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