'gcc/config.gcc'.
661
662 * GDB pretty printers have been added for mutex and condition variable
663 structures in POSIX Threads. When installed and loaded in gdb these pretty
664 printers show various pthread variables in human-readable form when read
665 using the 'print' or 'display' commands in gdb.
666
667 * Tunables feature added to allow tweaking of the runtime for an application
668 program. This feature can be enabled with the '--enable-tunables' configure
669 flag. The GNU C Library manual has details on usage and README.tunables has
670 instructions on adding new tunables to the library.
671
672 * A new version of condition variables functions have been implemented in
673 the NPTL implementation of POSIX Threads to provide stronger ordering
674 guarantees.
675
676 * A new version of pthread_rwlock functions have been implemented to use a more
677 scalable algorithm primarily through not using a critical section anymore to
678 make state changes.
679
680 Security related changes:
681
682 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
683 have been created with makecontext could fail to terminate due to a
684 missing .cantunwind annotation. This has been observed to lead to a hang
685 (denial of service) in some Go applications compiled with gccgo. Reported
686 by Andreas Schwab. (CVE-2016-6323)
687
688 * The DNS stub resolver functions would crash due to a NULL pointer
689 dereference when processing a query with a valid DNS question type which
690 was used internally in the implementation. The stub resolver now uses a
691 question type which is outside the range of valid question type values.
692 (CVE-2015-5180)
693
694 The following bugs are resolved with this release:
695
696 [4099] stdio: Overly agressive caching by stream i/o functions.
697 [7065] build: Support building glibc with -fstack-protector or -fstack-
698 protector-all
699 [9842] localedata: en_CA: incorrect date format
700 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
701 before it started waiting
702 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
703 [15765] nptl: sem_open is wrongly a cancellation point
704 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
705 [16458] libc: endian.h and netinet/in.h byte order macros return values of
706 wrong type
707 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
708 library linked with pthread
709 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
710 "FIXME: Ingo" issue)
711 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
712 [17252] libc: getrandom and getentropy syscall
713 [17730] malloc: thread-local storage is sometimes improperly free()'d
714 after being __libc_memalign()'d
715 [18241] stdio: failed fseek on memstream does not set errno and can fail
716 when it shouldnt
717 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
718 not
719 [18463] nptl: pthread_cond_broadcast issue when surrounded by
720 PTHREAD_PRIO_INHERIT mutex on ARM
721 [18784] network: res_query and related function crash for special record
722 type queries (CVE-2015-5180)
723 [19380] math: strtod does not raise "inexact"
724 [19387] string: Integer overflow in memchr
725 [19390] string: Integer overflow in strncat
726 [19398] build: linknamespace tests fail with massively parallel build
727 [19402] nptl: Deadlock with robust shared mutex and asynchronous
728 termination
729 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
730 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
731 [19514] libc: [PATCH] Fix spelling errors in spelling
732 "implement"/"implementation" in several places
733 [19582] network: Deprecate RES_USE_INET6
734 [19673] manual: clog10 docs appear to be erroneous
735 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
736 causes a segmentation fault
737 [19826] libc: invalid pointer returned from __tls_get_addr with static
738 linking
739 [20016] network: resolv: Remove hooks support from the API
740 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
741 uninitialized GOT
742 [20033] math: [x86_64] vectorized math function don't call the __finite
743 versions
744 [20116] nptl: use after free in pthread_create
745 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
746 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
747 is always true.
748 [20311] nptl: please install proc_service.h
749 [20366] libc: Compilation errors in installed headers in strict-compliance
750 modes
751 [20370] malloc: malloc: Arena free list management is still racy
752 (incorrect fix in bug 19243)
753 [20386] libc: assert (X = 0) does not result in GCC warning
754 [20432] malloc: malloc: Minimize interface required for interposition
755 [20435] libc: Missing unwind info in __startcontext causes infinite loop
756 in _Unwind_Backtrace (CVE-2016-6323)
757 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
758 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
759 libpthread.a
760 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
761 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
762 [20477] network: resolv: incorrect double-checked locking related to
763 _res_hconf
764 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
765 information.
766 [20495] math: x86_64 performance degradation due to AVX/SSE transition
767 penalty
768 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
769 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
770 cause transition penalty
771 [20517] math: sparcv9 missing fdiml compat symbol
772 [20524] manual: strverscmp is inconsistent
773 [20525] libc: <sys/quota.h> should be based on kernel headers
774 [20539] math: GCC 7 -static -lm fails to link at -O3
775 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
776 mcount.oS)
777 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
778 Checking
779 [20591] network: Remove obsolete DNSSEC support
780 [20592] network: DNS resource record type classification macros in
781 <arpa/nameser.h> are incorrect
782 [20593] network: Update DNS RR type definitions
783 [20611] network: getaddrinfo accepts invalid numeric scope IDs
784 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
785 multi-arch
786 [20629] network: libresolv: Remove support for bitstring labels
787 (RES_USEBSTRING)
788 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
789 assertion: "maxidx >= 2"
790 [20660] math: [arm] Use VSQRT
791 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
792 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
793 [20689] libc: Test for FMA should also check for AVX.
794 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
795 [20715] math: iszero macro breaks existing code
796 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
797 [20728] libc: powerpc: Missing TOC stub in clone
798 [20729] build: glibc-2.24 fails to build for i486 with -Os
799 [20750] build: Build fails with default PIE enabled gcc-6.x
800 [20768] math: [alpha] sqrt fegetenv namespace
801 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
802 glibc
803 [20787] math: float_t is defined as float incorrectly on x86_64 even if
804 __FLT_EVAL_METHOD__ is 2
805 [20790] network: rpcgen buffer overrun in get_prog_declaration
806 [20822] nptl: race condition in __lll_unlock_elision on powerpc
807 [20829] libc: crypt snprintf namespace
808 [20847] libc: tst-vfork3 failure
809 [20855] math: Default bits/mathdef.h has inappropriate float_t
810 [20859] math: [sh4] FP_ILOGB0 invalid
811 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
812 U+20AC), not same as GBK
813 [20915] dynamic-link: global-dynamic TLS broken on aarch64
814 [20916] math: pow handling of sNaN arguments
815 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
816 [20919] math: Bad pow (qNaN, 0) result with -lieee
817 [20940] math: hypot sNaN handling
818 [20947] math: fmax, fmin sNaN handling
819 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
820 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
821 RES_USE_INET6
822 [20971] string: powerpc64/power7 memchr overflows internal pointer check
823 [20973] nptl: robust mutexes: Lost wake-ups
824 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
825 [20978] nis: strlen on null pointer in nss_nisplus
826 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
827 wrong condition
828 [21014] string: i686 memchr overflows internal pointer check
829 [21019] libc: [mips] n32 lseek incorrect on overflow
830 [21022] libc: [microblaze] __backtrace get_frame_size namespace
831 [21026] libc: [MIPS] readahead syscall is broken on n64
832 [21028] math: Fallback fesetexceptflag should always succeed
833 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
834 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
835 clang
836 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
837 [21061] librt: [microblaze] librt lost clock_* exports
838 [21073] libc: tunables: insecure environment variables passed to
839 subprocesses with AT_SECURE
840 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
841 \f
842 Version 2.24
843
844 * The minimum Linux kernel version that this version of the GNU C Library
845 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
846 kernel version 2.6.32 or later suffices (on architectures that already
847 required kernel versions more recent than 3.2, those requirements remain
848 unchanged). Linux 3.2 or later kernel headers are required on all
849 architectures.
850
851 * The pap_AN locale has been deleted. This has been deprecated for a long
852 time. It has been replaced by pap_AW & pap_CW, both of which have long
853 been included in previous releases.
854
855 * The readdir_r and readdir64_r functions have been deprecated. It is
856 recommended to use readdir and readdir64 instead.
857
858 * The type “union wait” has been removed. It was deprecated in the early
859 1990s and never part of POSIX. Application code should use the int type
860 instead of “union wait”.
861
862 * A new NSS action is added to facilitate large distributed system
863 administration. The action, MERGE, allows remote user stores like LDAP
864 to be merged into local user stores like /etc/groups in order to provide
865 easy to use, updated, and managed sets of merged credentials. The new
866 action can be used by configuring it in /etc/nsswitch.conf:
867 group: files [SUCCESS=merge] nis
868 Implemented by Stephen Gallagher (Red Hat).
869
870 * The deprecated __malloc_initialize_hook variable has been removed from the
871 API.
872
873 * The long unused localedef --old-style option has been removed. It hasn't
874 done anything in over 16 years. Scripts using this option can safely
875 drop it.
876
877 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
878 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
879 return the next representable value in the direction of positive infinity
880 and the nextdown functions return the next representable value in the
881 direction of negative infinity. These are currently enabled as GNU
882 extensions.
883
884 Security related changes:
885
886 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
887 could result in a stack overflow when getnetbyname was called with an
888 overly long name. (CVE-2016-3075)
889
890 * Previously, getaddrinfo copied large amounts of address data to the stack,
891 even after the fix for CVE-2013-4458 has been applied, potentially
892 resulting in a stack overflow. getaddrinfo now uses a heap allocation
893 instead. Reported by Michael Petlan. (CVE-2016-3706)
894
895 * The glob function suffered from a stack-based buffer overflow when it was
896 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
897 Reported by Alexander Cherepanov. (CVE-2016-1234)
898
899 * The Sun RPC UDP client could exhaust all available stack space when
900 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
901 alloca plugin for GCC. (CVE-2016-4429)
902
903 * The IPv6 name server management code in libresolv could result in a memory
904 leak for each thread which is created, performs a failing naming lookup,
905 and exits. Over time, this could result in a denial of service due to
906 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
907
908 The following bugs are resolved with this release:
909
910 [1170] localedata: ne_NP: update Nepali locale definition file
911 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
912 of MS-DOS.
913 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
914 [6796] math: fdim() does not set errno on overflow
915 [10354] libc: posix_spawn should use vfork() in more cases than presently
916 [11213] localedata: localedata: add copyright disclaimer to locale files
917 [12143] localedata: chr_US: new Cherokee locale
918 [12450] localedata: sgs_LT: new locale
919 [12676] localedata: ln_CD: new locale
920 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
921 CLDR data
922 [13304] math: fma, fmaf, fmal produce wrong results
923 [14259] build: --localedir arg to configure is ignored
924 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
925 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
926 [14934] localedata: es_CL: wrong first weekday chilean locale
927 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
928 romanisation
929 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
930 and +/-
931 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
932 [15368] nptl: raise() is not async-signal-safe
933 [15479] math: ceil, floor, round and trunc raise inexact exception
934 [15578] localedata: kk_KZ: various updates
935 [16003] localedata: pap_AN: punt old locale
936 [16137] localedata: iw_IL: punt old locale
937 [16190] localedata: eo: new esperanto locale
938 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
939 [16742] malloc: race condition: pthread_atfork() called before first
940 malloc() results in unexpected locking behaviour/deadlocks
941 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
942 all locales
943 [16983] localedata: postal_fmt does not allow %l and %n modifiers
944 [17565] localedata: pt_PT: wrong (work-)week start
945 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
946 [17950] build: Build fails with -msse
947 [18205] localedata: be_BY*: wrong first_weekday and first_workday
948 [18433] libc: posix_spawn does not return correctly upon failure to
949 execute
950 [18453] localedata: charmaps/IBM875: incorrect codes
951 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
952 -Wsystem-headers
953 [18896] localedata: he_IL: improvements for currency
954 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
955 Romanian locale data
956 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
957 symbol
958 [19056] libc: Deprecate readdir_r
959 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
960 language
961 [19198] localedata: nl_NL: small improvements for Dutch locales
962 [19257] network: Per-thread memory leak in __res_vinit with IPv6
963 nameservers (CVE-2016-5417)
964 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
965 machine
966 [19400] locale: Language missing in "iso-639.def", trivial fix in
967 description
968 [19431] malloc: Deadlock between fflush, getdelim, and fork
969 [19505] libc: Incorrect file descriptor validity checks in
970 posix_spawn_file_actions_add{open,close,dup2}
971 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
972 when using RTLD_NEXT
973 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
974 `intl/{gettextP,loadinfo}.h'
975 [19534] libc: execle, execlp may use malloc
976 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
977 [19573] network: res_nclose and __res_maybe_init disagree about name
978 server initialization, breaking Hesiod
979 [19575] localedata: Status of GB18030 tables
980 [19581] localedata: sr_* date_fmt string contains additional newline
981 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
982 Excavator core
983 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
984 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
985 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
986 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
987 double range
988 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
989 part zero incorrect
990 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
991 equality tests
992 [19610] dynamic-link: ldconfig -X removes stale symbolic links
993 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
994 [19633] locale: strfmon_l applies global locale to number formatting
995 [19642] network: Memory leak in getnameinfo
996 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
997 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
998 glibc-2.22
999 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1000 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1001 'tst-numeric.c'
1002 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1003 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1004 low part
1005 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1006 result
1007 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1008 unaligned stack
1009 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1010 pointers and lengths in error-case.
1011 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1012 report errors on UTF-16 surrogates values.
1013 [19755] nscd: nscd assertion failure in gc
1014 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1015 [19759] libc: mempcpy shouldn't be inlined
1016 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1017 [19765] libc: s390 needs an optimized mempcpy
1018 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1019 NAME_MAX limit assumption (CVE-2016-1234)
1020 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1021 [19787] network: Missing and incorrect truncation checks in getnameinfo
1022 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1023 modes
1024 [19791] network: Assertion failure in res_query.c with un-connectable name
1025 server addresses
1026 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1027 [19822] math: libm.so install clobbers old version
1028 [19825] network: resolv: send_vc can return uninitialized data in second
1029 response to getaddrinfo
1030 [19830] network: nss_dns: should check RDATA length against buffer length
1031 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1032 confronted with A/AAAA records of invalid size
1033 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1034 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1035 [19853] stdio: Printing IBM long double in decimal with high precision is
1036 sometimes incorrect
1037 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1038 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1039 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1040 record types
1041 [19865] network: Assertion failure or memory leak in
1042 _nss_dns_getcanonname_r
1043 [19868] network: nss_dns: netent code does not skip over non-PTR records
1044 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1045 (CVE-2016-3075)
1046 [19881] string: Improve x86-64 memset
1047 [19907] string: Incorrect memcpy tests
1048 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1049 resolving symbols
1050 [19925] libc: termios.h XCASE namespace
1051 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1052 [19929] libc: limits.h NL_NMAX namespace
1053 [19931] stdio: Memory leak in vfprintf
1054 [19957] libc: clone(CLONE_VM) access invalid parent memory
1055 [19963] localedata: en_IL: New locale
1056 [19989] stdio: stdio.h cuserid namespace
1057 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1058 gethosts
1059 [19996] locale: langinfo.h nl_langinfo_l namespace
1060 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1061 [20010] network: getaddrinfo: Stack overflow in hostent translation
1062 (CVE-2016-3706)
1063 [20012] stdio: libio: fmemopen append mode failure
1064 [20014] stdio: stdio.h namespace for pre-threads POSIX
1065 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1066 [20023] libc: fcntl.h timespec namespace
1067 [20024] math: [x86_64] vectorized sincos trashes the stack
1068 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1069 [20041] time: sys/time.h timespec namespace
1070 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1071 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1072 [20051] libc: ttyslot in wrong header under wrong conditions
1073 [20054] libc: gethostname not declared for XPG4
1074 [20055] libc: termios.h missing tcgetsid for XPG4
1075 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1076 executable
1077 [20073] libc: sys/stat.h fchmod namespace
1078 [20074] libc: stdlib.h rand_r namespace
1079 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1080 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1081 XPG3
1082 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1083 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1084 (CVE-2016-4429)
1085 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1086 [20119] libc: Wrong mask for processors level type from CPUID
1087 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1088 AS not supporting AVX512
1089 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1090 sNaN argument
1091 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1092 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1093 argument
1094 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1095 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1096 [20178] libc: posix_spawn{p} should not call exit
1097 [20191] stdio: libio: vtables hardening
1098 [20195] string: FMA4 detection requires CPUID execution with register
1099 eax=0x80000001
1100 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1101 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1102 subnormals
1103 [20212] math: acos (sNaN) returns sNaN
1104 [20213] math: asin (sNaN) returns sNaN
1105 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1106 [20218] math: [i386] asinhl (sNaN) returns sNaN
1107 [20219] math: [i386] atanhl (sNaN) returns sNaN
1108 [20222] stdio: fopencookie: Mangle function pointers
1109 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1110 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1111 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1112 input
1113 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1114 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1115 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1116 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1117 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1118 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1119 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1120 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1121 [20240] math: modf (sNaN) returns sNaN
1122 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1123 [20250] math: frexp (sNaN) returns sNaN
1124 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1125 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1126 double rounding
1127 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1128 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1129 subscript is below array bounds [-Werror=array-bounds]
1130 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1131 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1132 (Only arm/linux)
1133 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1134 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1135 fallbacks
1136 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1137 "invalid" exceptions
1138 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1139 [20316] localedata: id_ID: Februari instead of Pebruari
1140 [20327] string: POWER8 strcasecmp returns incorrect result
1141 [20347] math: Failure: Test: j0_downward (0xap+0)
1142 [20348] libc: FAIL: misc/tst-preadvwritev64
1143 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1144 [20350] libc: There is no test for p{read,write}64
1145 [20357] math: Incorrect cos result for 1.5174239687223976
1146 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1147 \f
1148 Version 2.23
1149
1150 * Unicode 8.0.0 Support: Character encoding, character type info, and
1151 transliteration tables are all updated to Unicode 8.0.0, using new
1152 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1153 These updates cause user visible changes, such as the fixes for bugs
1154 89, 16061, and 18568.
1155
1156 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1157 kernel-internal CPU set size. This means that requests that change the
1158 CPU affinity which failed before (for example, an all-ones CPU mask) will
1159 now succeed. Applications that need to determine the effective CPU
1160 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1161 setting it because the kernel can adjust it (and the previous size check
1162 would not detect this in the majority of cases).
1163
1164 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1165 the following new symbols are used: fts64_children, fts64_close,
1166 fts64_open, fts64_read and fts64_set.
1167
1168 * getaddrinfo now detects certain invalid responses on an internal netlink
1169 socket. If such responses are received, an affected process will
1170 terminate with an error message of "Unexpected error <number> on netlink
1171 descriptor <number>" or "Unexpected netlink response of size <number> on
1172 descriptor <number>". The most likely cause for these errors is a
1173 multi-threaded application which erroneously closes and reuses the netlink
1174 file descriptor while it is used by getaddrinfo.
1175
1176 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1177 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1178 unnecessary serialization of memory allocation requests across threads.
1179 The defect is now corrected. Users should see a substantial increase in
1180 the concurent throughput of allocation requests for applications which
1181 trigger this bug. Affected applications typically create and
1182 destroy threads frequently. (Bug 19048 was reported and analyzed by
1183 Ericsson.)
1184
1185 * There is now a --disable-timezone-tools configure option for disabling the
1186 building and installing of the timezone related utilities (zic, zdump, and
1187 tzselect). This is useful for people who build the timezone data and code
1188 independent of the GNU C Library.
1189
1190 * The obsolete header <regexp.h> has been removed. Programs that require
1191 this header must be updated to use <regex.h> instead.
1192
1193 * The obsolete functions bdflush, create_module, get_kernel_syms,
1194 query_module and uselib are no longer available to newly linked binaries;
1195 the header <sys/kdaemon.h> has been removed. These functions and header
1196 were specific to systems using the Linux kernel and could not usefully be
1197 used with the GNU C Library on systems with version 2.6 or later of the
1198 Linux kernel.
1199
1200 * Optimized string, wcsmbs and memory functions for IBM z13.
1201 Implemented by Stefan Liebler.
1202
1203 * Newly linked programs that define a variable called signgam will no longer
1204 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1205 require signgam to be set by those functions must ensure that they use the
1206 variable provided by the GNU C Library and declared in <math.h>, without
1207 defining their own copy.
1208
1209 * The minimum GCC version that can be used to build this version of the GNU
1210 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1211 still be used to compile programs using the GNU C Library.
1212
1213 Security related changes:
1214
1215 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1216 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1217
1218 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1219 the pointer guard feature. It is always enabled. Previously,
1220 LD_POINTER_GUARD could be used to disable security hardening in binaries
1221 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1222 (CVE-2015-8777)
1223
1224 * An integer overflow in hcreate and hcreate_r could lead to an
1225 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1226
1227 * The catopen function no longer has unbounded stack usage. Reported by
1228 Max. (CVE-2015-8779)
1229
1230 * The nan, nanf and nanl functions no longer have unbounded stack usage
1231 depending on the length of the string passed as an argument to the
1232 functions. Reported by Joseph Myers. (CVE-2014-9761)
1233
1234 * A stack-based buffer overflow was found in libresolv when invoked from
1235 libnss_dns, allowing specially crafted DNS responses to seize control
1236 of execution flow in the DNS client. The buffer overflow occurs in
1237 the functions send_dg (send datagram) and send_vc (send TCP) for the
1238 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1239 family. The use of AF_UNSPEC triggers the low-level resolver code to
1240 send out two parallel queries for A and AAAA. A mismanagement of the
1241 buffers used for those queries could result in the response of a query
1242 writing beyond the alloca allocated buffer created by
1243 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1244 the overflow. Thanks to the Google Security Team and Red Hat for
1245 reporting the security impact of this issue, and Robert Holiday of
1246 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1247
1248 The following bugs are resolved with this release:
1249
1250 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1251 [887] math: Math library function "logb" and "nextafter" inconsistent
1252 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1253 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1254 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1255 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1256 use `mkstemp'
1257 [4404] localedata: German translation of "Alarm clock" is misleading
1258 [6799] math: nextafter() and nexttoward() doen't set errno on
1259 overflow/underflow errors
1260 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1261 overflow/underflow
1262 [10432] nis: _nss_nis_setnetgrent assertion failure
1263 [11460] libc: fts has no LFS support
1264 [12926] network: getaddrinfo()/make_request() may spin forever
1265 [13065] nptl: Race condition in pthread barriers
1266 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1267 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1268 are not contiguous
1269 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1270 [14912] libc: Rename non-installed bits/*.h headers
1271 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1272 [15367] math: Let gcc use __builtin_isinf
1273 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1274 [15421] math: lgamma wrongly sets signgam for ISO C
1275 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1276 FE_INVALID with argument out of range
1277 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1278 all exceptions
1279 [15786] dynamic-link: ifunc resolver functions can smash function
1280 arguments
1281 [15918] math: Unnecessary check for equality in hypotf()
1282 [16061] localedata: Review / update transliteration data
1283 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1284 should include
1285 [16141] time: strptime %z offset restriction
1286 [16171] math: drem should be alias of remainder
1287 [16296] math: fegetround is pure?
1288 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1289 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1290 GNU/Linux
1291 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1292 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1293 arguments
1294 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1295 "Exception "Inexact" set" on ppc32
1296 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1297 [16517] math: Missing underflow exception from tanf/tan/tanl
1298 [16519] math: Missing underflow exception from sinhf
1299 [16520] math: Missing underflow exception from tanhf
1300 [16521] math: Missing underflow exception from exp2
1301 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1302 rounding results
1303 [16734] stdio: fopen calls mmap to allocate its buffer
1304 [16961] math: nan function incorrect handling of bad sequences
1305 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1306 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1307 [16985] locale: localedef: confusing error message when opening output
1308 fails
1309 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1310 [17197] locale: Redundant shift character in iconv conversion output at
1311 block boundary
1312 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1313 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1314 [17250] dynamic-link: static linking breaks nss loading
1315 (getaddrinfo/getpwnam/etc...)
1316 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1317 4.7?
1318 [17441] math: isnan() should use __builtin_isnan() in GCC
1319 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1320 (related to lock elision)
1321 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1322 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1323 [17887] time: strptime should be able to parse "+01:00" style timezones
1324 [17905] libc: catopen() Multiple unbounded stack allocations
1325 (CVE-2015-8779)
1326 [18084] libc: backtrace (..., 0) dumps core on x86
1327 [18086] libc: nice() sets errno to 0 on success
1328 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1329 is too large (CVE-2015-8778)
1330 [18251] dynamic-link: SONAME missing when audit modules provides path
1331 [18265] libc: add attributes for wchar string and memory functions
1332 [18370] math: csqrt missing underflows
1333 [18421] libc: [hppa] read-only segment has dynamic relocations
1334 [18472] libc: Obsolete syscall wrappers should be compat symbols
1335 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1336 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1337 [18525] localedata: Remove locale timezone information
1338 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1339 [18568] localedata: Update locale data to Unicode 8.0
1340 [18589] locale: sort-test.sh fails at random
1341 [18595] math: ctan, ctanh missing underflows
1342 [18604] libc: assert macro-expands its argument
1343 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1344 contains a vector instruction exception.
1345 [18611] math: j1, jn missing errno setting on underflow
1346 [18618] localedata: sync Chechen locale definitions with other *_RU
1347 locales
1348 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1349 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1350 [18665] network: In send_dg, the recvfrom function is NOT always using the
1351 buffer size of a newly created buffer (CVE-2015-7547)
1352 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1353 missing break ?
1354 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1355 32bit processes
1356 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1357 [18699] math: tilegx cproj() for various complex infinities does not yield
1358 infinity
1359 [18724] libc: Harden put*ent functions against data injection
1360 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1361 [18755] build: build errors with -DNDEBUG
1362 [18757] stdio: fmemopen fails to set errno on failure
1363 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1364 be forced unloaded
1365 [18781] libc: openat64 lacks O_LARGEFILE
1366 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1367 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1368 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1369 [18790] math: [ldbl-128ibm] tanhl inaccurate
1370 [18795] libc: stpncpy fortification misses buffer lengths that are
1371 statically too large
1372 [18796] build: build fails for --disable-mathvec
1373 [18803] math: hypot missing underflows
1374 [18820] stdio: fmemopen may leak memory on failure
1375 [18823] math: csqrt spurious underflows
1376 [18824] math: fma spurious underflows
1377 [18825] math: pow missing underflows
1378 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1379 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1380 [18870] build: sem_open.c fails to compile with missing symbol
1381 FUTEX_SHARED
1382 [18872] stdio: Fix memory leak in printf_positional
1383 [18873] libc: posix_fallocate overflow check ineffective
1384 [18875] math: Excess precision leads incorrect libm
1385 [18877] libc: arm: mmap offset regression
1386 [18887] libc: memory corruption when using getmntent on blank lines
1387 [18918] localedata: hu_HU: change time to HH:MM:SS format
1388 [18921] libc: Regression: extraneous stat() and fstat() performed by
1389 opendir()
1390 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1391 binaries (CVE-2015-8777)
1392 [18951] math: tgamma missing underflows
1393 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1394 signgam
1395 [18953] localedata: lt_LT: change currency symbol to the euro
1396 [18956] math: powf inaccuracy
1397 [18961] math: [i386] exp missing underflows
1398 [18966] math: [i386] exp10 missing underflows
1399 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1400 [18969] build: multiple string test failures due to missing locale
1401 dependencies
1402 [18970] libc: Reference of pthread_setcancelstate in libc.a
1403 [18977] math: float / long double Bessel functions not in XSI POSIX
1404 [18980] math: i386 libm functions return with excess range and precision
1405 [18981] math: i386 scalb*, ldexp return with excess range and precision
1406 [18982] stdio: va_list and vprintf
1407 [18985] time: Passing out of range data to strftime() causes a segfault
1408 (CVE-2015-8776)
1409 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1410 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1411 [19012] locale: iconv_open leaks memory on error path
1412 [19016] math: clog, clog10 inaccuracy
1413 [19018] nptl: Mangle function pointers in tls_dtor_list
1414 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1415 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1416 [19048] malloc: malloc: arena free list can become cyclic, increasing
1417 contention
1418 [19049] math: [powerpc] erfc incorrect zero sign
1419 [19050] math: [powerpc] log* incorrect zero sign
1420 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1421 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1422 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1423 [19074] network: Data race in _res_hconf_reorder_addrs
1424 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1425 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1426 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1427 rounding modes
1428 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1429 ILP32
1430 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1431 [19086] manual: posix_fallocate64 documented argument order is wrong.
1432 [19088] math: lround, llround missing exceptions close to overflow
1433 threshold
1434 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1435 [19095] math: dbl-64 lrint incorrect for 64-bit long
1436 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1437 [19124] dynamic-link: ld.so failed to build with older assmebler
1438 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1439 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1440 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1441 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1442 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1443 pthread_setaffinity_np
1444 [19156] math: [ldbl-128] j0l spurious underflows
1445 [19164] nptl: tst-getcpu fails with many possible CPUs
1446 [19168] math: math/test-ildoubl and math/test-ldouble failure
1447 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1448 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1449 prelink
1450 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1451 (FE_NOMASK_ENV) do not clear SSE exceptions
1452 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1453 _int_new_arena/reused_arena
1454 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1455 [19201] math: dbl-64 remainder incorrect sign of zero result
1456 [19205] math: bits/math-finite.h conditions do not match math.h and
1457 bits/mathcalls.h
1458 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1459 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1460 for C99-based standards
1461 [19212] libc: features.h not -Wundef clean
1462 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1463 math-only
1464 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1465 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1466 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1467 disabled
1468 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1469 "inexact" exceptions
1470 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1471 arguments
1472 [19242] libc: strtol incorrect in Turkish locales
1473 [19243] malloc: reused_arena can pick an arena on the free list, leading
1474 to an assertion failure and reference count corruption
1475 [19253] time: tzset() ineffective when temporary TZ did not include DST
1476 rules
1477 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1478 [19270] math: [hppa] Shared libm missing __isnanl
1479 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1480 MAP_HUGETLB and MAP_STACK defines
1481 [19313] nptl: Wrong __cpu_mask for x32
1482 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1483 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1484 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1485 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1486 [19363] time: x32: times() return value wrongly truncates/sign extends
1487 from 32bit
1488 [19367] dynamic-link: Improve branch prediction on Silvermont
1489 [19369] network: Default domain name not reset by res_ninit when "search"
1490 / "domain" entry is removed from resolv.conf
1491 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1492 subnormals
1493 [19385] localedata: bg_BG: time separator should be colon, not comma
1494 [19408] libc: linux personality syscall wrapper may erroneously return an
1495 error on 32-bit architectures
1496 [19415] libc: dladdr returns wrong names on hppa
1497 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1498 IBM905, IBM907, and IBM909
1499 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1500 [19443] build: build failures with -DDEBUG
1501 [19451] build: Make check fails on test-double-vlen2
1502 [19462] libc: Glibc failed to build with -Os
1503 [19465] math: Wrong code with -Os
1504 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1505 -Os
1506 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1507 CPU's.
1508 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1509 [19486] math: S390: Math tests fail with "Exception Inexact set".
1510 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1511 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1512 architectures
1513 [19590] math: Fail to build shared objects that use libmvec.so functions.
1514 \f
1515 Version 2.22
1516
1517 * The following bugs are resolved with this release:
1518
1519 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1520 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1521 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1522 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1523 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1524 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1525 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1526 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1527 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1528 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1529 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1530 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1531 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1532 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1533 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1534 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1535 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1536 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1537 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1538 18657, 18676, 18694, 18696, 18887.
1539
1540 * Cache information can be queried via sysconf() function on s390 e.g. with
1541 _SC_LEVEL1_ICACHE_SIZE as argument.
1542
1543 * A buffer overflow in gethostbyname_r and related functions performing DNS
1544 requests has been fixed. If the NSS functions were called with a
1545 misaligned buffer, the buffer length change due to pointer alignment was
1546 not taken into account. This could result in application crashes or,
1547 potentially arbitrary code execution, using crafted, but syntactically
1548 valid DNS responses. (CVE-2015-1781)
1549
1550 * The time zone file parser has been made more robust against crafted time
1551 zone files, avoiding heap buffer overflows related to the processing of
1552 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1553 large time zone data files. Overly long time zone specifiers in the TZ
1554 variable no longer result in stack overflows and crashes.
1555
1556 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1557 for LD and GD on x86 and x86-64, has been implemented. You will need
1558 binutils-2.24 or later to enable this optimization.
1559
1560 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1561 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1562 Hat). These updates cause user visible changes, such as the fix for bug
1563 17998.
1564
1565 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1566 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1567 condition in some applications.
1568
1569 * Added vector math library named libmvec with the following vectorized x86_64
1570 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1571 pow, powf.
1572 The library can be disabled with --disable-mathvec. Use of the functions is
1573 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1574 Shared library libmvec.so is linked in as needed when using -lm (no need to
1575 specify -lmvec explicitly for not static builds).
1576 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1577
1578 * A new fmemopen implementation has been added with the goal of POSIX
1579 compliance. The new implementation fixes the following long-standing
1580 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1581 old implementation is still present for use be by existing binaries.
1582
1583 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1584 releases. It has been fixed to match 2.19 and older, but binaries built
1585 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1586
1587 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1588 Contributed by Roland McGrath (Google).
1589
1590 * The header <regexp.h> is deprecated, and will be removed in a future
1591 release. Use of this header will trigger a deprecation warning.
1592 Application developers should update their code to use <regex.h> instead.
1593
1594 This header was formerly part of SUS, but was deprecated in 1994 and
1595 removed from the standard in 2001. Also, the glibc implementation
1596 leaks memory. See BZ#18681 for more details.
1597 \f
1598 Version 2.21
1599
1600 * The following bugs are resolved with this release:
1601
1602 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1603 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1604 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1605 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1606 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1607 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1608 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1609 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1610 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1611 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1612 17892.
1613
1614 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1615 too little memory for the to-be-scanned arguments and overflow the
1616 allocated buffer. The implementation now correctly computes the required
1617 buffer size when using malloc, and switches to malloc from alloca as
1618 intended.
1619
1620 * A new semaphore algorithm has been implemented in generic C code for all
1621 machines. Previous custom assembly implementations of semaphore were
1622 difficult to reason about or ensure that they were safe. The new version
1623 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1624 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1625 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1626
1627 * Port to Altera Nios II has been contributed by Mentor Graphics.
1628
1629 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1630 implementations for powerpc64/powerpc64le.
1631 Implemented by Adhemerval Zanella (IBM).
1632
1633 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1634 and powerpc64le. This may improve lock scaling of existing programs on
1635 HTM capable systems. The lock elision code is only enabled with
1636 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1637 powerpc will issue a transaction abort on every syscall to avoid side
1638 effects being visible outside transactions.
1639
1640 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1641 AArch64. Contributed by ARM Ltd.
1642
1643 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1644
1645 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1646 under certain input conditions resulting in the execution of a shell for
1647 command substitution when the application did not request it. The
1648 implementation now checks WRDE_NOCMD immediately before executing the
1649 shell and returns the error WRDE_CMDSUB as expected.
1650
1651 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1652 processing format strings with a large number of format specifiers.
1653
1654 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1655 infinite loop if the DNS response contained a PTR record of an unexpected
1656 format.
1657
1658 * The minimum GCC version that can be used to build this version of the GNU
1659 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1660 still be used to compile programs using the GNU C Library.
1661
1662 * The GNU C Library is now built with -Werror by default. This can be
1663 disabled by configuring with --disable-werror.
1664
1665 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1666
1667 * The obsolete sigvec function has been removed. This was the original
1668 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1669 programs have been using instead for about 25 years. Of course, ABI
1670 compatibility for old binaries using sigvec remains intact.
1671
1672 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1673 with newer versions of bison.
1674
1675 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1676 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1677 registers overlay two consecutive single-precision registers. MIPS32R2
1678 introduced a new FPU mode (FR=1) where double-precision registers extend the
1679 corresponding single-precision registers which is incompatible with the
1680 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1681 require the use of FR=1 making a transition necessary. New o32 ABI
1682 extensions enable users to migrate over time from the original o32 ABI
1683 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1684 tracks the ABI of any loaded object and verifies that new objects are
1685 compatible. Mode transitions will also be requested as required and
1686 unsupportable objects will be rejected. The ABI checks include both soft and
1687 hard float ABIs for o32, n32 and n64.
1688
1689 GCC 5 with GNU binutils 2.25 onwards:
1690 It is strongly recommended that all o32 system libraries are built using the
1691 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1692 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1693 compiler using --with-fp-32=xx to set this by default.
1694 \f
1695 Version 2.20
1696
1697 * The following bugs are resolved with this release:
1698
1699 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1700 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1701 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1702 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1703 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1704 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1705 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1706 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1707 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1708 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1709 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1710 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1711 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1712 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1713 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1714
1715 * Reverted change of ABI data structures for s390 and s390x:
1716 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1717 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1718 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1719 are preserved pointing straight to the same implementation as the old ones.
1720 Given that, new callers will simply provide a too-big buffer to these
1721 functions. Any applications/libraries out there that embed jmp_buf or
1722 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1723 against 2.19 headers will have to rebuilt again. This is necessary in any
1724 case to revert the breakage in their ABI caused by the glibc change.
1725
1726 * Support for file description locks is added to systems running the
1727 Linux kernel. The standard file locking interfaces are extended to
1728 operate on file descriptions, not file descriptors, via the use of
1729 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1730 are associated with an open file instead of a process.
1731
1732 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1733
1734 * The minimum Linux kernel version that this version of the GNU C Library
1735 can be used with is 2.6.32.
1736
1737 * Running the testsuite no longer terminates as soon as a test fails.
1738 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1739 with PASS or FAIL lines for individual tests. A summary of the results is
1740 printed, including a list of failing lists, and "make check" exits with
1741 error status if there were any unexpected failures. "make check
1742 stop-on-test-failure=y" may be used to keep the old behavior.
1743
1744 * The am33 port, which had not worked for several years, has been removed
1745 from ports.
1746
1747 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1748 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1749 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1750 conflicted with POSIX (support for which was removed in 2.19), the
1751 interfaces those macros enabled remain available when compiling with
1752 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1753 test macros defined.
1754
1755 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1756
1757 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1758 This may improve lock scaling of existing programs on TX capable systems.
1759 The lock elision code is only built with --enable-lock-elision=yes and
1760 then requires a GCC version supporting the TX builtins. With lock elision
1761 default mutexes are elided via __builtin_tbegin, if the cpu supports
1762 transactions. By default lock elision is not enabled and the elision code
1763 is not built.
1764
1765 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1766 copy the path argument. This allowed programs to cause posix_spawn to
1767 deference a dangling pointer, or use an unexpected pathname argument if
1768 the string was modified after the posix_spawn_file_actions_addopen
1769 invocation.
1770
1771 * All supported architectures now use the main glibc sysdeps directory
1772 instead of some being in a separate "ports" directory (which was
1773 distributed separately before glibc 2.17).
1774
1775 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1776 On configurations that support it (all Linux configurations), it's now
1777 used regardless of the --enable-add-ons switch to configure. It is no
1778 longer possible to build such configurations without pthreads support.
1779
1780 * Locale names, including those obtained from environment variables (LANG
1781 and the LC_* variables), are more tightly checked for proper syntax.
1782 setlocale will now fail (with EINVAL) for locale names that are overly
1783 long, contain slashes without starting with a slash, or contain ".." path
1784 components. (CVE-2014-0475) Previously, some valid locale names were
1785 silently replaced with the "C" locale when running in AT_SECURE mode
1786 (e.g., in a SUID program). This is no longer necessary because of the
1787 additional checks.
1788
1789 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1790 with application code using Intel MPX instructions. (With all previous
1791 versions, the MPX register state could be clobbered when making calls
1792 into or out of a shared library.) Note that while the new dynamic
1793 linker is compatible with all known x86 hardware whether or not it
1794 supports Intel MPX, some x86 instruction-set emulators might fail to
1795 handle the new instruction encodings. This is known to affect Valgrind
1796 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1797 release), and might affect other tools that do instruction emulation.
1798
1799 * Support for loadable gconv transliteration modules has been removed.
1800 The support for transliteration modules has been non-functional for
1801 over a decade, and the removal is prompted by security defects. The
1802 normal gconv conversion modules are still supported. Transliteration
1803 with //TRANSLIT is still possible, and the //IGNORE specifier
1804 continues to be supported. (CVE-2014-5119)
1805
1806 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1807 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1808 resulting a denial-of-service security vulnerability in applications which
1809 use functions related to iconv. (CVE-2014-6040)
1810 \f
1811 Version 2.19
1812
1813 * The following bugs are resolved with this release:
1814
1815 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1816 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1817 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1818 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1819 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1820 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1821 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1822 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1823 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1824 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1825 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1826 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1827 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1828 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1829 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1830 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1831 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1832 16431, 16453, 16474, 16506, 16510, 16529
1833
1834 * Slovenian translations for glibc messages have been contributed by the
1835 Translation Project's Slovenian team of translators.
1836
1837 * The public headers no longer use __unused nor __block. This change is to
1838 support compiling programs that are derived from BSD sources and use
1839 __unused internally, and to support compiling with Clang's -fblock
1840 extension which uses __block.
1841
1842 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1843 large collation sequences to optimize multiple passes. This cache
1844 computation may overflow for large collation sequences and may cause a
1845 stack or buffer overflow. This is now fixed to use a slower algorithm
1846 which does not use a cache if there is an integer overflow.
1847
1848 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1849 rules for large collation sequences to optimize multiple passes and falls
1850 back to alloca if malloc fails, resulting in a possible stack overflow.
1851 The implementation now falls back to an uncached collation sequence lookup
1852 if malloc fails.
1853
1854 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1855 initialized for static applications resulting in the security feature
1856 being disabled. The pointer guard is now correctly initialized to a
1857 random value for static applications. Existing static applications need
1858 to be recompiled to take advantage of the fix (bug 15754).
1859
1860 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1861 to the d_name member of struct dirent, or omit the terminating NUL
1862 character. (Bugzilla #14699).
1863
1864 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1865 aligned_alloc functions could allocate too few bytes or corrupt the
1866 heap when passed very large allocation size values (Bugzilla #15855,
1867 #15856, #15857).
1868
1869 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1870 for AF_INET6 has been fixed (Bugzilla #16072).
1871
1872 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1873 pap_CW, quz_PE, the_NP.
1874
1875 * Substantially revised locales: gd_GB, ht_HT
1876
1877 * The LC_ADDRESS field was updated to support country_car for almost all
1878 supported locales.
1879
1880 * ISO 1427 definitions were updated.
1881
1882 * ISO 3166 definitions were updated.
1883
1884 * The localedef utility now supports --big-endian and --little-endian
1885 command-line options to generate locales for a different system from that
1886 for which the C library was built.
1887
1888 * Binary locale files now only depend on the endianness of the system for
1889 which they are generated and not on other properties of that system. As a
1890 consequence, binary files generated with new localedef may be incompatible
1891 with old versions of the GNU C Library, and binary files generated with
1892 old localedef may be incompatible with this version of the GNU C Library,
1893 in the following circumstances:
1894
1895 + Locale files may be incompatible on m68k systems.
1896
1897 + Locale archive files (but not separate files for individual locales) may
1898 be incompatible on systems where plain "char" is signed.
1899
1900 * The configure option --disable-versioning has been removed. Builds with
1901 --disable-versioning had not worked for several years.
1902
1903 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1904 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1905
1906 * SystemTap probes for malloc have been introduced.
1907
1908 * SystemTap probes for slow multiple precision fallback paths of
1909 transcendental functions have been introduced.
1910
1911 * Support for powerpc64le has been added.
1912
1913 * The soft-float powerpc port now supports e500 processors.
1914
1915 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1916
1917 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1918 set of header declarations that are enabled by default, even when other
1919 feature test macros or compiler options such as -std=c99 would otherwise
1920 disable some of those declarations.
1921
1922 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1923 conflict with POSIX. The libbsd-compat library (which was a dummy library
1924 that did nothing) has also been removed.
1925
1926 * Preliminary documentation about Multi-Thread, Async-Signal and
1927 Async-Cancel Safety has been added.
1928
1929 * Change of ABI data structures for s390 and s390x:
1930 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1931 allow for future hardware extensions. All library functions that accept or
1932 return these structures were versioned in order to provide backwards
1933 compatibility with existing code. However, not all cases can be handled
1934 with interface versioning. If an application embeds either structure into
1935 another structure and that structure is passed to another compilation unit
1936 which expects the newer larger-sized object then any access to the new
1937 fields will result in undefined behaviour. Similarly any access to fields
1938 that were moved by the enlarging of the structures will cause undefined
1939 behaviour. To fix the undefined behaviour all objects that are part of the
1940 final in-memory image must agree on the size of structures, and this may
1941 require recompilation.
1942 \f
1943 Version 2.18
1944
1945 * The following bugs are resolved with this release:
1946
1947 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1948 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1949 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1950 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1951 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1952 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1953 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1954 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1955 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1956 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1957 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1958 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1959 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1960 15755, 15759.
1961
1962 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1963 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1964 Distributions can re-enable building and using pt_chown via the new configure
1965 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1966 considerable security risks and should only be used if the distribution
1967 understands and accepts the risks.
1968
1969 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1970 #15078).
1971
1972 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1973 fixed (Bugzilla #15330).
1974
1975 * Add support for calling C++11 thread_local object destructors on thread
1976 and program exit. This needs compiler support for offloading C++11
1977 destructor calls to glibc.
1978
1979 * Improved worst case performance of libm functions with double inputs and
1980 output.
1981
1982 * Support for priority inherited mutexes in pthread condition variables on
1983 non-x86 architectures.
1984
1985 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1986
1987 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1988
1989 * Optimized string functions for ARM. Implemented by Will Newton and
1990 Richard Henderson.
1991
1992 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1993
1994 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1995 Richard Henderson.
1996
1997 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1998 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1999
2000 * Added a benchmark framework to track performance of functions in glibc.
2001
2002 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2003 It is based on draft TS 18661 and currently enabled as a GNU extension.
2004
2005 * On Linux, the clock function now uses the clock_gettime system call
2006 for improved precision, rather than old times system call.
2007
2008 * Added support for version-3 tz format files. This is needed when using
2009 the tz database release 2013e or later, and affects a few unusual cases --
2010 currently only TZ='America/Godthab' for time stamps after 2037.
2011
2012 * Added new API functions pthread_getattr_default_np and
2013 pthread_setattr_default_np to get and set the default pthread
2014 attributes of a process.
2015
2016 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2017 This may improve lock scaling of existing programs on TSX capable systems.
2018 When the --enable-lock-elision=yes parameter is specified at configure
2019 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2020 mutexes.
2021
2022 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2023 availability). Implemented by Ryan S. Arnold.
2024
2025 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2026 \f
2027 Version 2.17
2028
2029 * The following bugs are resolved with this release:
2030
2031 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2032 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2033 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2034 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2035 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2036 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2037 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2038 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2039 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2040 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2041 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2042 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2043 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2044 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2045 14879, 14889, 14893, 14898, 14914.
2046
2047 * Optimization of memcpy for MIPS.
2048
2049 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2050 EMFILE has been fixed (Bugzilla #14889).
2051
2052 * The add-on ports collection is now distributed in the "ports" subdirectory
2053 of the main GNU C Library distribution, rather than separately.
2054
2055 * Port to ARM AArch64 contributed by Linaro.
2056
2057 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2058 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2059 zEnterprise z196.
2060 Implemented by Andreas Krebbel.
2061
2062 * The new function secure_getenv allows secure access to the environment,
2063 returning NULL if running in a SUID/SGID process. This function replaces
2064 the internal function __secure_getenv.
2065
2066 * SystemTap static probes have been added into the dynamic linker.
2067 Implemented by Gary Benson.
2068
2069 * Optimizations of string functions strstr, strcasestr and memmem.
2070 Implemented by Maxim Kuvyrkov.
2071
2072 * The minimum Linux kernel version that this version of the GNU C Library
2073 can be used with is 2.6.16.
2074
2075 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2076 powerpc POWER7. Implemented by Will Schmidt.
2077
2078 * New configure option --disable-nscd builds the C library such that it
2079 never attempts to contact the Name Service Caching Daemon (nscd).
2080 New configure option --disable-build-nscd avoids building nscd itself;
2081 this is the default if --disable-nscd is used.
2082
2083 * Improved support for cross-compilation, including cross-testing and
2084 bootstrap builds without a previously built glibc.
2085
2086 * Several testsuite tests are now able to test multiple IFUNC variants of an
2087 interface, rather than just testing the one that would be chooen by
2088 default.
2089
2090 * New configure options --with-bugurl and --with-pkgversion, for
2091 distributors to use to embed their bug-reporting and package version
2092 information in --help and --version output.
2093
2094 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2095 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2096 allows creation of chroots without the procfs mounted on /proc.
2097
2098 * The `crypt' function now fails if passed salt bytes that violate the
2099 specification for those values. On Linux, the `crypt' function will
2100 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2101 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2102 when the mode is enabled.
2103
2104 * The `clock_*' suite of functions (declared in <time.h>) is now available
2105 directly in the main C library. Previously it was necessary to link with
2106 -lrt to use these functions. This change has the effect that a
2107 single-threaded program that uses a function such as `clock_gettime' (and
2108 is not linked with -lrt) will no longer implicitly load the pthreads
2109 library at runtime and so will not suffer the overheads associated with
2110 multi-thread support in other code such as the C++ runtime library.
2111
2112 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2113 sat_IN, and szl_PL.
2114 \f
2115 Version 2.16
2116
2117 * The following bugs are resolved with this release:
2118
2119 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2120 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2121 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2122 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2123 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2124 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2125 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2126 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2127 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2128 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2129 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2130 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2131 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2132 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2133 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2134 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2135 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2136 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2137 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2138 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2139 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2140 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2141 14277, 14278.
2142
2143 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2144 configuring glibc with:
2145 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2146 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2147 Implemented by H.J. Lu.
2148
2149 * ISO C11 support:
2150
2151 + define static_assert
2152
2153 + do not declare gets
2154
2155 + declare at_quick_exit and quick_exit also for ISO C11
2156
2157 + aligned_alloc. NB: The code is deliberately allows the size parameter
2158 to not be a multiple of the alignment. This is a moronic requirement
2159 in the standard but it is only a requirement on the caller, not the
2160 implementation.
2161
2162 + timespec_get added
2163
2164 + uchar.h support added
2165
2166 + CMPLX, CMPLXF, CMPLXL added
2167
2168 Implemented by Ulrich Drepper.
2169
2170 * Support for the IA-64 has been moved to ports.
2171
2172 * Remove support for anything but ELF binary format
2173
2174 * Checking versions of poll, ppoll added.
2175 Implemented by Ulrich Drepper.
2176
2177 * More generic and 64-bit performance optimizations to math functions.
2178 Implemented by Ulrich Drepper.
2179
2180 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2181 headers and functions available at compile time as they were before
2182 version 2.14. This option will be removed at some time in the future
2183 after the TI-RPC library becomes fully sufficient for the needs of
2184 existing applications.
2185
2186 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2187 Note that glibc is not expected to work with any Linux kernel version
2188 before 2.6.
2189
2190 * New header <sys/auxv.h> and function getauxval allowing easy access to
2191 the AT_* key-value pairs passed from the Linux kernel. The header also
2192 defines the HWCAP_* bits associated with the AT_HWCAP key.
2193
2194 * New locales: mag_IN
2195
2196 * New configure option --enable-systemtap builds SystemTap static probes
2197 into libc for setjmp and longjmp and into libpthread for various operations.
2198 So far the setjmp/longjmp probes and some of the libpthread probes are
2199 provided only for i*86 and x86_64.
2200 Implemented by Roland McGrath and Rayson Ho.
2201
2202 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2203
2204 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2205 and Will Schmidt.
2206
2207 * More optimized functions for SPARC. Implemented by David S. Miller.
2208
2209 * Improved support for cross-compilation, especially bootstrap builds
2210 without a previously built glibc.
2211
2212 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2213 Chris Metcalf from Tilera.
2214
2215 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2216 now supported for ARM processors.
2217
2218 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2219 as the name of the dynamic linker, to distinguish it from the
2220 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2221
2222 * Support for CRIS has been removed from ports.
2223
2224 * A new class of installed header has been documented for low-level
2225 platform-specific functionality. PowerPC added the first instance with a
2226 function to provide time base register access. Contributed by Tulio
2227 Magno Quites Machado Filho.
2228
2229 * ix86 configurations now install header files that are consistent with
2230 what x86-64 configurations install. These same header files can be used
2231 for -m32, -m64, or -mx32 builds.
2232 Contributed by H.J. Lu.
2233
2234 * Math library bug fixes. A thorough audit of all open math library bugs was
2235 conducted by Joseph Myers. Significant progress was made on many math
2236 library bugs resulting in more accurate exceptions and function results.
2237 Many thanks to all those that contributed including Andreas Jaeger for his
2238 patch review and work on the x87 trigonometric instruction issues.
2239
2240 * Timezone data is no longer installed. Timezone-related binaries and scripts
2241 will continue to be installed. Users should obtain their timezone data from
2242 their distribution provider or from the tzdata package at
2243 <ftp://munnari.oz.au/pub/>.
2244 \f
2245 Version 2.15
2246
2247 * The following bugs are resolved with this release:
2248
2249 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2250 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2251 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2252 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2253 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2254 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2255 13484, 13506, 13515, 13523, 13524, 13538, 13540
2256
2257 * New program pldd to list loaded object of a process
2258 Implemented by Ulrich Drepper.
2259
2260 * Add nss_db support back to glibc. No more dependency on Berkeley db
2261 and support for initgroups lookups.
2262 Implemented by Ulrich Drepper.
2263
2264 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2265 Contributed by HJ Lu.
2266
2267 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2268 Contributed by HJ Lu.
2269
2270 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2271 on x86-32 and x86-64.
2272 Contributed by Liubov Dmitrieva.
2273
2274 * Optimized strchr and strrchr for SSE on x86-32.
2275 Contributed by Liubov Dmitrieva.
2276
2277 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2278 for x86-64 and x86-32.
2279 Contributed by Liubov Dmitrieva.
2280
2281 * New interfaces: scandirat, scandirat64
2282 Implemented by Ulrich Drepper.
2283
2284 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2285 Implemented by Ulrich Drepper.
2286
2287 * nscd now also caches the netgroup database.
2288 Implemented by Ulrich Drepper.
2289
2290 * Integrate libm with gcc's -ffinite-math-only option.
2291 Implemented by Ulrich Drepper.
2292
2293 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2294 to math functions. Implemented by Ulrich Drepper.
2295
2296 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2297 Implemented by Ulrich Drepper.
2298
2299 * New Linux interfaces: process_vm_readv, process_vm_writev
2300
2301 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2302 Implemented by Ulrich Drepper.
2303
2304 * Optimized nearbyint and strcasecmp for PPC.
2305 Implemented by Adhemerval Zanella.
2306
2307 * New locales: bho_IN, unm_US, es_CU, ta_LK
2308 \f
2309 Version 2.14
2310
2311 * The following bugs are resolved with this release:
2312
2313 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2314 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2315 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2316 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2317 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2318 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2319 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2320 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2321 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2322 12795, 12811, 12813, 12814, 12841
2323
2324 * The RPC implementation in libc is obsoleted. Old programs keep working
2325 but new programs cannot be linked with the routines in libc anymore.
2326 Programs in need of RPC functionality must be linked against TI-RPC.
2327 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2328
2329 Visible changes of this change include (obviously) the inability to link
2330 programs using RPC functions without referencing the TI-RPC library and the
2331 removal of the RPC headers from the glibc headers.
2332 Implemented by Ulrich Drepper.
2333
2334 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2335 syncfs, setns, sendmmsg
2336
2337 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2338 yue_HK, lij_IT, mhr_RU
2339
2340 * New encodings: CP770, CP771, CP772, CP773, CP774
2341
2342 * New program sotruss to trace calls through PLTs
2343 Implemented by Ulrich Drepper.
2344
2345 * The malloc hook implementation is marked deprecated and will be removed
2346 from the default implementation in the next version. The design never
2347 worked ever since the introduction of threads. Even programs which do
2348 not create threads themselves can use multiple threads created internally.
2349 \f
2350 Version 2.13
2351
2352 * The following bugs are resolved with this release:
2353
2354 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2355 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2356 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2357 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2358 12378, 12394, 12397
2359
2360 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2361
2362 * POWER7 optimizations: memset, memcmp, strncmp
2363
2364 * New optimized string functions for x86-64: strnlen (SSE2),
2365 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2366 Implemented by Ulrich Drepper.
2367 \f
2368 Version 2.12
2369
2370 * The following bugs are resolved with this release:
2371
2372 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2373 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2374 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2375 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2376 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2377 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2378 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2379 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2380
2381 * New interfaces: pthread_getname_np, pthread_setname_np
2382
2383 * New Linux interface: recvmmsg
2384
2385 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2386
2387 * The dynamic linker now recognizes supported ABI versions from the
2388 EI_ABIVERSION field in the ELF header.
2389 Implemented by Ulrich Drepper.
2390
2391 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2392 will not be used to fill in password fields in the passwd.byname replies.
2393 Instead it is used to synthesize the shadow.byname table, should it be
2394 missing. This is a useful mode in some installations involving Solaris.
2395 Implemented by Ulrich Drepper.
2396
2397 * New locales: kok_IN, sq_MK, cv_RU
2398 \f
2399 Version 2.11
2400
2401 * The following bugs are resolved with this release:
2402
2403 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2404 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2405 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2406 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2407 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2408 10789, 10817, 10823, 10840, 10847
2409
2410 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2411 mkostemps64
2412 Implemented by Ulrich Drepper.
2413
2414 * Checking version of longjmp added that fails if an uninitialized stack
2415 frame would be created. Implemented by Ulrich Drepper.
2416
2417 * STT_GNU_IFUNC is now supported in static executables.
2418 Implemented by H.J. Lu.
2419
2420 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2421
2422 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2423 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2424 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2425 Contributed by H.J. Lu.
2426
2427 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2428 Implemented by Ulrich Drepper.
2429
2430 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2431 strstr, strcasestr.
2432 Contributed by H.J. Lu.
2433
2434 * Support for fma instruction in AVX on x86-64.
2435 Implemented by H.J. Lu and Ulrich Drepper.
2436
2437 * AVX support in x86-64 auditing support in ld.so.
2438 Implemented by H.J. Lu.
2439
2440 * STB_GNU_UNIQUE support added.
2441 Implemented by Ulrich Drepper.
2442
2443 * Implement second fallback mode for optimized DNS lookup for even more
2444 broken environments. If two requests from the same port are not
2445 handled correctly close the socket and open a new one before sending
2446 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2447 can be used to select this mode right away, instead of rediscovering the
2448 necessity is every process again.
2449 Implemented by Ulrich Drepper.
2450
2451 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2452 Implemented by Adam Tkac.
2453
2454 * Optimized iconv conversions for S390x.
2455 Implemented by Andreas Krebbel.
2456
2457 * Using condvars with PI mutexes is now more efficient due to kernel
2458 support for requeueing to PI futexes. NPTL support added for x86-64.
2459 Implemented by Ulrich Drepper.
2460
2461 * New locales: ps_AF, my_MM
2462 \f
2463 Version 2.10
2464
2465 * The following bugs are resolved with this release:
2466
2467 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2468 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2469 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2470 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2471 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2472
2473 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2474 Implemented by Ulrich Drepper.
2475
2476 * New Linux interfaces: accept4, fallocate, fallocate64.
2477 Implemented by Ulrich Drepper.
2478
2479 * Correct declarations of string function when used in C++ code. This
2480 could lead to compile errors for invalid C++ code.
2481
2482 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2483 now in POSIX.
2484
2485 * New POSIX 2008 interface: psiginfo
2486 Implemented by Ulrich Drepper.
2487
2488 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2489 Implemented by Ulrich Drepper.
2490
2491 * Support for selecting between multiple function definitions at runtime
2492 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2493
2494 * The libcrypt library can now use the hash function implementations in
2495 NSS. Implemented by Ulrich Drepper.
2496
2497 * The malloc implementation can be compiled to be less memory efficient
2498 but higher performing in multi-threaded programs.
2499 Implemented by Ulrich Drepper.
2500
2501 * New locales: nan_TW@latin, ks_IN
2502
2503 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2504 Implemented by Ulrich Drepper.
2505
2506 * Extended printf hook support. It is possible to use user-defined types
2507 and extend existing format specifiers.
2508 Implemented by Ulrich Drepper.
2509
2510 * Handling for group shadow files has been added.
2511 Implemented by Ulrich Drepper.
2512
2513 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2514 servers (the case, e.g., for some people using the built-in DNS
2515 server in ADSL modems/routers). There is a once-per-process timeout
2516 in case of a broken server. To avoid it, users can run nscd or put
2517 'options single-request' in /etc/resolv.conf.
2518 Implemented by Ulrich Drepper.
2519 \f
2520 Version 2.9
2521
2522 * The following bugs are resolved with this release:
2523
2524 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2525 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2526 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2527 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2528 6968, 6974, 6980, 6995, 7008, 7009, 7029
2529
2530 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2531 up at the same time. Implemented by Ulrich Drepper.
2532
2533 * TLS descriptors for LD and GD on x86 and x86-64.
2534 Implemented by Alexandre Oliva.
2535
2536 * getaddrinfo now handles DCCP and UDPlite.
2537 Implemented by Ulrich Drepper.
2538
2539 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2540 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2541 Implemented by Ulrich Drepper.
2542
2543 * New implementation of memmem, strstr, and strcasestr which is O(n).
2544 Implemented by Eric Blake.
2545
2546 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2547
2548 * Implement "e" option for popen to open file descriptor with the
2549 close-on-exec flag set. Implemented by Ulrich Drepper.
2550
2551 * Implement "b" mode for fmemopen. In this mode writes writes don't
2552 implicitly add a NUL byte and seeks from the end of the buffer really
2553 use the buffer end, not the string length as the basis.
2554 Implemented by Ulrich Drepper.
2555
2556 * Many functions, exported and internal, now atomically set the close-on-exec
2557 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2558
2559 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2560 Sinhala)
2561 Implemented by Pravin Satpute.
2562
2563 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2564 \f
2565 Version 2.8
2566
2567 * The following bugs are resolved with this release:
2568
2569 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2570 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2571 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2572 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2573 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2574 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2575 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2576
2577 * New locales: bo_CN, bo_IN, shs_CA.
2578
2579 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2580
2581 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2582 Implemented by Pravin Satpute.
2583
2584 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2585
2586 * nscd caches now all timeouts for DNS entries
2587 Implemented by Ulrich Drepper.
2588
2589 * nscd is more efficient and wakes up less often.
2590 Implemented by Ulrich Drepper.
2591
2592 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2593 vdprintf, and obstack_vprintf.
2594 Implemented by Jakub Jelinek.
2595
2596 * Faster memset for x86-64.
2597 Implemented by Harsha Jagasia and H.J. Lu.
2598
2599 * Faster memcpy on x86.
2600 Implemented by Ulrich Drepper.
2601
2602 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2603 Implemented by Ulrich Drepper.
2604
2605 * Faster sqrt and sqrtf implemention for some PPC variants.
2606 Implemented by Steven Munroe.
2607 \f
2608 Version 2.7
2609
2610 * The following bugs are resolved with this release:
2611
2612 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2613 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2614 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2615 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2616 5071, 5103, 5104, 5112, 5113, 5184, 5186
2617
2618 * More checking functions: fread, fread_unlocked, open*, mq_open.
2619 Implemented by Jakub Jelinek and Ulrich Drepper.
2620
2621 * Extend fortification to C++. Implemented by Jakub Jelinek.
2622
2623 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2624 by not recognizing 'a' as a modifier when those specs are requested.
2625 Implemented by Jakub Jelinek.
2626
2627 * PPC optimizations to math and string functions.
2628 Implemented by Steven Munroe.
2629
2630 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2631 options to be passed. Implemented by Ulrich Drepper.
2632
2633 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2634
2635 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2636 Implemented by Ulrich Drepper.
2637
2638 * Handle private futexes in the NPTL implementation.
2639 Implemented by Jakub Jelinek and Ulrich Drepper.
2640
2641 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2642 Implemented by Roland McGrath and Ulrich Drepper.
2643
2644 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2645
2646 * SHA-256 and SHA-512 based password encryption.
2647 Implemented by Ulrich Drepper.
2648
2649 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2650 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2651 yo_NG.
2652
2653 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2654 Implemented by Ulrich Drepper.
2655 \f
2656 Version 2.6
2657
2658 * The following bugs are resolved with this release:
2659
2660 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2661 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2662 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2663 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2664 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2665 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2666 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2667 4702, 4858
2668
2669 * New Linux interfaces: epoll_pwait, sched_getcpu.
2670
2671 * New generic interfaces: strerror_l.
2672
2673 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2674 \f
2675 Version 2.5
2676
2677 * The following bugs are resolved with this release:
2678
2679 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2680 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2681 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2682 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2683 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2684 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2685 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2686 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2687 3225, 3251, 3252, 3253, 3273
2688
2689 * For Linux, the sorting of addresses returned by getaddrinfo now also
2690 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2691 Implemented by Ulrich Drepper.
2692
2693 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2694 Implemented by Ulrich Drepper.
2695
2696 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2697
2698 * New iconv module for MIK. Contributed by Alexander Shopov.
2699
2700 * For sites with broken group and/or passwd database, the auto-propagate
2701 option of nscd can prevent creating ID lookup entries from the results
2702 of a name lookup and vice versa. This usually is no problem but some
2703 site might have problems with the default behavior.
2704 Implemented by Ulrich Drepper.
2705
2706 * Iterating over entire database in NIS can be slow. With the
2707 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2708 to trade time for memory. The entire database will be read at once.
2709 Implemented by Ulrich Drepper.
2710
2711 * The interfaces introduced in RFC 3542 have been implemented by
2712 Ulrich Drepper.
2713
2714 * Support for the new ELF hash table format was added by Ulrich Drepper.
2715
2716 * Support for priority inheritance mutexes added by Jakub Jelinek and
2717 Ulrich Drepper.
2718
2719 * Support for priority protected mutexes added by Jakub Jelinek.
2720
2721 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2722 or_IN, csb_PL, fy_NL, sr_ME.
2723 \f
2724 Version 2.4
2725
2726 * More overflow detection functions.
2727
2728 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2729 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2730
2731 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2732 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2733 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2734 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2735 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2736 by Masahide Washizawa.
2737
2738 * It is now possible to install audit libraries for the dynamic linker using
2739 LD_AUDIT. Implemented by Ulrich Drepper.
2740
2741 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2742 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2743 For a libc and libpthread that works well on Linux 2.4 kernels, we
2744 recommend using the stable 2.3 branch.
2745
2746 * The new function open_wmemstream acts like open_memstream,
2747 but uses a wchar_t wide character buffer.
2748
2749 * The new function ppoll is an improved version of poll, similar to pselect.
2750
2751 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2752 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2753 renameat, unlinkat, linkat, symlinkat, readlinkat.
2754
2755 * New Linux kernel system calls: unshare,
2756 inotify_init, inotify_add_watch, inotify_rm_watch.
2757
2758 * The euidaccess function is now also known by the alias eaccess,
2759 for compatibility with some other systems.
2760
2761 * Timezone data updated to 2006b version.
2762 \f
2763 Version 2.3.6
2764
2765 * The following bugs are resolved with this release:
2766
2767 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2768 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2769 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2770 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2771 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2772 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2773
2774 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2775
2776 * As of this release, GCC 4 can be used to compile the C Library.
2777
2778 * Timezone data updated to 2005m version.
2779 \f
2780 Version 2.3.5
2781
2782 * The following bugs are resolved with this release:
2783
2784 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2785 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2786 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2787 777, 787, 821, 822, 823, 825
2788
2789 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2790 \f
2791 Version 2.3.4
2792
2793 * Support for RFC 3678. Real implementations exist only for Linux so far.
2794 Implemented by Ulrich Drepper.
2795
2796 * nscd can now cache entries persistently. Expiring entries are reloaded.
2797 For speedups the cache can be shared in memory with client processes.
2798 Implemented by Ulrich Drepper.
2799
2800 * nscd can now perform SELinux checks.
2801 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2802
2803 * getaddrinfo queries are now cached. Canonical name lookup is performed
2804 efficiently.
2805 Implemented by Ulrich Drepper.
2806
2807 * The nothrow function attribute is used when headers are used by gcc when
2808 compiling C code. This can avoid significant amounts of exception
2809 handling data.
2810
2811 * The malloc functions perform more error checking and are stricter when
2812 it comes to reacting on errors. The default action is to terminate
2813 the process after showing an error message. Implemented by Ulrich Drepper.
2814
2815 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2816 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2817
2818 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2819 namespaces using the new function dlmopen(). This feature is of course,
2820 like most other dynamic loading functionality, not available in statically
2821 linked applications. Implemented by Ulrich Drepper.
2822
2823 * Low-overhead boundary checking variants of string and some stdio functions
2824 were added. These are to be used in conjunction with a gcc patch by
2825 Jakub Jelinek which adds calls to these functions if possible.
2826 Implemented by Jakub Jelinek and Ulrich Drepper.
2827
2828 * Old code for several operating systems and machine architectures that
2829 have not been in working condition in a long time have been removed from
2830 the main source tree maintained by the GNU C Library's maintainers.
2831 These files are now reside in the separate `ports' source module
2832 that is usable as an add-on when building the library.
2833 \f
2834 Version 2.3.3
2835
2836 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2837 interrogate the dynamic linker, compatible with the Solaris interface.
2838
2839 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2840 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2841
2842 * getifaddrs now uses the netlink interface on Linux to get its information.
2843 Implemented by Thorsten Kukuk.
2844
2845 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2846 Implemented by Ulrich Drepper.
2847
2848 * support for non-executable stacks on x86 has been added. Changes mostly
2849 by Roland McGrath.
2850
2851 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2852 and Ulrich Drepper.
2853
2854 * getaddrinfo now performs destination address selection according to
2855 RFC 3484.
2856 \f
2857 Version 2.3.2
2858
2859 * Thread-safe interfaces for many functions that access locale data
2860 were added in version 2.3, but these features were omitted from NEWS.
2861 Many functions have variants with an `_l' suffix that take a `locale_t'
2862 object as a parameter rather than consulting the current locale.
2863 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2864 create and maintain `locale_t' objects. Additionally, the new function
2865 `uselocale' sets "the current locale" (as used by functions not so
2866 parameterized) set for an individual thread. These features were added
2867 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2868
2869 * The functions getresuid, getresgid, setresuid, and setresgid, which
2870 have long been available on Linux, are now declared in <unistd.h>
2871 and are now also available on the Hurd.
2872
2873 * ELF thread-local storage support (TLS) now works on x86-64.
2874
2875 * The new dynamic string token $LIB is expanded in shared library names.
2876 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2877
2878 * Aldy Hernandez contributed complete software floating point support for
2879 PowerPC machines with no FPU.
2880
2881 * fexecve is implemented on Linux.
2882
2883 * The `btowc' function should work at least twice as fast due to
2884 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2885
2886 * With approriate thread add-ons cancelable functions are now implemented
2887 in libc.so as well. No need to call the function in libpthread. This
2888 change allowed to finally disable the incorrect and expensive handling
2889 of weak definition in ld.so.
2890
2891 * Yet more PLT entries in libc.so have been removed. We finally arrived
2892 at the bare minimum. Startup times improved appropriately.
2893
2894 * Support for the new Linux/x86 system call interface was added. The
2895 AT_SYSINFO auxiliary vector entry is recognized and handled.
2896 \f
2897 Version 2.3
2898
2899 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2900 charsets.
2901
2902 * iconv (the program and the interface) now accepts empty names (excluding
2903 options like //TRANSLIT) to mean "use charset of current locale".
2904
2905 * localedef can now transliterate characters in strings which are not in
2906 the provided charmap. The information from the input locale is used.
2907
2908 * Prelinking support was added for ELF targets. This requires additional
2909 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2910
2911 * Read-only stdio streams now use mmap to speed up operation by eliminating
2912 copying and buffer underflows. To use add 'm' to the mode string of
2913 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2914
2915 * The malloc functions were completely rewritten by Wolfram Gloger based
2916 on Doug Lea's malloc-2.7.0.c.
2917
2918 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2919 implementation of regex.
2920
2921 * Bruno Haible upgraded the iconv and locale implementation to support
2922 Unicode 3.2.
2923
2924 * Contents of the LC_* and LANG environment variables in the CEN style are
2925 not recognized anymore. It never was used. Change by Ulrich Drepper.
2926
2927 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2928 thread-local storage (TLS) ABI on some platforms.
2929 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2930
2931 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2932 EUC-JISX0213, and TSCII.
2933
2934 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2935 BSD-compatible interface for getting all network interface addresses.
2936 Implementation for IPv4 by Roland McGrath.
2937
2938 * Loading of locale data is faster due to the introduction of a locale
2939 archive. Implemented by Roland McGrath and Ulrich Drepper.
2940
2941 * Startup times are significantly reduced by not using exported functions
2942 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2943 and Ulrich Drepper.
2944
2945 * Steven Munroe contributed a port to PowerPC64/Linux.
2946 \f
2947 Version 2.2.6
2948
2949 * The Hurd now uses the GNU libio implementation of stdio rather than the
2950 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2951
2952 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2953 with the same behavior as the Linux system call of the same name.
2954 \f
2955 Version 2.2.5
2956
2957 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2958 128-bit long double format.
2959
2960 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2961 IBM1161, and IBM1162 charsets.
2962
2963 * Andreas Jaeger contributed a port to x86-64/Linux.
2964
2965 * Peter Bruin contributed a port to PowerPC/Hurd.
2966
2967 * libc functions using I/O streams now can handle wide-oriented streams
2968 as well.
2969
2970 * optimizations in the dynamic linker. Binaries created by recent binutils
2971 versions start up quicker due to reduced time spent on relocations.
2972
2973 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2974 \f
2975 Version 2.2.4
2976
2977 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2978 asinh, atanh, j0 for the 128-bit long double format.
2979
2980 * Bruno Haible updated all the code handling Unicode in some form to
2981 support Unicode 3.1.
2982
2983 * Speed of regex for single-byte locales is back to previous levels.
2984 Patch by Isamu Hasegawa.
2985
2986 * Alpha, SPARC, and IA-64 now also using floating stacks.
2987
2988 * Startup time of internationalized applications greatly improved through
2989 iconv cache. Use iconvconfig to generate the cache file.
2990 Contributed by Ulrich Drepper.
2991
2992 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2993 severe performance problems. Patch by David Mosberger.
2994
2995 * The Hurd port got a lot more functionality like AIO, various stdio
2996 extensions, etc. Mainly done by Roland McGrath.
2997
2998 * mtrace can now lookup symbols in shared libraries.
2999 \f
3000 Version 2.2.3
3001
3002 * Intel's IA-64 math library is largely integrated. It provides fast and
3003 accurate implementatations for most basic and standard math functions
3004 in float, double, and long double format.
3005
3006 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3007 and asin for the 96-bit long double format and asin, log, tan for the
3008 128-bit long double format.
3009
3010 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3011 The basic double functions exist today. Contributed by Abraham Ziv
3012 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3013 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3014
3015 * An asynchronous name lookup library was added. The interface is designed
3016 after POSIX AIO. The proposal was circulated beforehand to get comments.
3017 No negative ones came in. Implemented by Ulrich Drepper.
3018
3019 * Port to S390/64bit contributed by Martin Schwidefsky
3020 <schwidefsky@de.ibm.com>.
3021
3022 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3023 of functions for Linux/IA-64.
3024
3025 * The RPC code is now thread safe. Threads can now use the same service
3026 of different services at the same time. Patch by Eric Norum
3027 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3028
3029 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3030 family of functions for Linux/S390.
3031
3032 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3033 of functions for Linux/x86.
3034
3035 * Port to Linux/CRIS contributed by Axis Communications.
3036 \f
3037 Version 2.2.2
3038
3039 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3040 we can now check for namespace violations and missing declarations. The
3041 result is that almost all headers are now Unix-compliant (as defined in
3042 the upcoming XPG6). The negative side is that some programs might need
3043 corrections, too, if they depend on the incorrect form of the headers in
3044 previous versions which defined too many symbols and included too many
3045 other headers.
3046
3047 * regex now handles multibyte character sets correctly.
3048 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3049
3050 * iconv (the program) does now conform to the upcoming XPG6 and handles
3051 charmaps. Instead of the charset names the path of charmaps can be
3052 provided and the conversion happens based on this data.
3053 Contributed by Ulrich Drepper.
3054
3055 * The locale program now provides detailed information about the installed
3056 locales. While
3057
3058 locale -a
3059
3060 only lists the names of the supported locales
3061
3062 locale -a --verbose
3063
3064 provides details such as country, language, and codeset name.
3065 Contributed by Ulrich Drepper.
3066 \f
3067 Version 2.2.1
3068
3069 * The gencat program now parses the input file according to the charset
3070 selected by the LC_CTYPE category. This is important for stateful
3071 character sets. To make generating catalogs easier there is a way
3072 to overwrite the charset selected by the locale: before the first
3073 message or $ quote line the catalog can contain a line like
3074
3075 $ codeset=ISO-8859-2
3076
3077 to select the charset (ISO-8859-2 in this case).
3078
3079 Implemented by Shinya Hanataka and Ulrich Drepper.
3080
3081 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3082 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3083 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3084
3085 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3086 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3087
3088 * The LANGUAGE environment variable is now ignored unless the locale is
3089 changed from the default "C" locale.
3090
3091 * The usual bug fixes.
3092 \f
3093 Version 2.2
3094
3095 * Greg McGary added runtime support for bounds checking using gcc's
3096 new -fbounded-pointers option. ix86 target is complete. PowerPC
3097 is in progress.
3098
3099 * Thorsten Kukuk added secure mode support to nscd.
3100
3101 * The Berkeley db libraries have been removed.
3102
3103 Related, the nss_db module is now in a separate package since it
3104 obviously requires a database library being available.
3105
3106 * Wide character I/O streams implemented by Ulrich Drepper.
3107
3108 * Functions from the extended socket API added by Ulrich Drepper.
3109
3110 * Functions feenableexcept and fedisableexcept to control the
3111 behaviour of individual exceptions have been added by Andreas Jaeger.
3112
3113 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3114
3115 * The resolver code has been updated from bind 8.2.3-T5B which supports
3116 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3117 and Mark Kettenis.
3118
3119 This change could in some situations effect backward compatibility. Since
3120 now `_res' is a thread-local instead of a global variable, modifying it
3121 in one thread does not have any effect in other threads.
3122
3123 The resolver library was also extended to allow IPv6 as the transport
3124 protocol for the requests. This work was done by Stig Venaas.
3125
3126 * Compatibility code for K&R C compilers has been removed from the
3127 header files. A ISO C compiler is needed to use the library
3128 (conforming to either C89 or C99 standard).
3129
3130 * Complete rewrite of the localedef program to support multibyte character
3131 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3132 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3133 Implemented by Ulrich Drepper.
3134
3135 Bruno Haible significantly improved the generation and use of the data
3136 structures for the wide character tables.
3137
3138 * Plural handling in gettext implemented by Ulrich Drepper.
3139
3140 * The utmp daemon has been removed.
3141
3142 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3143
3144 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3145 and Yutaka Niibe.
3146
3147 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3148
3149 * POSIX barriers implemented by Kaz Kylheku.
3150
3151 * POSIX spawn function family implemented by Ulrich Drepper.
3152
3153 * POSIX spinlocks are now available.
3154
3155 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3156
3157 * the configure option --enable-kernel=X.Y.Z allows to strip out
3158 compatibility for kernel versions before X.Y.Z. This is currently only
3159 implemented for Linux.
3160
3161 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3162 field sin6_scope_id. This means that all programs using IPv6 should be
3163 recompiled. Don't expect binary compatibility with previous glibc
3164 versions.
3165
3166 * various conversion modules for IBM character sets contributed by
3167 Masahide Washizawa.
3168
3169 * IA-64 port by Jes Sorensen and HJ Lu.
3170 \f
3171 Version 2.1.3
3172
3173 * bug fixes
3174
3175 \f
3176 Version 2.1.2
3177
3178 * bug fixes
3179
3180 \f
3181 Version 2.1.1
3182
3183 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3184
3185 * New xdr functions are added; some rpc functions are now 64bit clean.
3186
3187 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3188
3189 * Fixed known incompatibilities with glibc 2.0.
3190
3191 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3192
3193 * Optimized a number of functions (especially the ELF dynamic loader).
3194
3195 * Update timezone data files.
3196
3197 * lots of charmaps corrections
3198
3199 * some new locale definitions and charmaps
3200
3201 \f
3202 Version 2.1
3203
3204 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3205 conform to POSIX member type requirements. Symbol versions have been
3206 adjusted as needed within the library, and for direct use by applications,
3207 but there is potential for problems if third-party libraries use
3208 struct timeval as part of their interface. This does not present
3209 a problem for X and other "essential" system libraries.
3210
3211 * An additional locale model to support C++ Standard Library locale
3212 model and probably more was implemented by Ulrich Drepper.
3213
3214 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3215 symbol level.
3216
3217 * Miles Bader provided the `argp' function family to support hierachical
3218 command line argument parsing, layered on top of getopt.
3219
3220 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3221
3222 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3223 numbers.
3224
3225 * scanf recognizes the %a and %A format for scanning floating point numbers.
3226
3227 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3228 information and interfaces for the available integer types.
3229
3230 * about 130 new math functions were added to implement the ISO C9x math
3231 library.
3232
3233 * the new header <complex.h> contains definitions of the complex math
3234 functions from ISO C 9X.
3235
3236 * the new header <tgmath.h> defines generic macros to use complex or
3237 real valued functions.
3238
3239 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3240
3241 * Andreas Jaeger provided a test suite for the math library.
3242
3243 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3244
3245 * Ulrich Drepper added character set conversion functions (iconv).
3246
3247 * Optimized string functions have been added.
3248
3249 * The localedata addon is now part of glibc.
3250
3251 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3252
3253 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3254 daemon for NSS (nscd).
3255
3256 Missing a better place here are some numbers on improvements. Under
3257 Linux 2.1.125 un-tar-ing the kernel sources takes
3258
3259 user system wall
3260
3261 using local files 12.19s 6.88s 22.91s
3262
3263 using NIS 13.92s 8.91s 26.34s
3264
3265 using NIS & nscd 10.37s 7.34s 25.30s
3266
3267 using NIS+ 27.57s 30.37s 640.46s
3268
3269 using NIS+ & nscd 10.25s 7.83s 26.51s
3270
3271 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3272
3273 Keep in mind that non-namelookup related operations dominate above times.
3274 It was just a common complain that using NIS+ unpacking the kernel is
3275 horribly slow.
3276
3277 [1] The old nscd implementation is not available anymore since it was
3278 distributed with glibc up to version 2.0.98 and thus is now replaced.
3279
3280 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3281
3282 * Mark Kettenis provided a Hesiod NSS module.
3283
3284 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3285 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3286
3287 * rcmd can now handle netgroups (Dick Streefland).
3288
3289 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3290 Bambrough.
3291
3292 * Support for the IPv6 protocol has been added to the socket API, as per the
3293 latest draft standards.
3294
3295 * Support for Linux 2.2 has been added.
3296
3297 * Interface changes relative to the latest 2.0.x release:
3298 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3299 addseverity NEW: Unix98
3300 alphasort64 NEW: LFS
3301 argp_err_exit_status NEW: argp, GNU ext
3302 argp_error NEW: argp, GNU ext
3303 argp_failure NEW: argp, GNU ext
3304 argp_help NEW: argp, GNU ext
3305 argp_parse NEW: argp, GNU ext
3306 argp_program_bug_address NEW: argp, GNU ext
3307 argp_program_version NEW: argp, GNU ext
3308 argp_program_version_hook NEW: argp, GNU ext
3309 argp_state_help NEW: argp, GNU ext
3310 argp_usage NEW: argp, GNU ext
3311 authdes_create NEW: Secure RPC
3312 authdes_getucred NEW: Secure RPC
3313 authdes_pk_create NEW: Secure RPC
3314 backtrace NEW: GNU ext.
3315 backtrace_symbols NEW: GNU ext.
3316 backtrace_symbols_fd NEW: GNU ext.
3317 cacos NEW: ISO C 9x
3318 cacosf NEW: ISO C 9x
3319 cacosh NEW: ISO C 9x
3320 cacoshf NEW: ISO C 9x
3321 cacoshl NEW: ISO C 9x
3322 cacosl NEW: ISO C 9x
3323 capget NEW: kernel
3324 capset NEW: kernel
3325 carg NEW: ISO C 9x
3326 cargf NEW: ISO C 9x
3327 cargl NEW: ISO C 9x
3328 casin NEW: ISO C 9x
3329 casinf NEW: ISO C 9x
3330 casinh NEW: ISO C 9x
3331 casinhf NEW: ISO C 9x
3332 casinhl NEW: ISO C 9x
3333 casinl NEW: ISO C 9x
3334 catan NEW: ISO C 9x
3335 catanf NEW: ISO C 9x
3336 catanh NEW: ISO C 9x
3337 catanhf NEW: ISO C 9x
3338 catanhl NEW: ISO C 9x
3339 catanl NEW: ISO C 9x
3340 cbc_crypt NEW: Secure RPC
3341 ccos NEW: ISO C 9x
3342 ccosf NEW: ISO C 9x
3343 ccosh NEW: ISO C 9x
3344 ccoshf NEW: ISO C 9x
3345 ccoshl NEW: ISO C 9x
3346 ccosl NEW: ISO C 9x
3347 cexp NEW: ISO C 9x
3348 cexpf NEW: ISO C 9x
3349 cexpl NEW: ISO C 9x
3350 cimag NEW: ISO C 9x
3351 cimagf NEW: ISO C 9x
3352 cimagl NEW: ISO C 9x
3353 clearerr_locked REMOVED
3354 clntunix_create NEW: sunrpc ext
3355 clog NEW: ISO C 9x
3356 clog10 NEW: ISO C 9x
3357 clog10f NEW: ISO C 9x
3358 clog10l NEW: ISO C 9x
3359 clogf NEW: ISO C 9x
3360 clogl NEW: ISO C 9x
3361 conj NEW: ISO C 9x
3362 conjf NEW: ISO C 9x
3363 conjl NEW: ISO C 9x
3364 cpow NEW: ISO C 9x
3365 cpowf NEW: ISO C 9x
3366 cpowl NEW: ISO C 9x
3367 cproj NEW: ISO C 9x
3368 cprojf NEW: ISO C 9x
3369 cprojl NEW: ISO C 9x
3370 creal NEW: ISO C 9x
3371 crealf NEW: ISO C 9x
3372 creall NEW: ISO C 9x
3373 creat64 NEW: LFS
3374 csin NEW: ISO C 9x
3375 csinf NEW: ISO C 9x
3376 csinh NEW: ISO C 9x
3377 csinhf NEW: ISO C 9x
3378 csinhl NEW: ISO C 9x
3379 csinl NEW: ISO C 9x
3380 csqrt NEW: ISO C 9x
3381 csqrtf NEW: ISO C 9x
3382 csqrtl NEW: ISO C 9x
3383 ctan NEW: ISO C 9x
3384 ctanf NEW: ISO C 9x
3385 ctanh NEW: ISO C 9x
3386 ctanhf NEW: ISO C 9x
3387 ctanhl NEW: ISO C 9x
3388 ctanl NEW: ISO C 9x
3389 des_setparity NEW: Secure RPC
3390 ecb_crypt NEW: Secure RPC
3391 endutxent NEW: Unix98
3392 exp10 NEW: ISO C 9x
3393 exp10f NEW: ISO C 9x
3394 exp10l NEW: ISO C 9x
3395 exp2 NEW: ISO C 9x
3396 exp2f NEW: ISO C 9x
3397 exp2l NEW: ISO C 9x
3398 fattach NEW: STREAMS
3399 fdetach NEW: STREAMS
3400 fdim NEW: ISO C 9x
3401 fdimf NEW: ISO C 9x
3402 fdiml NEW: ISO C 9x
3403 feclearexcept NEW: ISO C 9x
3404 fegetenv NEW: ISO C 9x
3405 fegetexceptflag NEW: ISO C 9x
3406 fegetround NEW: ISO C 9x
3407 feholdexcept NEW: ISO C 9x
3408 feof_locked REMOVED
3409 feraiseexcept NEW: ISO C 9x
3410 ferror_locked REMOVED
3411 fesetenv NEW: ISO C 9x
3412 fesetexceptflag NEW: ISO C 9x
3413 fesetround NEW: ISO C 9x
3414 fetestexcept NEW: ISO C 9x
3415 feupdateenv NEW: ISO C 9x
3416 fflush_locked REMOVED
3417 ffsl NEW: GNU ext.
3418 ffsll NEW: GNU ext.
3419 fgetpos64 NEW: LFS
3420 fgets_unlocked NEW: GNU ext.
3421 fileno_locked REMOVED
3422 fma NEW: ISO C 9x
3423 fmaf NEW: ISO C 9x
3424 fmal NEW: ISO C 9x
3425 fmax NEW: ISO C 9x
3426 fmaxf NEW: ISO C 9x
3427 fmaxl NEW: ISO C 9x
3428 fmin NEW: ISO C 9x
3429 fminf NEW: ISO C 9x
3430 fminl NEW: ISO C 9x
3431 fmtmsg NEW: Unix98
3432 fopen64 NEW: LFS
3433 fputc_locked REMOVED
3434 fputs_unlocked NEW: GNU ext.
3435 fread_unlocked NEW: GNU ext.
3436 freopen64 NEW: LFS
3437 fseeko NEW: Unix98
3438 fsetpos64 NEW: LFS
3439 fstatfs64 NEW: LFS
3440 fstatvfs NEW: Unix98
3441 fstatvfs64 NEW: LFS
3442 ftello NEW: Unix98
3443 ftello64 NEW: LFS
3444 ftruncate64 NEW: LFS
3445 ftw64 NEW: LFS
3446 fwrite_unlocked NEW: GNU ext.
3447 gai_strerror NEW: IPv6
3448 gamma_r REMOVED
3449 gammaf_r REMOVED
3450 gammal_r REMOVED
3451 getchar_locked REMOVED
3452 getdate NEW: Unix98
3453 getdate_err NEW: Unix98
3454 getdate_r NEW: GNU ext.
3455 getmsg NEW: STREAMS
3456 getnameinfo NEW: IPv6
3457 getnetname NEW: Secure RPC
3458 getpmsg NEW: STREAMS
3459 getpt NEW: Unix98 PTY
3460 getrlimit64 NEW: LFS
3461 getutxent NEW: Unix98
3462 getutxid NEW: Unix98
3463 getutxline NEW: Unix98
3464 glob64 NEW: GNU ext.
3465 globfree64 NEW: GNU ext.
3466 gnu_get_libc_release NEW: GNU ext.
3467 gnu_get_libc_version NEW: GNU ext.
3468 grantpt NEW: Unix98 PTY
3469 host2netname NEW: Secure RPC
3470 iconv NEW: iconv
3471 iconv_close NEW: iconv
3472 iconv_open NEW: iconv
3473 if_freenameindex NEW: IPv6
3474 if_indextoname NEW: IPv6
3475 if_nameindex NEW: IPv6
3476 if_nametoindex NEW: IPv6
3477 in6addr_any NEW: IPv6
3478 in6addr_loopback NEW: IPv6
3479 inet6_isipv4mapped NEW: IPv6
3480 isastream NEW: STREAMS
3481 iswblank NEW: GNU ext.
3482 key_decryptsession NEW: Secure RPC
3483 key_decryptsession_pk NEW: Secure RPC
3484 key_encryptsession NEW: Secure RPC
3485 key_encryptsession_pk NEW: Secure RPC
3486 key_gendes NEW: Secure RPC
3487 key_get_conv NEW: Secure RPC
3488 key_secretkey_is_set NEW: Secure RPC
3489 key_setnet NEW: Secure RPC
3490 key_setsecret NEW: Secure RPC
3491 llrint NEW: ISO C 9x
3492 llrintf NEW: ISO C 9x
3493 llrintl NEW: ISO C 9x
3494 llround NEW: ISO C 9x
3495 llroundf NEW: ISO C 9x
3496 llroundl NEW: ISO C 9x
3497 log2 NEW: ISO C 9x
3498 log2f NEW: ISO C 9x
3499 log2l NEW: ISO C 9x
3500 lrint NEW: ISO C 9x
3501 lrintf NEW: ISO C 9x
3502 lrintl NEW: ISO C 9x
3503 lround NEW: ISO C 9x
3504 lroundf NEW: ISO C 9x
3505 lroundl NEW: ISO C 9x
3506 lseek64 NEW: LFS
3507 makecontext NEW: Unix98
3508 mempcpy NEW: GNU ext.
3509 mmap64 NEW: LFS
3510 moncontrol REMOVED
3511 modify_ldt NEW: kernel
3512 nan NEW: ISO C 9x
3513 nanf NEW: ISO C 9x
3514 nanl NEW: ISO C 9x
3515 nearbyint NEW: ISO C 9x
3516 nearbyintf NEW: ISO C 9x
3517 nearbyintl NEW: ISO C 9x
3518 netname2host NEW: Secure RPC
3519 netname2user NEW: Secure RPC
3520 nexttoward NEW: ISO C 9x
3521 nexttowardf NEW: ISO C 9x
3522 nexttowardl NEW: ISO C 9x
3523 nftw NEW: Unix98
3524 nftw64 NEW: LFS
3525 open64 NEW: LFS
3526 passwd2des NEW: Secure RPC
3527 pow10 NEW: GNU ext.
3528 pow10f NEW: GNU ext.
3529 pow10l NEW: GNU ext.
3530 pread NEW: Unix98
3531 pread64 NEW: LFS
3532 printf_size NEW: GNU ext.
3533 printf_size_info NEW: GNU ext.
3534 profil_counter REMOVED
3535 pthread_mutexattr_getkind_np REPLACED
3536 pthread_mutexattr_setkind_np REPLACED
3537 ptsname NEW: Unix98 PTY
3538 ptsname_r NEW: Unix98 PTY
3539 putc_locked REMOVED
3540 putchar_locked REMOVED
3541 putgrent NEW: GNU ext.
3542 putmsg NEW: STREAMS
3543 putpmsg NEW: STREAMS
3544 pututxline NEW: Unix98
3545 pwrite NEW: Unix98
3546 pwrite64 NEW: LFS
3547 readdir64 NEW: LFS
3548 readdir64_r NEW: LFS
3549 remquo NEW: ISO C 9x
3550 remquof NEW: ISO C 9x
3551 remquol NEW: ISO C 9x
3552 round NEW: ISO C 9x
3553 roundf NEW: ISO C 9x
3554 roundl NEW: ISO C 9x
3555 rtime NEW: GNU ext.
3556 scalbln NEW: ISO C 9x
3557 scalblnf NEW: ISO C 9x
3558 scalblnl NEW: ISO C 9x
3559 scandir64 NEW: LFS
3560 sendfile NEW: kernel
3561 setcontext NEW: Unix98
3562 setrlimit64 NEW: LFS
3563 setutxent NEW: Unix98
3564 sighold NEW: Unix98
3565 sigignore NEW: Unix98
3566 sigqueue NEW: POSIX.1b
3567 sigrelse NEW: Unix98
3568 sigset NEW: POSIX.1b
3569 sigtimedwait NEW: POSIX.1b
3570 sigwaitinfo NEW: POSIX.1b
3571 sincos NEW: GNU ext.
3572 sincosf NEW: GNU ext.
3573 sincosl NEW: GNU ext.
3574 statfs64 NEW: LFS
3575 statvfs NEW: Unix98
3576 statvfs64 NEW: LFS
3577 strcasestr NEW: GNU ext.
3578 strtoimax NEW: ISO C 9x
3579 strtoumax NEW: ISO C 9x
3580 strverscmp NEW: GNU ext.
3581 svcauthdes_stats NEW: Secure RPC
3582 svcunix_create NEW: sunrpc etx
3583 svcunixfd_create NEW: sunrpc ext
3584 swapcontext NEW: Unix98
3585 tcgetsid NEW: Unix98 PTY
3586 tdestroy NEW: GNU ext.
3587 tgamma NEW: ISO C 9x
3588 tgammaf NEW: ISO C 9x
3589 tgammal NEW: ISO C 9x
3590 tmpfile64 NEW: LFS
3591 trunc NEW: ISO C 9x
3592 truncate64 NEW: LFS
3593 truncf NEW: ISO C 9x
3594 truncl NEW: ISO C 9x
3595 umount2 NEW: kernel
3596 unlockpt NEW: Unix98 PTY
3597 updwtmpx NEW: Unix98
3598 user2netname NEW: Secure RPC
3599 utmpxname NEW: Unix98
3600 versionsort NEW: GNU ext.
3601 versionsort64 NEW: GNU ext.
3602 waitid NEW: Unix98
3603 wcscasecmp NEW: GNU ext.
3604 wcsncasecmp NEW: GNU ext.
3605 wcsnlen NEW: GNU ext.
3606 wcstoimax NEW: ISO C 9x
3607 wcstoll NEW: ISO C 9x
3608 wcstoull NEW: ISO C 9x
3609 wcstoumax NEW: ISO C 9x
3610 wcswcs NEW: Unix98
3611 wordexp NEW: POSIX.2
3612 wordfree NEW: POSIX.2
3613 write_profiling REMOVED
3614 xdecrypt NEW: Secure RPC
3615 xdr_authdes_cred NEW: Secure RPC
3616 xdr_authdes_verf NEW: Secure RPC
3617 xdr_cryptkeyarg NEW: Secure RPC
3618 xdr_cryptkeyarg2 NEW: Secure RPC
3619 xdr_cryptkeyres NEW: Secure RPC
3620 xdr_getcredres NEW: Secure RPC
3621 xdr_key_netstarg NEW: Secure RPC
3622 xdr_key_netstres NEW: Secure RPC
3623 xdr_keybuf NEW: Secure RPC
3624 xdr_keystatus NEW: Secure RPC
3625 xdr_netnamestr NEW: Secure RPC
3626 xdr_sizeof NEW: Secure RPC
3627 xdr_unixcred NEW: sunrpc ext
3628 xencrypt NEW: Secure RPC
3629 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3630 \f
3631 Version 2.0.6
3632
3633 * more bug fixes
3634
3635 \f
3636 Version 2.0.5
3637
3638 * more bug fixes
3639
3640 * inet_ntoa is thread-safe
3641
3642 * updwtmp is moved from libutil to libc
3643
3644 * rewrite of cbrt function
3645
3646 * update of timezone data
3647 \f
3648 Version 2.0.4
3649
3650 * more bug fixes
3651 \f
3652 Version 2.0.3
3653
3654 * more bug fixes
3655 \f
3656 Version 2.0.2
3657
3658 * more bug fixes
3659
3660 * add atoll function
3661
3662 * fix complex problems in Berkeley DB code
3663
3664 * fix math functions
3665 \f
3666 Version 2.0.1
3667
3668 * fixed lots of header problems (especially Linux/GNU specific)
3669
3670 * dynamic loader preserves all registers
3671
3672 * Roland McGrath provided support for handling of auxiliary objects in
3673 the ELF dynamic loader.
3674
3675 * support for parallel builds is improved
3676 \f
3677 Version 2.0
3678
3679 * GNU extensions are no longer declared by default. To enable them you
3680 must define the macro `_GNU_SOURCE' in your program or compile with
3681 `-D_GNU_SOURCE'.
3682
3683 * The library has changed from using GNU ld symbol aliases to using weak
3684 symbols where available. The ELF object file format supports weak
3685 symbols; GNU ld also supports weak symbols in the a.out format. (There
3686 is also now support for other GNU ld extensions in ELF. Use the
3687 `--with-elf' option to configure to indicate you have ELF, and
3688 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3689 of many files which contained only symbol aliases, reducing the size of
3690 the source and the compiled library; many other files were renamed to
3691 less cryptic names previously occupied by the symbol alias files.
3692 There is a new header file <elf.h> for programs which operate on
3693 files in the ELF format.
3694
3695 * Converted to Autoconf version 2, so `configure' has more options.
3696 Run `configure --help' to see the details.
3697
3698 * The library can now be configured to build profiling, highly-optimized
3699 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3700 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3701 `configure' enable building these extra libraries. The shared library is
3702 built by default when using both ELF and GNU ld. When shared libraries
3703 are enabled, the new library `-ldl' is available for arbitrary run-time
3704 loading of shared objects; its interface is defined in <dlfcn.h>. The
3705 new header file <link.h> gives access to the internals of the run-time
3706 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3707 application of same name on other systems and it provides information
3708 about dynamically linked binaries.
3709
3710 * The C library now provides the run-time support code for profiling
3711 executables compiled with `-pg'. Programs can control the profiling code
3712 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3713 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3714 the support for this file format was contributed by David Mosberger-Tang.
3715
3716 * The math code has been replaced with a math library based on fdlibm from
3717 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3718 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3719 functions. The math functions now reside in a separate library, so
3720 programs using them will need to use `-lm' their linking commands.
3721
3722 * John C. Bowman contributed optimized ix87 assembler inline functions.
3723
3724 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3725 mechanism similar to that found in Solaris 2. This is now used for the
3726 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3727 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3728 file controls what services are used for each individual database. This
3729 works by loading shared libraries with names specified in `nsswitch.conf',
3730 so service modules can be changed or added at any time without even
3731 relinking any program. Currently there are the file, db, and NIS based
3732 NSS services available.
3733
3734 * The new functions `strtoq' and `strtouq' parse integer values from
3735 strings, like `strtol' and `strtoul', but they return `long long int' and
3736 `unsigned long long int' values, respectively (64-bit quantities).
3737
3738 * The new functions `strtof' and `strtold' parse floating-point values from
3739 strings, like `strtod', but they return `float' and `long double' values,
3740 respectively (on some machines `double' and `long double' are the same).
3741
3742 * Ulrich Drepper has contributed new implementations of the floating-point
3743 printing and reading code used in the `printf' family of functions and
3744 `strtod', `strtof', and `strtold'. These new functions are perfectly
3745 accurate, and much faster than the old ones.
3746
3747 * The implementation of the POSIX locale model was completely rewritten by
3748 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3749 to compile the POSIX locale definition.
3750
3751 * The former dummy implementations of the strcoll and strxfrm function are
3752 now replaced by fully functional code contributed by Ulrich Drepper. The
3753 collation information comes from the POSIX locale definitions.
3754
3755 * The new header <langinfo.h> defines an interface for accessing
3756 various locale-dependent data (using the locale chosen with `setlocale').
3757
3758 * Ulrich Drepper has contributed a new suite of functions for operation on
3759 wide-character and multibyte-character strings, in <wchar.h>;
3760 and classification and case conversion of wide characters, in <wctype.h>.
3761 These new functions are conforming to the ISO C, Amendement 1 specification.
3762
3763 * There is now a second implementation of the standard I/O library available.
3764 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3765 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3766
3767 * You can now use positional parameter specifications in format strings
3768 for the `printf' and `scanf' families of functions. For example,
3769 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3770 ``Number 6, Mr Jones''. This is mainly useful when providing different
3771 format strings for different languages, whose grammars may dictate
3772 different orderings of the values being printed. To support this
3773 feature, the interface for `register_printf_handler' has changed; see
3774 the header file <printf.h> for details.
3775
3776 * The `printf' and `scanf' families of functions now understand a new
3777 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3778 to group numbers as indicated by the locale; for `scanf' and friends, this
3779 says to accept as valid only a number with all the proper grouping
3780 separators in the right places. In the default "C" locale, numbers are
3781 not grouped; but locales for specific countries will define the usual
3782 conventions (i.e. separate thousands with `,' in the US locale).
3783
3784 * The pgrp functions have been regularized, slightly incompatibly but much
3785 less confusingly. The core functions are now `getpgid' and `setpgid',
3786 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3787 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3788 provided for compatibility. There is no longer an incompatible `getpgrp'
3789 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3790
3791 * The new header file <fts.h> and suite of functions simplify programs that
3792 operate on directory trees. This code comes from 4.4 BSD.
3793
3794 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3795 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3796 NSS scheme used in glibc.
3797
3798 * There is a new malloc debugging hook `__memalign_hook'.
3799
3800 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3801 `unsigned int' in <sys/types.h>. These are for compatibility only and
3802 their use is discouraged.
3803
3804 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3805 done differently, so that it works even without GNU ld.
3806
3807 * New function `euidaccess' checks allowed access to a file like `access',
3808 but using the effective IDs instead of the real IDs.
3809
3810 * The time zone data files have been updated for the latest and greatest
3811 local time conventions of the countries of the world.
3812
3813 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3814 see <dirent.h>.
3815
3816 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3817 for formatting floating-point numbers. They are provided only for
3818 compatibility; new programs should use `sprintf' instead. There are
3819 also equivalent function for the `long double' floating-point type and
3820 all functions also exist in a reentrant form.
3821
3822 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3823 functions for maintaining the login-record files (primarily of use to
3824 system programs such as `login'), and convenient functions for
3825 allocating and initializing a pseudo-terminal (pty) device.
3826
3827 * Ulrich Drepper has contributed new support for System V style
3828 shared memory and IPC on systems that support it.
3829
3830 * Ulrich Drepper has contributed several miscellaneous new functions found
3831 in System V: The `hsearch' family of functions provide an effective
3832 implementation of hash tables; `a64l' and `l64a' provide a very simple
3833 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3834 number generator.
3835
3836 * Ulrich Drepper has contributed new reentrant counterparts for the
3837 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3838
3839 * Ulrich Drepper has contributed new, highly-optimized versions of several
3840 string functions for the i486/Pentium family of processors.
3841
3842 * Ulrich Drepper has updated the Linux-specific code, based largely
3843 on work done in Hongjiu Lu's version of GNU libc for Linux.
3844 The GNU library now supports Linux versions 2.0.10 and later,
3845 using the ELF object file format (i[3456]86-*-linux).
3846
3847 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3848
3849 * David Mosberger-Tang and Richard Henderson have ported the C library
3850 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3851 dynamic linking support for ELF/Alpha.
3852
3853 * Richard Henderson contributed several Alpha optimized assembler function
3854 for arithmetic and string handling.
3855
3856 * Ulrich Drepper has contributed a new set of message catalog functions to
3857 support multiple languages using the <libintl.h> interface, for use with
3858 his new package GNU gettext. Translation volunteers have contributed
3859 catalogs of the library's messages in Spanish, German, and Korean.
3860
3861 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3862 program and the `catgets' function for reading the catalog files it
3863 creates. (The <libintl.h> interface is preferred; we include the
3864 <nl_types.h> interface using `catgets' only for source compatibility with
3865 programs already written to use it.)
3866
3867 * New header file <values.h> gives SVID-compatible names for <limits.h>
3868 constants.
3869
3870 * Various new macros, declarations, and small header files for compatibility
3871 with 4.4 BSD.
3872
3873 * New function `group_member' is a convenient way to check if a process has
3874 a given effective group ID.
3875
3876 * When using GCC 2.7 and later, the socket functions are now declared in a
3877 special way so that passing an argument of type `struct sockaddr_in *',
3878 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3879 `struct sockaddr *' type, does not generate a type-clash warning.
3880
3881 * New function `error' declared in header file <error.h> is a convenient
3882 function for printing error messages and optionally exiting; this is the
3883 canonical function used in GNU programs. The new functions `err', `warn',
3884 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3885 doing the same thing.
3886
3887 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3888 POSIX.2 `glob' function to do ~ and {...} expansion.
3889
3890 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3891 `clearenv' which is used in POSIX.9 is also available.
3892
3893 * New function `getsid' returns session ID number on systems that support it.
3894
3895 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3896 files <db.h> and <mpool.h> provide a rich set of functions for several
3897 types of simple databases stored in memory and in files, and <ndbm.h> is
3898 an old `ndbm'-compatible interface using the `db' functions. Link with
3899 `-ldb' to get these functions.
3900
3901 * New macro `strdupa' copies a string like `strdup', but uses local stack
3902 space from `alloca' instead of dynamic heap space from `malloc'.
3903
3904 * New function `strnlen' is like `strlen' but searches only a given maximum
3905 number of characters for the null terminator. `stpncpy', `strndup' and
3906 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3907 function.
3908
3909 * New function `statfs' in header <sys/statfs.h>.
3910
3911 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3912 provide convenient functions for operating on blocks of null-terminated
3913 strings.
3914
3915 * A new suite of functions in <utmp.h> handle all the details of reading
3916 and writing the utmp file.
3917
3918 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3919 Thorsten Kukuk.
3920
3921 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3922 completely POSIX compliant and also implemented the extended functionality
3923 to handle alternate digit representation and alternate era date formats.
3924
3925 * Ulrich Drepper provided an implementation of the `strptime' function
3926 defined in XPG4.2 which transforms a string into a `struct tm' value.
3927
3928 * Paul Eggert provided the tzselect shell script as part of the timezone
3929 code. The shell script makes it easy to select the correct timezone
3930 specification.
3931
3932 * The implementation of the malloc family of functions is completely replaced
3933 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3934 The implementation uses the mmap function (if available) and it is
3935 optimized for the use in multi threaded programs.
3936
3937 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3938 functions. This new functionality is usable by specifying a special
3939 salt string and it is compatible with implementation on *BSD systems.
3940
3941 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3942 `getsubopt' to handle second level command line options, `bsd_signal'
3943 to access BSD style `signal' functionality, the obsolete `regexp' style
3944 expression matcher.
3945
3946 * the `lchown' function is available on system which support this
3947 functionality.
3948
3949 * The implementation of the shadow password handling function was contributed
3950 by Ulrich Drepper.
3951
3952 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3953
3954 * POSIX.1g support was added. The <sys/select.h> header is available,
3955 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3956 implementation of `getaddrinfo'.
3957 \f
3958 Version 1.09
3959
3960 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3961
3962 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3963 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3964
3965 * The new function `daemon' from 4.4 BSD is useful for server programs that
3966 want to put themselves in the background.
3967
3968 * Joel Sherrill has contributed support for several standalone boards that
3969 run without an operating system.
3970
3971 * `printf', `scanf' and friends now accept a `q' type modifier for long
3972 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3973
3974 * All of the code taken from BSD (notably most of the math and networking
3975 routines) has been updated from the BSD 4.4-Lite release.
3976
3977 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3978
3979 * The new functions `getdomainname' and `setdomainname' fetch or change the
3980 YP/NIS domain name. These are system calls which exist on systems which
3981 have YP (aka NIS).
3982
3983 * The time zone data files have been updated for the latest international
3984 conventions.
3985
3986 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3987 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3988 \f
3989 Version 1.08
3990
3991 * The C library now includes support for Sun RPC, from Sun's free
3992 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3993 are included. (There is still no support for YP.)
3994
3995 * Tom Quinn has contributed a port of the C library to SGI machines running
3996 Irix 4 (mips-sgi-irix4).
3997
3998 * The new `lockf' function is a simplified interface to the locking
3999 facilities of `fcntl', included for compatibility.
4000
4001 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4002
4003 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4004
4005 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4006 compatibility.
4007
4008 * The error message printed by `assert' for a failed assertion now includes
4009 the name of the program (if using GNU ld) and the name of the calling
4010 function (with versions of GCC that support this).
4011
4012 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4013
4014 * The library now includes the <sys/mman.h> header file and memory
4015 management functions `mmap', `munmap', `mprotect', `msync', and
4016 `madvise', on systems that support those facilities.
4017
4018 * The interface for `mcheck' has changed slightly: the function called to
4019 abort the program when an allocation inconsistency is detected now takes
4020 an argument that indicates the type of failure. The new function
4021 `mprobe' lets you request a consistency check for a particular block at
4022 any time (checks are normally done only when you call `free' or `realloc'
4023 on a block).
4024
4025 * It is now possible to easily cross-compile the C library, building on one
4026 system a library to run on another machine and/or operating system. All
4027 you need to do is set the variable `HOST_CC' in `configparms' to the
4028 native compiler for programs to run on the machine you are building on (a
4029 few generator programs are used on Unix systems); set `CC' to the
4030 cross-compiler.
4031
4032 * The new function `fexecve' (only implemented on the GNU system) executes
4033 a program file given a file descriptor already open on the file.
4034 \f
4035 Version 1.07
4036
4037 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4038 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4039
4040 * You can set the variable `libprefix' in `configparms' to specify a prefix
4041 to be prepended to installed library files; this makes it easy to install
4042 the GNU C library to be linked as `-lgnuc' or whatever.
4043
4044 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4045 copies a limited number of characters from a string, and returns the
4046 address of the last character written.
4047
4048 * You no longer need to check for whether the installed `stddef.h' is
4049 compatible with the GNU C library. configure now checks for you.
4050
4051 * You can now define a per-stream `fileno' function to convert the
4052 stream's cookie into an integral file descriptor.
4053
4054 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4055 allocates zero bytes of storage, and returns a unique pointer which
4056 you can pass to `realloc' or `free'. The behavior is undefined if
4057 you dereference this pointer.
4058
4059 * The C library now runs on Sony NEWS m68k machines running either
4060 NewsOS 3 or NewsOS 4.
4061
4062 * The new `syscall' function is a system-dependent primitive function
4063 for invoking system calls. It has the canonical behavior on Unix
4064 systems, including unreliable return values for some calls (such as
4065 `pipe', `fork' and `getppid').
4066
4067 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4068 to `EAGAIN', which is the preferred name. On systems whose kernels
4069 use two distinct codes, the C library now translates EWOULDBLOCK to
4070 EAGAIN in every system call function.
4071 \f
4072 Version 1.06
4073
4074 * The GNU C Library Reference Manual is now distributed with the library.
4075 `make dvi' will produce a DVI file of the printed manual.
4076 `make info' will produce Info files that you can read on line using C-h i
4077 in Emacs or the `info' program.
4078 Please send comments on the manual to bug-glibc-manual@gnu.org.
4079
4080 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4081
4082 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4083
4084 * Jason Merrill has contributed a port to the Sequent Symmetry running
4085 Dynix version 3 (i386-sequent-dynix).
4086
4087 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4088 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4089
4090 * New function `memory_warnings' lets you arrange to get warnings when
4091 malloc is running out of memory to allocate, like Emacs gives you.
4092
4093 * The C library now contains the relocating allocator used in Emacs 19 for
4094 its editing buffers. This allocator (ralloc) minimizes allocation
4095 overhead and fragmentation by moving allocated regions around whenever it
4096 needs to. You always refer to a ralloc'd region with a "handle" (a
4097 pointer to a pointer--an object of type `void **').
4098
4099 * There is a new `printf' format: `%m' gives you the string corresponding
4100 to the error code in `errno'.
4101
4102 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4103 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4104 pass, the `a' modifier says to fill in a `char **' you pass with a
4105 malloc'd string.
4106
4107 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4108 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4109 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4110
4111 * `mkstemp' is a traditional Unix function to atomically create and open a
4112 uniquely-named temporary file.
4113 \f
4114 Version 1.05
4115
4116 * The standard location for the file that says what the local timezone is
4117 has changed again. It is now `/usr/local/etc/localtime' (or more
4118 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4119
4120 * The distribution no longer contains any files with names longer than 14
4121 characters.
4122
4123 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4124 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4125
4126 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4127 \f
4128 Version 1.04
4129
4130 * The configuration process has changed quite a bit. The `configure'
4131 script is now used just like the configuration scripts for other GNU
4132 packages. The `sysdeps' directory hierarchy is much rearranged.
4133 The file `INSTALL' explains the new scheme in detail.
4134
4135 * The header files no longer need to be processed into ANSI C and
4136 traditional C versions. There is just one set of files to install, and
4137 it will work with ANSI or old C compilers (including `gcc -traditional').
4138
4139 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4140 MIPS DECStation running Ultrix 4.
4141
4142 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4143 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4144 made itself into a shared library.
4145
4146 * Yet further improved support for the i386, running 4.3 BSD-like systems
4147 (such as Mach 3 with the Unix single-server), or System V.
4148
4149 * New function `strncasecmp' to do case-insensitive string comparison
4150 with limited length.
4151
4152 * New function `strsep' is a reentrant alternative to `strtok'.
4153
4154 * New functions `scandir' and `alphasort' for searching directories.
4155
4156 * New function `setenv' is a better interface to `putenv'.
4157
4158 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4159 function for traversing a directory tree.
4160
4161 * The GNU obstack package is now also part of the C library.
4162 The new function `open_obstack_stream' creates a stdio stream that
4163 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4164 formatted output directly to an obstack.
4165
4166 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4167 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4168
4169 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4170
4171 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4172 things to your strings.
4173
4174 * There are some new test programs: test-fseek, testmb, and testrand.
4175
4176 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4177 These ports are not finished, but are a good starting place for really
4178 supporting those systems.
4179
4180 * `/etc/localtime' is now the standard location for the file that says what
4181 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4182 This follows the general principle that `/etc' is the place for all local
4183 configuration files.
4184
4185 * The C library header files now use `extern "C"' when used by the C++
4186 compiler, so the C library should now work with C++ code.
4187
4188 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4189 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4190 in <strings.h>.)
4191
4192 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4193 function which uses the `merge sort' algorithm, and is said to be
4194 significantly faster than the old GNU `qsort' function. Merge sort is
4195 now the standard `qsort' function. The new algorithm can require a lot
4196 of temporary storage; so, the old sorting function is called when the
4197 required storage is not available.
4198
4199 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4200 provides the Unix `crypt' function, plus some other entry points.
4201
4202 * The code and header files taken from 4.4 BSD have been updated with the
4203 latest files released from Berkeley.
4204 \f
4205 ----------------------------------------------------------------------
4206 Copyright information:
4207
4208 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4209
4210 Permission is granted to anyone to make or distribute verbatim copies
4211 of this document as received, in any medium, provided that the
4212 copyright notice and this permission notice are preserved,
4213 thus giving the recipient permission to redistribute in turn.
4214
4215 Permission is granted to distribute modified versions
4216 of this document, or of portions of it,
4217 under the above conditions, provided also that they
4218 carry prominent notices stating who last changed them.
4219 \f
4220 Local variables:
4221 version-control: never
4222 fill-column: 76
4223 End: