[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Commits] r24093 - in /fsf/trunk/libc: ./ conform/data/ locale/programs/ localedata/ localedata/locales/ manual/ ports/ ports/sysdeps/...
- To: commits@xxxxxxxxxx
- Subject: [Commits] r24093 - in /fsf/trunk/libc: ./ conform/data/ locale/programs/ localedata/ localedata/locales/ manual/ ports/ ports/sysdeps/...
- From: eglibc@xxxxxxxxxx
- Date: Wed, 25 Sep 2013 00:02:26 -0000
Author: eglibc
Date: Wed Sep 25 00:02:24 2013
New Revision: 24093
Log:
Import glibc-mainline for 2013-09-25
Added:
fsf/trunk/libc/localedata/locales/cmn_TW
fsf/trunk/libc/localedata/locales/hak_TW
fsf/trunk/libc/ports/sysdeps/unix/sysv/linux/aarch64/sigcontextinfo.h
Modified:
fsf/trunk/libc/ChangeLog
fsf/trunk/libc/NEWS
fsf/trunk/libc/conform/data/pthread.h-data
fsf/trunk/libc/locale/programs/ld-collate.c
fsf/trunk/libc/locale/programs/localedef.c
fsf/trunk/libc/locale/programs/locarchive.c
fsf/trunk/libc/locale/programs/locfile.c
fsf/trunk/libc/locale/programs/locfile.h
fsf/trunk/libc/localedata/ChangeLog
fsf/trunk/libc/localedata/SUPPORTED
fsf/trunk/libc/manual/freemanuals.texi
fsf/trunk/libc/manual/libc.texinfo
fsf/trunk/libc/ports/ChangeLog.aarch64
fsf/trunk/libc/ports/ChangeLog.arm
fsf/trunk/libc/ports/sysdeps/aarch64/machine-gmon.h
fsf/trunk/libc/ports/sysdeps/arm/nptl/tls.h
fsf/trunk/libc/stdio-common/tst-long-dbl-fphex.c
fsf/trunk/libc/sysdeps/mach/hurd/i386/tls.h
Modified: fsf/trunk/libc/ChangeLog
==============================================================================
--- fsf/trunk/libc/ChangeLog (original)
+++ fsf/trunk/libc/ChangeLog Wed Sep 25 00:02:24 2013
@@ -1,3 +1,72 @@
+2013-09-24 Olivier Langlois <olivier@xxxxxxxxxxxxxxxxxxx>
+
+ * stdio-common/tst-long-dbl-fphex.c (do_test): Fix swprintf length
+ argument calculation.
+
+2013-09-24 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+
+ * conform/data/pthread.h-data [POSIX] (PTHREAD_MUTEX_INITIALIZER):
+ Expect macro.
+ [POSIX] (pthread_attr_t): Do not require type.
+ [POSIX] (pthread_cond_t): Likewise.
+ [POSIX] (pthread_condattr_t): Likewise.
+ [POSIX] (pthread_key_t): Likewise.
+ [POSIX] (pthread_mutex_t): Likewise.
+ [POSIX] (pthread_mutexattr_t): Likewise.
+ [POSIX] (pthread_once_t): Likewise.
+ [POSIX] (pthread_t): Likewise.
+ [POSIX-based standards] (pthread_atfork): Expect function.
+
+2013-09-24 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+ Richard Sandiford <richard@xxxxxxxxxxxxxxxx>
+
+ * locale/programs/locfile.h: Include <byteswap.h> and <stdbool.h>.
+ (swap_endianness_p): New extern variable.
+ (set_big_endian): New inline function.
+ (maybe_swap_uint32): Likewise.
+ (maybe_swap_uint32_array): Likewise.
+ (maybe_swap_uint32_obstack): Likewise.
+ * locale/programs/locfile.c: Include <stdbool.h>.
+ (swap_endianness_p): New variable.
+ (add_locale_uint32): Call maybe_swap_uint32.
+ (add_locale_uint32_array): Call maybe_swap_uint32_obstack.
+ (write_locale_data): Call maybe_swap_uint32_array.
+ * locale/programs/ld-collate.c (obstack_int32_grow): Call
+ maybe_swap_uint32.
+ (obstack_int32_grow_fast): Likewise.
+ (output_weightwc): Call maybe_swap_uint32_obstack.
+ (collate_output): Likewise.
+ * locale/programs/localedef.c (OPT_LITTLE_ENDIAN): New macro.
+ (OPT_LITTLE_ENDIAN): Likewise.
+ (options): Add --little-endian and --big-endian options.
+ (parse_opt): Handle OPT_LITTLE_ENDIAN and OPT_BIG_ENDIAN.
+ * locale/programs/locarchive.c: Include "locfile.h".
+ (GET): New macro.
+ (SET): Likewise.
+ (INC): Likewise.
+ (create_archive): Use the new macros to access fields of
+ structures directly mapped from or written to locale archives.
+ (oldlocrecentcmp): Likewise.
+ (enlarge_archive): Likewise.
+ (insert_name): Likewise.
+ (add_alias): Likewise.
+ (add_locale): Likewise.
+ (delete_locales_from_archive): Likewise.
+ (show_archive_content): Likewise.
+ (add_locale_to_archive): Likewise. Use maybe_swap_uint32 on
+ locale data.
+
+2013-09-24 Roland McGrath <roland@xxxxxxxxxxxxx>
+
+ * manual/freemanuals.texi: Updated from (newly) canonical copy at
+ http://www.gnu.org/doc/freemanuals.texi.
+ * manual/libc.texinfo (Free Manuals): Put the @appendix line here.
+
+2013-09-24 Will Newton <will.newton@xxxxxxxxxx>
+
+ * sysdeps/mach/hurd/i386/tls.h (TLS_INIT_TP_EXPENSIVE): Remove
+ macro.
+
2013-09-23 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
* locale/hashval.h (compute_hashval): Interpret bytes of key as
Modified: fsf/trunk/libc/NEWS
==============================================================================
--- fsf/trunk/libc/NEWS (original)
+++ fsf/trunk/libc/NEWS Wed Sep 25 00:02:24 2013
@@ -9,10 +9,10 @@
* The following bugs are resolved with this release:
- 13985, 14155, 14547, 14699, 15427, 15522, 15531, 15532, 15608, 15640,
- 15681, 15736, 15748, 15749, 15754, 15797, 15844, 15849, 15855, 15856,
- 15857, 15859, 15867, 15886, 15887, 15890, 15892, 15893, 15895, 15897,
- 15905, 15909, 15919, 15921, 15923, 15939, 15963, 15966.
+ 13985, 14155, 14547, 14699, 15427, 15522, 15531, 15532, 15608, 15609,
+ 15610, 15640, 15681, 15736, 15748, 15749, 15754, 15797, 15844, 15849,
+ 15855, 15856, 15857, 15859, 15867, 15886, 15887, 15890, 15892, 15893,
+ 15895, 15897, 15905, 15909, 15919, 15921, 15923, 15939, 15963, 15966.
* CVE-2012-4412 The strcoll implementation caches indices and rules for
large collation sequences to optimize multiple passes. This cache
@@ -41,12 +41,16 @@
heap when passed very large allocation size values (Bugzilla #15855,
#15856, #15857).
-* New locales: ak_GH, lzh_TW, nan_TW, quz_PE.
+* New locales: ak_GH, cmn_TW, hak_TW, lzh_TW, nan_TW, quz_PE.
* The LC_ADDRESS field was updated to support country_car for almost all
supported locales.
* ISO 1427 definitions were updated.
+
+* The localedef utility now supports --big-endian and --little-endian
+ command-line options to generate locales for a different system from that
+ for which the C library was built.
* The configure option --disable-versioning has been removed. Builds with
--disable-versioning had not worked for several years.
Modified: fsf/trunk/libc/conform/data/pthread.h-data
==============================================================================
--- fsf/trunk/libc/conform/data/pthread.h-data (original)
+++ fsf/trunk/libc/conform/data/pthread.h-data Wed Sep 25 00:02:24 2013
@@ -17,11 +17,11 @@
constant PTHREAD_PROCESS_PRIVATE
constant PTHREAD_SCOPE_PROCESS
constant PTHREAD_SCOPE_SYSTEM
+macro PTHREAD_MUTEX_INITIALIZER
# ifndef POSIX
constant PTHREAD_MUTEX_DEFAULT
constant PTHREAD_MUTEX_ERRORCHECK
constant PTHREAD_MUTEX_NORMAL
-macro PTHREAD_MUTEX_INITIALIZER
constant PTHREAD_MUTEX_RECURSIVE
macro PTHREAD_RWLOCK_INITIALIZER
# endif
@@ -30,26 +30,27 @@
constant PTHREAD_MUTEX_STALLED
# endif
+# ifndef POSIX
type pthread_attr_t
-# if !defined POSIX && !defined UNIX98
+# ifndef UNIX98
type pthread_barrier_t
type pthread_barrierattr_t
-# endif
+# endif
type pthread_cond_t
type pthread_condattr_t
type pthread_key_t
type pthread_mutex_t
type pthread_mutexattr_t
type pthread_once_t
-# if !defined POSIX
type pthread_rwlock_t
type pthread_rwlockattr_t
+# ifndef UNIX98
+type pthread_spinlock_t
+# endif
+type pthread_t
# endif
-# if !defined POSIX && !defined UNIX98
-type pthread_spinlock_t
-# endif
-type pthread_t
+function int pthread_atfork (void (*) (void), void (*) (void), void (*) (void))
function int pthread_attr_destroy (pthread_attr_t*)
function int pthread_attr_getdetachstate (const pthread_attr_t*, int*)
# ifndef POSIX
Modified: fsf/trunk/libc/locale/programs/ld-collate.c
==============================================================================
--- fsf/trunk/libc/locale/programs/ld-collate.c (original)
+++ fsf/trunk/libc/locale/programs/ld-collate.c Wed Sep 25 00:02:24 2013
@@ -44,6 +44,7 @@
__attribute ((always_inline))
obstack_int32_grow (struct obstack *obstack, int32_t data)
{
+ data = maybe_swap_uint32 (data);
if (sizeof (int32_t) == sizeof (int))
obstack_int_grow (obstack, data);
else
@@ -54,6 +55,7 @@
__attribute ((always_inline))
obstack_int32_grow_fast (struct obstack *obstack, int32_t data)
{
+ data = maybe_swap_uint32 (data);
if (sizeof (int32_t) == sizeof (int))
obstack_int_grow_fast (obstack, data);
else
@@ -1955,6 +1957,7 @@
obstack_int32_grow (pool, j);
obstack_grow (pool, buf, j * sizeof (int32_t));
+ maybe_swap_uint32_obstack (pool, j);
}
return retval | ((elem->section->ruleidx & 0x7f) << 24);
@@ -2479,6 +2482,7 @@
obstack_int32_grow (&extrapool, runp->nwcs);
obstack_grow (&extrapool, runp->wcs,
runp->nwcs * sizeof (uint32_t));
+ maybe_swap_uint32_obstack (&extrapool, runp->nwcs);
obstack_int32_grow (&extrapool, runp->wcseqorder);
}
Modified: fsf/trunk/libc/locale/programs/localedef.c
==============================================================================
--- fsf/trunk/libc/locale/programs/localedef.c (original)
+++ fsf/trunk/libc/locale/programs/localedef.c Wed Sep 25 00:02:24 2013
@@ -112,6 +112,8 @@
#define OPT_REPLACE 307
#define OPT_DELETE_FROM_ARCHIVE 308
#define OPT_LIST_ARCHIVE 309
+#define OPT_LITTLE_ENDIAN 400
+#define OPT_BIG_ENDIAN 401
/* Definitions of arguments for argp functions. */
static const struct argp_option options[] =
@@ -144,6 +146,10 @@
{ "list-archive", OPT_LIST_ARCHIVE, NULL, 0, N_("List content of archive") },
{ "alias-file", 'A', N_("FILE"), 0,
N_("locale.alias file to consult when making archive")},
+ { "little-endian", OPT_LITTLE_ENDIAN, NULL, 0,
+ N_("Generate little-endian output") },
+ { "big-endian", OPT_BIG_ENDIAN, NULL, 0,
+ N_("Generate big-endian output") },
{ NULL, 0, NULL, 0, NULL }
};
@@ -325,6 +331,12 @@
break;
case OPT_LIST_ARCHIVE:
list_archive = true;
+ break;
+ case OPT_LITTLE_ENDIAN:
+ set_big_endian (false);
+ break;
+ case OPT_BIG_ENDIAN:
+ set_big_endian (true);
break;
case 'c':
force_output = 1;
Modified: fsf/trunk/libc/locale/programs/locarchive.c
==============================================================================
--- fsf/trunk/libc/locale/programs/locarchive.c (original)
+++ fsf/trunk/libc/locale/programs/locarchive.c Wed Sep 25 00:02:24 2013
@@ -46,6 +46,7 @@
#include "../localeinfo.h"
#include "../locarchive.h"
#include "localedef.h"
+#include "locfile.h"
/* Define the hash function. We define the function as static inline.
We must change the name so as not to conflict with simple-hash.h. */
@@ -72,6 +73,13 @@
#define INITIAL_SIZE_STRINGS 7500
#define INITIAL_NUM_LOCREC 420
#define INITIAL_NUM_SUMS 2000
+
+
+/* Get and set values (possibly endian-swapped) in structures mapped
+ from or written directly to locale archives. */
+#define GET(FIELD) maybe_swap_uint32 (FIELD)
+#define SET(FIELD, VALUE) ((FIELD) = maybe_swap_uint32 (VALUE))
+#define INC(FIELD, INCREMENT) SET (FIELD, GET (FIELD) + (INCREMENT))
/* Size of the reserved address space area. */
@@ -125,27 +133,31 @@
error (EXIT_FAILURE, errno, _("cannot create temporary file: %s"), fname);
/* Create the initial content of the archive. */
- head.magic = AR_MAGIC;
- head.serial = 0;
- head.namehash_offset = sizeof (struct locarhead);
- head.namehash_used = 0;
- head.namehash_size = next_prime (INITIAL_NUM_NAMES);
-
- head.string_offset = (head.namehash_offset
- + head.namehash_size * sizeof (struct namehashent));
- head.string_used = 0;
- head.string_size = INITIAL_SIZE_STRINGS;
-
- head.locrectab_offset = head.string_offset + head.string_size;
- head.locrectab_used = 0;
- head.locrectab_size = INITIAL_NUM_LOCREC;
-
- head.sumhash_offset = (head.locrectab_offset
- + head.locrectab_size * sizeof (struct locrecent));
- head.sumhash_used = 0;
- head.sumhash_size = next_prime (INITIAL_NUM_SUMS);
-
- total = head.sumhash_offset + head.sumhash_size * sizeof (struct sumhashent);
+ SET (head.magic, AR_MAGIC);
+ SET (head.serial, 0);
+ SET (head.namehash_offset, sizeof (struct locarhead));
+ SET (head.namehash_used, 0);
+ SET (head.namehash_size, next_prime (INITIAL_NUM_NAMES));
+
+ SET (head.string_offset,
+ (GET (head.namehash_offset)
+ + GET (head.namehash_size) * sizeof (struct namehashent)));
+ SET (head.string_used, 0);
+ SET (head.string_size, INITIAL_SIZE_STRINGS);
+
+ SET (head.locrectab_offset,
+ GET (head.string_offset) + GET (head.string_size));
+ SET (head.locrectab_used, 0);
+ SET (head.locrectab_size, INITIAL_NUM_LOCREC);
+
+ SET (head.sumhash_offset,
+ (GET (head.locrectab_offset)
+ + GET (head.locrectab_size) * sizeof (struct locrecent)));
+ SET (head.sumhash_used, 0);
+ SET (head.sumhash_size, next_prime (INITIAL_NUM_SUMS));
+
+ total = (GET (head.sumhash_offset)
+ + GET (head.sumhash_size) * sizeof (struct sumhashent));
/* Write out the header and create room for the other data structures. */
if (TEMP_FAILURE_RETRY (write (fd, &head, sizeof (head))) != sizeof (head))
@@ -240,10 +252,10 @@
for (cnt = 0; cnt < __LC_LAST; ++cnt)
if (cnt != LC_ALL)
{
- if (la->record[cnt].offset < start_a)
- start_a = la->record[cnt].offset;
- if (la->record[cnt].offset + la->record[cnt].len > end_a)
- end_a = la->record[cnt].offset + la->record[cnt].len;
+ if (GET (la->record[cnt].offset) < start_a)
+ start_a = GET (la->record[cnt].offset);
+ if (GET (la->record[cnt].offset) + GET (la->record[cnt].len) > end_a)
+ end_a = GET (la->record[cnt].offset) + GET (la->record[cnt].len);
}
assert (start_a != (uint32_t)-1);
assert (end_a != 0);
@@ -251,10 +263,10 @@
for (cnt = 0; cnt < __LC_LAST; ++cnt)
if (cnt != LC_ALL)
{
- if (lb->record[cnt].offset < start_b)
- start_b = lb->record[cnt].offset;
- if (lb->record[cnt].offset + lb->record[cnt].len > end_b)
- end_b = lb->record[cnt].offset + lb->record[cnt].len;
+ if (GET (lb->record[cnt].offset) < start_b)
+ start_b = GET (lb->record[cnt].offset);
+ if (GET (lb->record[cnt].offset) + GET (lb->record[cnt].len) > end_b)
+ end_b = GET (lb->record[cnt].offset) + GET (lb->record[cnt].len);
}
assert (start_b != (uint32_t)-1);
assert (end_b != 0);
@@ -371,38 +383,42 @@
/* Create the new archive header. The sizes of the various tables
should be double from what is currently used. */
- newhead.namehash_size = MAX (next_prime (2 * newhead.namehash_used),
- newhead.namehash_size);
+ SET (newhead.namehash_size,
+ MAX (next_prime (2 * GET (newhead.namehash_used)),
+ GET (newhead.namehash_size)));
if (verbose)
printf ("name: size: %u, used: %d, new: size: %u\n",
- head->namehash_size, head->namehash_used, newhead.namehash_size);
-
- newhead.string_offset = (newhead.namehash_offset
- + (newhead.namehash_size
- * sizeof (struct namehashent)));
+ GET (head->namehash_size),
+ GET (head->namehash_used), GET (newhead.namehash_size));
+
+ SET (newhead.string_offset, (GET (newhead.namehash_offset)
+ + (GET (newhead.namehash_size)
+ * sizeof (struct namehashent))));
/* Keep the string table size aligned to 4 bytes, so that
all the struct { uint32_t } types following are happy. */
- newhead.string_size = MAX ((2 * newhead.string_used + 3) & -4,
- newhead.string_size);
-
- newhead.locrectab_offset = newhead.string_offset + newhead.string_size;
- newhead.locrectab_size = MAX (2 * newhead.locrectab_used,
- newhead.locrectab_size);
-
- newhead.sumhash_offset = (newhead.locrectab_offset
- + (newhead.locrectab_size
- * sizeof (struct locrecent)));
- newhead.sumhash_size = MAX (next_prime (2 * newhead.sumhash_used),
- newhead.sumhash_size);
-
- total = (newhead.sumhash_offset
- + newhead.sumhash_size * sizeof (struct sumhashent));
+ SET (newhead.string_size, MAX ((2 * GET (newhead.string_used) + 3) & -4,
+ GET (newhead.string_size)));
+
+ SET (newhead.locrectab_offset,
+ GET (newhead.string_offset) + GET (newhead.string_size));
+ SET (newhead.locrectab_size, MAX (2 * GET (newhead.locrectab_used),
+ GET (newhead.locrectab_size)));
+
+ SET (newhead.sumhash_offset, (GET (newhead.locrectab_offset)
+ + (GET (newhead.locrectab_size)
+ * sizeof (struct locrecent))));
+ SET (newhead.sumhash_size,
+ MAX (next_prime (2 * GET (newhead.sumhash_used)),
+ GET (newhead.sumhash_size)));
+
+ total = (GET (newhead.sumhash_offset)
+ + GET (newhead.sumhash_size) * sizeof (struct sumhashent));
/* The new file is empty now. */
- newhead.namehash_used = 0;
- newhead.string_used = 0;
- newhead.locrectab_used = 0;
- newhead.sumhash_used = 0;
+ SET (newhead.namehash_used, 0);
+ SET (newhead.string_used, 0);
+ SET (newhead.locrectab_used, 0);
+ SET (newhead.sumhash_used, 0);
/* Write out the header and create room for the other data structures. */
if (TEMP_FAILURE_RETRY (write (fd, &newhead, sizeof (newhead)))
@@ -453,17 +469,17 @@
/* Walk through the hash name hash table to find out what data is
still referenced and transfer it into the new file. */
oldnamehashtab = (struct namehashent *) ((char *) ah->addr
- + head->namehash_offset);
+ + GET (head->namehash_offset));
/* Sort the old locrec table in order of data position. */
- struct oldlocrecent oldlocrecarray[head->namehash_size];
- for (cnt = 0, loccnt = 0; cnt < head->namehash_size; ++cnt)
- if (oldnamehashtab[cnt].locrec_offset != 0)
+ struct oldlocrecent oldlocrecarray[GET (head->namehash_size)];
+ for (cnt = 0, loccnt = 0; cnt < GET (head->namehash_size); ++cnt)
+ if (GET (oldnamehashtab[cnt].locrec_offset) != 0)
{
oldlocrecarray[loccnt].cnt = cnt;
oldlocrecarray[loccnt++].locrec
= (struct locrecent *) ((char *) ah->addr
- + oldnamehashtab[cnt].locrec_offset);
+ + GET (oldnamehashtab[cnt].locrec_offset));
}
qsort (oldlocrecarray, loccnt, sizeof (struct oldlocrecent),
oldlocrecentcmp);
@@ -479,9 +495,9 @@
for (idx = 0; idx < __LC_LAST; ++idx)
if (idx != LC_ALL)
{
- old_data[idx].size = oldlocrec->record[idx].len;
+ old_data[idx].size = GET (oldlocrec->record[idx].len);
old_data[idx].addr
- = ((char *) ah->addr + oldlocrec->record[idx].offset);
+ = ((char *) ah->addr + GET (oldlocrec->record[idx].offset));
__md5_buffer (old_data[idx].addr, old_data[idx].size,
old_data[idx].sum);
@@ -491,20 +507,23 @@
{
const char *oldname
= ((char *) ah->addr
- + oldnamehashtab[oldlocrecarray[cnt - 1].cnt].name_offset);
-
- add_alias (&new_ah,
- ((char *) ah->addr
- + oldnamehashtab[oldlocrecarray[cnt].cnt].name_offset),
- 0, oldname, &last_locrec_offset);
+ + GET (oldnamehashtab[oldlocrecarray[cnt
+ - 1].cnt].name_offset));
+
+ add_alias
+ (&new_ah,
+ ((char *) ah->addr
+ + GET (oldnamehashtab[oldlocrecarray[cnt].cnt].name_offset)),
+ 0, oldname, &last_locrec_offset);
continue;
}
last_locrec_offset =
- add_locale (&new_ah,
- ((char *) ah->addr
- + oldnamehashtab[oldlocrecarray[cnt].cnt].name_offset),
- old_data, 0);
+ add_locale
+ (&new_ah,
+ ((char *) ah->addr
+ + GET (oldnamehashtab[oldlocrecarray[cnt].cnt].name_offset)),
+ old_data, 0);
if (last_locrec_offset == 0)
error (EXIT_FAILURE, 0, _("cannot extend locale archive file"));
}
@@ -672,26 +691,28 @@
{
const struct locarhead *const head = ah->addr;
struct namehashent *namehashtab
- = (struct namehashent *) ((char *) ah->addr + head->namehash_offset);
+ = (struct namehashent *) ((char *) ah->addr
+ + GET (head->namehash_offset));
unsigned int insert_idx, idx, incr;
/* Hash value of the locale name. */
uint32_t hval = archive_hashval (name, name_len);
insert_idx = -1;
- idx = hval % head->namehash_size;
- incr = 1 + hval % (head->namehash_size - 2);
+ idx = hval % GET (head->namehash_size);
+ incr = 1 + hval % (GET (head->namehash_size) - 2);
/* If the name_offset field is zero this means this is a
deleted entry and therefore no entry can be found. */
- while (namehashtab[idx].name_offset != 0)
- {
- if (namehashtab[idx].hashval == hval
- && strcmp (name,
- (char *) ah->addr + namehashtab[idx].name_offset) == 0)
+ while (GET (namehashtab[idx].name_offset) != 0)
+ {
+ if (GET (namehashtab[idx].hashval) == hval
+ && (strcmp (name,
+ (char *) ah->addr + GET (namehashtab[idx].name_offset))
+ == 0))
{
/* Found the entry. */
- if (namehashtab[idx].locrec_offset != 0 && ! replace)
+ if (GET (namehashtab[idx].locrec_offset) != 0 && ! replace)
{
if (! be_quiet)
error (0, 0, _("locale '%s' already exists"), name);
@@ -701,26 +722,27 @@
break;
}
- if (namehashtab[idx].hashval == hval && ! be_quiet)
+ if (GET (namehashtab[idx].hashval) == hval && ! be_quiet)
{
error (0, 0, "hash collision (%u) %s, %s",
- hval, name, (char *) ah->addr + namehashtab[idx].name_offset);
+ hval, name,
+ (char *) ah->addr + GET (namehashtab[idx].name_offset));
}
/* Remember the first place we can insert the new entry. */
- if (namehashtab[idx].locrec_offset == 0 && insert_idx == -1)
+ if (GET (namehashtab[idx].locrec_offset) == 0 && insert_idx == -1)
insert_idx = idx;
idx += incr;
- if (idx >= head->namehash_size)
- idx -= head->namehash_size;
+ if (idx >= GET (head->namehash_size))
+ idx -= GET (head->namehash_size);
}
/* Add as early as possible. */
if (insert_idx != -1)
idx = insert_idx;
- namehashtab[idx].hashval = hval; /* no-op if replacing an old entry. */
+ SET (namehashtab[idx].hashval, hval); /* no-op if replacing an old entry. */
return &namehashtab[idx];
}
@@ -736,12 +758,13 @@
if (namehashent == NULL && ! replace)
return;
- if (namehashent->name_offset == 0)
+ if (GET (namehashent->name_offset) == 0)
{
/* We are adding a new hash entry for this alias.
Determine whether we have to resize the file. */
- if (head->string_used + name_len + 1 > head->string_size
- || 100 * head->namehash_used > 75 * head->namehash_size)
+ if (GET (head->string_used) + name_len + 1 > GET (head->string_size)
+ || (100 * GET (head->namehash_used)
+ > 75 * GET (head->namehash_size)))
{
/* The current archive is not large enough. */
enlarge_archive (ah, head);
@@ -749,9 +772,9 @@
/* The locrecent might have moved, so we have to look up
the old name afresh. */
namehashent = insert_name (ah, oldname, strlen (oldname), true);
- assert (namehashent->name_offset != 0);
- assert (namehashent->locrec_offset != 0);
- *locrec_offset_p = namehashent->locrec_offset;
+ assert (GET (namehashent->name_offset) != 0);
+ assert (GET (namehashent->locrec_offset) != 0);
+ *locrec_offset_p = GET (namehashent->locrec_offset);
/* Tail call to try the whole thing again. */
add_alias (ah, alias, replace, oldname, locrec_offset_p);
@@ -759,26 +782,27 @@
}
/* Add the name string. */
- memcpy (ah->addr + head->string_offset + head->string_used,
+ memcpy (ah->addr + GET (head->string_offset) + GET (head->string_used),
alias, name_len + 1);
- namehashent->name_offset = head->string_offset + head->string_used;
- head->string_used += name_len + 1;
-
- ++head->namehash_used;
- }
-
- if (namehashent->locrec_offset != 0)
+ SET (namehashent->name_offset,
+ GET (head->string_offset) + GET (head->string_used));
+ INC (head->string_used, name_len + 1);
+
+ INC (head->namehash_used, 1);
+ }
+
+ if (GET (namehashent->locrec_offset) != 0)
{
/* Replacing an existing entry.
Mark that we are no longer using the old locrecent. */
struct locrecent *locrecent
= (struct locrecent *) ((char *) ah->addr
- + namehashent->locrec_offset);
- --locrecent->refs;
+ + GET (namehashent->locrec_offset));
+ INC (locrecent->refs, -1);
}
/* Point this entry at the locrecent installed for the main name. */
- namehashent->locrec_offset = locrec_offset;
+ SET (namehashent->locrec_offset, locrec_offset);
}
static int /* qsort comparator used below */
@@ -819,7 +843,7 @@
head = ah->addr;
sumhashtab = (struct sumhashent *) ((char *) ah->addr
- + head->sumhash_offset);
+ + GET (head->sumhash_offset));
memset (file_offsets, 0, sizeof (file_offsets));
@@ -877,10 +901,10 @@
table. */
hval = archive_hashval (data[cnt].sum, 16);
- idx = hval % head->sumhash_size;
- incr = 1 + hval % (head->sumhash_size - 2);
-
- while (sumhashtab[idx].file_offset != 0)
+ idx = hval % GET (head->sumhash_size);
+ incr = 1 + hval % (GET (head->sumhash_size) - 2);
+
+ while (GET (sumhashtab[idx].file_offset) != 0)
{
if (memcmp (data[cnt].sum, sumhashtab[idx].sum, 16) == 0)
{
@@ -890,40 +914,42 @@
Unfortunately the sumhashent record does not include
the size of the stored data. So we have to search for
it. */
- locrecent = (struct locrecent *) ((char *) ah->addr
- + head->locrectab_offset);
+ locrecent
+ = (struct locrecent *) ((char *) ah->addr
+ + GET (head->locrectab_offset));
size_t iloc;
- for (iloc = 0; iloc < head->locrectab_used; ++iloc)
- if (locrecent[iloc].refs != 0
- && (locrecent[iloc].record[cnt].offset
- == sumhashtab[idx].file_offset))
+ for (iloc = 0; iloc < GET (head->locrectab_used); ++iloc)
+ if (GET (locrecent[iloc].refs) != 0
+ && (GET (locrecent[iloc].record[cnt].offset)
+ == GET (sumhashtab[idx].file_offset)))
break;
- if (iloc != head->locrectab_used
- && data[cnt].size == locrecent[iloc].record[cnt].len
+ if (iloc != GET (head->locrectab_used)
+ && data[cnt].size == GET (locrecent[iloc].record[cnt].len)
/* We have to compare the content. Either we can
have the data mmaped or we have to read from
the file. */
- && (file_data_available_p (ah, sumhashtab[idx].file_offset,
- data[cnt].size)
+ && (file_data_available_p
+ (ah, GET (sumhashtab[idx].file_offset),
+ data[cnt].size)
? memcmp (data[cnt].addr,
(char *) ah->addr
- + sumhashtab[idx].file_offset,
+ + GET (sumhashtab[idx].file_offset),
data[cnt].size) == 0
: compare_from_file (ah, data[cnt].addr,
- sumhashtab[idx].file_offset,
+ GET (sumhashtab[idx].file_offset),
data[cnt].size) == 0))
{
/* Found it. */
- file_offsets[cnt] = sumhashtab[idx].file_offset;
+ file_offsets[cnt] = GET (sumhashtab[idx].file_offset);
--num_new_offsets;
break;
}
}
idx += incr;
- if (idx >= head->sumhash_size)
- idx -= head->sumhash_size;
+ if (idx >= GET (head->sumhash_size))
+ idx -= GET (head->sumhash_size);
}
}
@@ -933,11 +959,14 @@
return 0;
/* Determine whether we have to resize the file. */
- if (100 * (head->sumhash_used + num_new_offsets) > 75 * head->sumhash_size
- || (namehashent->locrec_offset == 0
- && (head->locrectab_used == head->locrectab_size
- || head->string_used + name_len + 1 > head->string_size
- || 100 * head->namehash_used > 75 * head->namehash_size)))
+ if ((100 * (GET (head->sumhash_used) + num_new_offsets)
+ > 75 * GET (head->sumhash_size))
+ || (GET (namehashent->locrec_offset) == 0
+ && (GET (head->locrectab_used) == GET (head->locrectab_size)
+ || (GET (head->string_used) + name_len + 1
+ > GET (head->string_size))
+ || (100 * GET (head->namehash_used)
+ > 75 * GET (head->namehash_size)))))
{
/* The current archive is not large enough. */
enlarge_archive (ah, head);
@@ -1000,20 +1029,20 @@
/* Add the hash value to the hash table. */
md5hval = archive_hashval (data[cnt].sum, 16);
- idx = md5hval % head->sumhash_size;
- incr = 1 + md5hval % (head->sumhash_size - 2);
-
- while (sumhashtab[idx].file_offset != 0)
+ idx = md5hval % GET (head->sumhash_size);
+ incr = 1 + md5hval % (GET (head->sumhash_size) - 2);
+
+ while (GET (sumhashtab[idx].file_offset) != 0)
{
idx += incr;
- if (idx >= head->sumhash_size)
- idx -= head->sumhash_size;
+ if (idx >= GET (head->sumhash_size))
+ idx -= GET (head->sumhash_size);
}
memcpy (sumhashtab[idx].sum, data[cnt].sum, 16);
- sumhashtab[idx].file_offset = file_offsets[cnt];
-
- ++head->sumhash_used;
+ SET (sumhashtab[idx].file_offset, file_offsets[cnt]);
+
+ INC (head->sumhash_used, 1);
}
lastoffset = file_offsets[LC_ALL];
@@ -1024,25 +1053,28 @@
lastoffset += (data[cnt].size + 15) & -16;
}
- if (namehashent->name_offset == 0)
+ if (GET (namehashent->name_offset) == 0)
{
/* Add the name string. */
- memcpy ((char *) ah->addr + head->string_offset + head->string_used,
+ memcpy ((char *) ah->addr + GET (head->string_offset)
+ + GET (head->string_used),
name, name_len + 1);
- namehashent->name_offset = head->string_offset + head->string_used;
- head->string_used += name_len + 1;
- ++head->namehash_used;
- }
-
- if (namehashent->locrec_offset == 0)
+ SET (namehashent->name_offset,
+ GET (head->string_offset) + GET (head->string_used));
+ INC (head->string_used, name_len + 1);
+ INC (head->namehash_used, 1);
+ }
+
+ if (GET (namehashent->locrec_offset == 0))
{
/* Allocate a name location record. */
- namehashent->locrec_offset = (head->locrectab_offset
- + (head->locrectab_used++
- * sizeof (struct locrecent)));
+ SET (namehashent->locrec_offset, (GET (head->locrectab_offset)
+ + (GET (head->locrectab_used)
+ * sizeof (struct locrecent))));
+ INC (head->locrectab_used, 1);
locrecent = (struct locrecent *) ((char *) ah->addr
- + namehashent->locrec_offset);
- locrecent->refs = 1;
+ + GET (namehashent->locrec_offset));
+ SET (locrecent->refs, 1);
}
else
{
@@ -1050,27 +1082,29 @@
we still need a new one. If not, reuse the old one. */
locrecent = (struct locrecent *) ((char *) ah->addr
- + namehashent->locrec_offset);
- if (locrecent->refs > 1)
- {
- --locrecent->refs;
- namehashent->locrec_offset = (head->locrectab_offset
- + (head->locrectab_used++
- * sizeof (struct locrecent)));
- locrecent = (struct locrecent *) ((char *) ah->addr
- + namehashent->locrec_offset);
- locrecent->refs = 1;
+ + GET (namehashent->locrec_offset));
+ if (GET (locrecent->refs) > 1)
+ {
+ INC (locrecent->refs, -1);
+ SET (namehashent->locrec_offset, (GET (head->locrectab_offset)
+ + (GET (head->locrectab_used)
+ * sizeof (struct locrecent))));
+ INC (head->locrectab_used, 1);
+ locrecent
+ = (struct locrecent *) ((char *) ah->addr
+ + GET (namehashent->locrec_offset));
+ SET (locrecent->refs, 1);
}
}
/* Fill in the table with the locations of the locale data. */
for (cnt = 0; cnt < __LC_LAST; ++cnt)
{
- locrecent->record[cnt].offset = file_offsets[cnt];
- locrecent->record[cnt].len = data[cnt].size;
- }
-
- return namehashent->locrec_offset;
+ SET (locrecent->record[cnt].offset, file_offsets[cnt]);
+ SET (locrecent->record[cnt].len, data[cnt].size);
+ }
+
+ return GET (namehashent->locrec_offset);
}
@@ -1129,7 +1163,8 @@
unsigned int strindex[0];
} *filedata = data[LC_CTYPE].addr;
codeset = (char *) filedata
- + filedata->strindex[_NL_ITEM_INDEX (_NL_CTYPE_CODESET_NAME)];
+ + maybe_swap_uint32 (filedata->strindex[_NL_ITEM_INDEX
+ (_NL_CTYPE_CODESET_NAME)]);
char *normalized_codeset_name = NULL;
normalized_codeset = _nl_normalize_codeset (codeset, strlen (codeset));
@@ -1492,7 +1527,7 @@
head = ah.addr;
namehashtab = (struct namehashent *) ((char *) ah.addr
- + head->namehash_offset);
+ + GET (head->namehash_offset));
while (nlist-- > 0)
{
@@ -1504,30 +1539,31 @@
/* Search for this locale in the archive. */
hval = archive_hashval (locname, strlen (locname));
- idx = hval % head->namehash_size;
- incr = 1 + hval % (head->namehash_size - 2);
+ idx = hval % GET (head->namehash_size);
+ incr = 1 + hval % (GET (head->namehash_size) - 2);
/* If the name_offset field is zero this means this is no
deleted entry and therefore no entry can be found. */
- while (namehashtab[idx].name_offset != 0)
- {
- if (namehashtab[idx].hashval == hval
+ while (GET (namehashtab[idx].name_offset) != 0)
+ {
+ if (GET (namehashtab[idx].hashval) == hval
&& (strcmp (locname,
- (char *) ah.addr + namehashtab[idx].name_offset)
+ ((char *) ah.addr
+ + GET (namehashtab[idx].name_offset)))
== 0))
{
/* Found the entry. Now mark it as removed by zero-ing
the reference to the locale record. */
- namehashtab[idx].locrec_offset = 0;
+ SET (namehashtab[idx].locrec_offset, 0);
break;
}
idx += incr;
- if (idx >= head->namehash_size)
- idx -= head->namehash_size;
- }
-
- if (namehashtab[idx].name_offset == 0 && ! be_quiet)
+ if (idx >= GET (head->namehash_size))
+ idx -= GET (head->namehash_size);
+ }
+
+ if (GET (namehashtab[idx].name_offset) == 0 && ! be_quiet)
error (0, 0, _("locale \"%s\" not in archive"), locname);
}
@@ -1590,17 +1626,17 @@
head = ah.addr;
- names = (struct nameent *) xmalloc (head->namehash_used
+ names = (struct nameent *) xmalloc (GET (head->namehash_used)
* sizeof (struct nameent));
namehashtab = (struct namehashent *) ((char *) ah.addr
- + head->namehash_offset);
- for (cnt = used = 0; cnt < head->namehash_size; ++cnt)
- if (namehashtab[cnt].locrec_offset != 0)
+ + GET (head->namehash_offset));
+ for (cnt = used = 0; cnt < GET (head->namehash_size); ++cnt)
+ if (GET (namehashtab[cnt].locrec_offset) != 0)
{
- assert (used < head->namehash_used);
- names[used].name = ah.addr + namehashtab[cnt].name_offset;
- names[used++].locrec_offset = namehashtab[cnt].locrec_offset;
+ assert (used < GET (head->namehash_used));
+ names[used].name = ah.addr + GET (namehashtab[cnt].name_offset);
+ names[used++].locrec_offset = GET (namehashtab[cnt].locrec_offset);
}
/* Sort the names. */
@@ -1612,17 +1648,17 @@
struct sumhashent *sumhashtab;
int sumused;
- files = (struct dataent *) xmalloc (head->sumhash_used
+ files = (struct dataent *) xmalloc (GET (head->sumhash_used)
* sizeof (struct dataent));
sumhashtab = (struct sumhashent *) ((char *) ah.addr
- + head->sumhash_offset);
- for (cnt = sumused = 0; cnt < head->sumhash_size; ++cnt)
- if (sumhashtab[cnt].file_offset != 0)
+ + GET (head->sumhash_offset));
+ for (cnt = sumused = 0; cnt < GET (head->sumhash_size); ++cnt)
+ if (GET (sumhashtab[cnt].file_offset) != 0)
{
- assert (sumused < head->sumhash_used);
+ assert (sumused < GET (head->sumhash_used));
files[sumused].sum = (const unsigned char *) sumhashtab[cnt].sum;
- files[sumused].file_offset = sumhashtab[cnt].file_offset;
+ files[sumused].file_offset = GET (sumhashtab[cnt].file_offset);
files[sumused++].nlink = 0;
}
@@ -1638,18 +1674,19 @@
locrec = (struct locrecent *) ((char *) ah.addr
+ names[cnt].locrec_offset);
for (idx = 0; idx < __LC_LAST; ++idx)
- if (locrec->record[LC_ALL].offset != 0
+ if (GET (locrec->record[LC_ALL].offset) != 0
? (idx == LC_ALL
- || (locrec->record[idx].offset
- < locrec->record[LC_ALL].offset)
- || (locrec->record[idx].offset + locrec->record[idx].len
- > (locrec->record[LC_ALL].offset
- + locrec->record[LC_ALL].len)))
+ || (GET (locrec->record[idx].offset)
+ < GET (locrec->record[LC_ALL].offset))
+ || ((GET (locrec->record[idx].offset)
+ + GET (locrec->record[idx].len))
+ > (GET (locrec->record[LC_ALL].offset)
+ + GET (locrec->record[LC_ALL].len))))
: idx != LC_ALL)
{
struct dataent *data, dataent;
- dataent.file_offset = locrec->record[idx].offset;
+ dataent.file_offset = GET (locrec->record[idx].offset);
data = (struct dataent *) bsearch (&dataent, files, sumused,
sizeof (struct dataent),
dataentcmp);
@@ -1671,21 +1708,24 @@
{
struct dataent *data, dataent;
- dataent.file_offset = locrec->record[idx].offset;
- if (locrec->record[LC_ALL].offset != 0
- && dataent.file_offset >= locrec->record[LC_ALL].offset
- && (dataent.file_offset + locrec->record[idx].len
- <= (locrec->record[LC_ALL].offset
- + locrec->record[LC_ALL].len)))
- dataent.file_offset = locrec->record[LC_ALL].offset;
+ dataent.file_offset = GET (locrec->record[idx].offset);
+ if (GET (locrec->record[LC_ALL].offset) != 0
+ && (dataent.file_offset
+ >= GET (locrec->record[LC_ALL].offset))
+ && (dataent.file_offset + GET (locrec->record[idx].len)
+ <= (GET (locrec->record[LC_ALL].offset)
+ + GET (locrec->record[LC_ALL].len))))
+ dataent.file_offset = GET (locrec->record[LC_ALL].offset);
data = (struct dataent *) bsearch (&dataent, files, sumused,
sizeof (struct dataent),
dataentcmp);
printf ("%6d %7x %3d%c ",
- locrec->record[idx].len, locrec->record[idx].offset,
+ GET (locrec->record[idx].len),
+ GET (locrec->record[idx].offset),
data->nlink,
- dataent.file_offset == locrec->record[LC_ALL].offset
+ (dataent.file_offset
+ == GET (locrec->record[LC_ALL].offset))
? '+' : ' ');
for (i = 0; i < 16; i += 4)
printf ("%02x%02x%02x%02x",
Modified: fsf/trunk/libc/locale/programs/locfile.c
==============================================================================
--- fsf/trunk/libc/locale/programs/locfile.c (original)
+++ fsf/trunk/libc/locale/programs/locfile.c Wed Sep 25 00:02:24 2013
@@ -22,6 +22,7 @@
#include <dirent.h>
#include <errno.h>
#include <fcntl.h>
+#include <stdbool.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
@@ -538,6 +539,10 @@
return ret;
}
+/* True if the locale files use the opposite endianness to the
+ machine running localedef. */
+bool swap_endianness_p;
+
/* When called outside a start_locale_structure/end_locale_structure
or start_locale_prelude/end_locale_prelude block, record that the
next byte in FILE's obstack will be the first byte of a new element.
@@ -624,6 +629,7 @@
{
align_locale_data (file, sizeof (uint32_t));
record_offset (file);
+ value = maybe_swap_uint32 (value);
obstack_grow (&file->data, &value, sizeof (value));
}
@@ -636,6 +642,7 @@
align_locale_data (file, sizeof (uint32_t));
record_offset (file);
obstack_grow (&file->data, data, n_elems * sizeof (uint32_t));
+ maybe_swap_uint32_obstack (&file->data, n_elems);
}
/* Record that FILE's next element is the single byte given by VALUE. */
@@ -708,6 +715,8 @@
vec[1].iov_base = file->offsets;
vec[2].iov_len = obstack_object_size (&file->data);
vec[2].iov_base = obstack_finish (&file->data);
+ maybe_swap_uint32_array (vec[0].iov_base, 2);
+ maybe_swap_uint32_array (vec[1].iov_base, file->n_elements);
n_elem = 3;
if (! no_archive)
{
Modified: fsf/trunk/libc/locale/programs/locfile.h
==============================================================================
--- fsf/trunk/libc/locale/programs/locfile.h (original)
+++ fsf/trunk/libc/locale/programs/locfile.h Wed Sep 25 00:02:24 2013
@@ -18,6 +18,8 @@
#ifndef _LOCFILE_H
#define _LOCFILE_H 1
+#include <byteswap.h>
+#include <stdbool.h>
#include <stdint.h>
#include <sys/uio.h>
@@ -66,6 +68,41 @@
const struct charmap_t *charmap,
const char *locname,
const char *output_path);
+
+extern bool swap_endianness_p;
+
+/* Change the output to be big-endian if BIG_ENDIAN is true and
+ little-endian otherwise. */
+static inline void
+set_big_endian (bool big_endian)
+{
+ swap_endianness_p = (big_endian != (__BYTE_ORDER == __BIG_ENDIAN));
+}
+
+/* Munge VALUE so that, when stored, it has the correct byte order
+ for the output files. */
+static inline uint32_t
+maybe_swap_uint32 (uint32_t value)
+{
+ return swap_endianness_p ? bswap_32 (value) : value;
+}
+
+/* Likewise, but munge an array of N uint32_ts starting at ARRAY. */
+static inline void
+maybe_swap_uint32_array (uint32_t *array, size_t n)
+{
+ if (swap_endianness_p)
+ while (n-- > 0)
+ array[n] = bswap_32 (array[n]);
+}
+
+/* Like maybe_swap_uint32_array, but the array of N elements is at
+ the end of OBSTACK's current object. */
+static inline void
+maybe_swap_uint32_obstack (struct obstack *obstack, size_t n)
+{
+ maybe_swap_uint32_array ((uint32_t *) obstack_next_free (obstack) - n, n);
+}
/* Write out the data. */
extern void init_locale_data (struct locale_file *file, size_t n_elements);
Modified: fsf/trunk/libc/localedata/ChangeLog
==============================================================================
--- fsf/trunk/libc/localedata/ChangeLog (original)
+++ fsf/trunk/libc/localedata/ChangeLog Wed Sep 25 00:02:24 2013
@@ -1,10 +1,22 @@
-2013-09-21 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+2013-09-23 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+
+ [BZ#15609]
+ * locales/cmn_TW: New locale for cmn_TW.
+ * SUPPORTED: Add cmn_TW.
+
+2013-09-23 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
[BZ#15608]
- * locales/nan_TW: New locale for lzh_TW.
+ * locales/hak_TW: New locale for hak_TW.
+ * SUPPORTED: Add hak_TW.
+
+2013-09-23 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+
+ [BZ#15608]
+ * locales/lzh_TW: New locale for lzh_TW.
* SUPPORTED: Add lzh_TW.
-2013-09-21 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+2013-09-23 Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
[BZ#15608]
* locales/nan_TW: New locale for nan_TW.
Modified: fsf/trunk/libc/localedata/SUPPORTED
==============================================================================
--- fsf/trunk/libc/localedata/SUPPORTED (original)
+++ fsf/trunk/libc/localedata/SUPPORTED Wed Sep 25 00:02:24 2013
@@ -81,6 +81,7 @@
ca_FR/ISO-8859-15 \
ca_IT.UTF-8/UTF-8 \
ca_IT/ISO-8859-15 \
+cmn_TW/UTF-8 \
crh_UA/UTF-8 \
cs_CZ.UTF-8/UTF-8 \
cs_CZ/ISO-8859-2 \
@@ -230,6 +231,7 @@
gv_GB.UTF-8/UTF-8 \
gv_GB/ISO-8859-1 \
ha_NG/UTF-8 \
+hak_TW/UTF-8 \
he_IL.UTF-8/UTF-8 \
he_IL/ISO-8859-8 \
hi_IN/UTF-8 \
Added: fsf/trunk/libc/localedata/locales/cmn_TW
==============================================================================
--- fsf/trunk/libc/localedata/locales/cmn_TW (added)
+++ fsf/trunk/libc/localedata/locales/cmn_TW Wed Sep 25 00:02:24 2013
@@ -1,0 +1,220 @@
+comment_char %
+escape_char /
+% Charset: UTF-8
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% Mandarin Chinese locale for the Republic of China
+%
+% Prepared and contributed to glibc by Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+%
+% build with: localedef -f UTF-8 -i cmn_TW cmn_TW
+%
+% This file is a part of GNU C Library (glibc) and contains locale data. The
+% Free Software Foundation does not claim any copyright interest in the
+% locale data contained in this file. The foregoing does not affect the
+% license of GNU C Library (glibc) as a whole. It does not exempt you from the
+% conditions of the license if your use would otherwise be governed by
+% that license.
+%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+LC_IDENTIFICATION
+title "Mandarin Chinese locale for the Republic of China"
+source ""
+address ""
+contact ""
+email "bug-glibc-locales@xxxxxxx"
+tel ""
+fax ""
+language "Mandarin Chinese"
+territory "Republic of China"
+revision "0.1"
+date "2013-06-02"
+%
+category "cmn_TW:2000";LC_IDENTIFICATION
+category "cmn_TW:2000";LC_CTYPE
+category "cmn_TW:2000";LC_COLLATE
+category "cmn_TW:2000";LC_TIME
+category "cmn_TW:2000";LC_NUMERIC
+category "cmn_TW:2000";LC_MONETARY
+category "cmn_TW:2000";LC_PAPER
+category "cmn_TW:2000";LC_MEASUREMENT
+category "cmn_TW:2000";LC_MESSAGES
+category "cmn_TW:2000";LC_NAME
+category "cmn_TW:2000";LC_ADDRESS
+category "cmn_TW:2000";LC_TELEPHONE
+END LC_IDENTIFICATION
+
+LC_CTYPE
+copy "i18n"
+class "hanzi"; /
+<U3007>;/
+<U3400>..<U4DBF>;/
+<U4E00>..<U9FA5>;/
+<UF900>..<UFA6A>;/
+<U00020000>..<U0002A6D6>;/
+<U0002F800>..<U0002FA1D>
+END LC_CTYPE
+
+LC_COLLATE
+copy "iso14651_t1"
+END LC_COLLATE
+
+LC_TIME
+% ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ
+mon "<U4E00><U6708>";/
+ "<U4E8C><U6708>";/
+ "<U4E09><U6708>";/
+ "<U56DB><U6708>";/
+ "<U4E94><U6708>";/
+ "<U516D><U6708>";/
+ "<U4E03><U6708>";/
+ "<U516B><U6708>";/
+ "<U4E5D><U6708>";/
+ "<U5341><U6708>";/
+ "<U5341><U4E00><U6708>";/
+ "<U5341><U4E8C><U6708>"
+% 1ÃÂÂ, 2ÃÂÂ, 3ÃÂÂ, 4ÃÂÂ, 5ÃÂÂ, 6ÃÂÂ, 7ÃÂÂ, 8ÃÂÂ, 9ÃÂÂ, 10ÃÂÂ, 11ÃÂÂ, 12ÃÂÂ
+abmon "<U0020><U0031><U6708>";/
+ "<U0020><U0032><U6708>";/
+ "<U0020><U0033><U6708>";/
+ "<U0020><U0034><U6708>";/
+ "<U0020><U0035><U6708>";/
+ "<U0020><U0036><U6708>";/
+ "<U0020><U0037><U6708>";/
+ "<U0020><U0038><U6708>";/
+ "<U0020><U0039><U6708>";/
+ "<U0031><U0030><U6708>";/
+ "<U0031><U0031><U6708>";/
+ "<U0031><U0032><U6708>"
+% ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ
+day "<U661F><U671F><U65E5>";/
+ "<U661F><U671F><U4E00>";/
+ "<U661F><U671F><U4E8C>";/
+ "<U661F><U671F><U4E09>";/
+ "<U661F><U671F><U56DB>";/
+ "<U661F><U671F><U4E94>";/
+ "<U661F><U671F><U516D>"
+% ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ
+abday "<U65E5>";/
+ "<U4E00>";/
+ "<U4E8C>";/
+ "<U4E09>";/
+ "<U56DB>";/
+ "<U4E94>";/
+ "<U516D>"
+% %YÃÂÂ%mÃÂÂ%dÃÂÂ (%A) %HÃÂÂ%MÃÂÂ%SÃÂÂ
+d_t_fmt "<U0025><U0059><U5E74><U0025><U006D><U6708><U0025><U0064><U65E5><U0020><U0028><U0025><U0041><U0029><U0020><U0025><U0048><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% %YÃÂÂ%mÃÂÂ%dÃÂÂ
+d_fmt "<U0025><U0059><U5E74><U0025><U006D><U6708><U0025><U0064><U65E5>"
+% %HÃÂÂ%MÃÂÂ%SÃÂÂ
+t_fmt "<U0025><U0048><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% ÃÂÂÃÂÂ, ÃÂÂÃÂÂ
+am_pm "<U4E0A><U5348>";/
+ "<U4E0B><U5348>"
+% %p %IÃÂÂ%MÃÂÂ%SÃÂÂ
+t_fmt_ampm "<U0025><U0070><U0020><U0025><U0049><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% %YÃÂÂ %b %eÃÂÂ %A %H:%M:%S %Z
+date_fmt "<U0025><U0059><U5E74><U0020><U0025><U0062><U0020><U0025><U0065><U65E5><U0020><U0025><U0041><U0020><U0025><U0048><U003A><U0025><U004D><U003A><U0025><U0053><U0020><U0025><U005A>"
+END LC_TIME
+
+LC_NUMERIC
+% .
+decimal_point "<U002E>"
+% ,
+thousands_sep "<U002C>"
+grouping 4
+END LC_NUMERIC
+
+LC_MONETARY
+% NT$
+currency_symbol "<U004E><U0054><U0024>"
+% TWD
+int_curr_symbol "<U0054><U0057><U0044><U0020>"
+% .
+mon_decimal_point "<U002E>"
+% ,
+mon_thousands_sep "<U002C>"
+mon_grouping 4
+positive_sign ""
+% -
+negative_sign "<U002D>"
+frac_digits 2
+int_frac_digits 2
+p_cs_precedes 1
+p_sep_by_space 0
+n_cs_precedes 1
+n_sep_by_space 0
+p_sign_posn 1
+n_sign_posn 1
+int_p_cs_precedes 1
+int_p_sep_by_space 0
+int_n_cs_precedes 1
+int_n_sep_by_space 0
+int_p_sign_posn 1
+int_n_sign_posn 1
+END LC_MONETARY
+
+LC_PAPER
+height 297
+width 210
+END LC_PAPER
+
+LC_MEASUREMENT
+measurement 1
+END LC_MEASUREMENT
+
+LC_MESSAGES
+% ^[yYÃÂÂ]
+yesexpr "<U005E><U005B><U0079><U0059><U662F><U005D>"
+% ^[nNÃÂÂÃÂÂ]
+noexpr "<U005E><U005B><U006E><U004E><U4E0D><U5426><U005D>"
+END LC_MESSAGES
+
+LC_NAME
+% %f%t%g%t%d
+name_fmt "<U0025><U0066><U0025><U0074><U0025><U0067><U0025><U0074><U0025><U0064>"
+% ÃÂÂ
+name_gen "<U541B>"
+% ÃÂÂÃÂÂ
+name_miss "<U5C0F><U59D0>"
+% ÃÂÂÃÂÂ
+name_mr "<U5148><U751F>"
+% ÃÂÂÃÂÂ
+name_mrs "<U592B><U4EBA>"
+% ÃÂÂÃÂÂ
+name_ms "<U5973><U58EB>"
+END LC_NAME
+
+LC_ADDRESS
+% %c%N%T%N%s %h %e %r%N%b%N%d%N%f%N%a%N
+postal_fmt "<U0025><U0063><U0025><U004E><U0025><U0054><U0025><U004E><U0025><U0073><U0020><U0025><U0068><U0020><U0025><U0065><U0020><U0025><U0072><U0025><U004E><U0025><U0062><U0025><U004E><U0025><U0064><U0025><U004E><U0025><U0066><U0025><U004E><U0025><U0061><U0025><U004E>"
+% ÃÂÂÃÂÂÃÂÂÃÂÂ
+country_name "<U4E2D><U83EF><U6C11><U570B>"
+% TW
+country_post "<U0054><U0057>"
+% TW
+country_ab2 "<U0054><U0057>"
+% TWN
+country_ab3 "<U0054><U0057><U004E>"
+country_num 158
+country_isbn 957
+% ÃÂÂÃÂÂÃÂÂÃÂÂ
+lang_name "<U6F22><U8A9E><U5B98><U8A71>"
+% cmn
+lang_term "<U0063><U006D><U006E>"
+% cmn
+lang_lib "<U0063><U006D><U006E>"
+END LC_ADDRESS
+
+LC_TELEPHONE
+% +%c-%a-%l
+tel_int_fmt "<U002B><U0025><U0063><U002D><U0025><U0061><U002D><U0025><U006C>"
+% %A-%l
+tel_dom_fmt "<U0025><U0041><U002D><U0025><U006C>"
+% 00
+int_select "<U0030><U0030>"
+% 886
+int_prefix "<U0038><U0038><U0036>"
+END LC_TELEPHONE
Added: fsf/trunk/libc/localedata/locales/hak_TW
==============================================================================
--- fsf/trunk/libc/localedata/locales/hak_TW (added)
+++ fsf/trunk/libc/localedata/locales/hak_TW Wed Sep 25 00:02:24 2013
@@ -1,0 +1,219 @@
+comment_char %
+escape_char /
+% Charset: UTF-8
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% Hakka Chinese locale for the Republic of China
+%
+% Prepared and contributed to glibc by Wei-Lun Chao <bluebat@xxxxxxxxxxxxxx>
+%
+% build with: localedef -f UTF-8 -i hak_TW hak_TW
+%
+% This file is a part of GNU C Library (glibc) and contains locale data. The
+% Free Software Foundation does not claim any copyright interest in the
+% locale data contained in this file. The foregoing does not affect the
+% license of GNU C Library (glibc) as a whole. It does not exempt you from the
+% conditions of the license if your use would otherwise be governed by
+% that license.
+%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+LC_IDENTIFICATION
+title "Hakka Chinese locale for the Republic of China"
+source ""
+address ""
+contact ""
+email "bug-glibc-locales@xxxxxxx"
+tel ""
+fax ""
+language "Hakka Chinese"
+territory "Republic of China"
+revision "0.1"
+date "2013-06-02"
+%
+category "hak_TW:2000";LC_IDENTIFICATION
+category "hak_TW:2000";LC_CTYPE
+category "hak_TW:2000";LC_COLLATE
+category "hak_TW:2000";LC_TIME
+category "hak_TW:2000";LC_NUMERIC
+category "hak_TW:2000";LC_MONETARY
+category "hak_TW:2000";LC_PAPER
+category "hak_TW:2000";LC_MEASUREMENT
+category "hak_TW:2000";LC_MESSAGES
+category "hak_TW:2000";LC_NAME
+category "hak_TW:2000";LC_ADDRESS
+category "hak_TW:2000";LC_TELEPHONE
+END LC_IDENTIFICATION
+
+LC_CTYPE
+copy "i18n"
+class "hanzi"; /
+<U3007>;/
+<U3400>..<U4DBF>;/
+<U4E00>..<U9FA5>;/
+<UF900>..<UFA6A>;/
+<U00020000>..<U0002A6D6>;/
+<U0002F800>..<U0002FA1D>
+END LC_CTYPE
+
+LC_COLLATE
+copy "iso14651_t1"
+END LC_COLLATE
+
+LC_TIME
+% ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ
+mon "<U4E00><U6708>";/
+ "<U4E8C><U6708>";/
+ "<U4E09><U6708>";/
+ "<U56DB><U6708>";/
+ "<U4E94><U6708>";/
+ "<U516D><U6708>";/
+ "<U4E03><U6708>";/
+ "<U516B><U6708>";/
+ "<U4E5D><U6708>";/
+ "<U5341><U6708>";/
+ "<U5341><U4E00><U6708>";/
+ "<U5341><U4E8C><U6708>"
+% 1ÃÂÂ, 2ÃÂÂ, 3ÃÂÂ, 4ÃÂÂ, 5ÃÂÂ, 6ÃÂÂ, 7ÃÂÂ, 8ÃÂÂ, 9ÃÂÂ, 10ÃÂÂ, 11ÃÂÂ, 12ÃÂÂ
+abmon "<U0020><U0031><U6708>";/
+ "<U0020><U0032><U6708>";/
+ "<U0020><U0033><U6708>";/
+ "<U0020><U0034><U6708>";/
+ "<U0020><U0035><U6708>";/
+ "<U0020><U0036><U6708>";/
+ "<U0020><U0037><U6708>";/
+ "<U0020><U0038><U6708>";/
+ "<U0020><U0039><U6708>";/
+ "<U0031><U0030><U6708>";/
+ "<U0031><U0031><U6708>";/
+ "<U0031><U0032><U6708>"
+% ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ, ÃÂÂÃÂÂÃÂÂ
+day "<U79AE><U62DC><U65E5>";/
+ "<U79AE><U62DC><U4E00>";/
+ "<U79AE><U62DC><U4E8C>";/
+ "<U79AE><U62DC><U4E09>";/
+ "<U79AE><U62DC><U56DB>";/
+ "<U79AE><U62DC><U4E94>";/
+ "<U79AE><U62DC><U516D>"
+% ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ, ÃÂÂ
+abday "<U65E5>";/
+ "<U4E00>";/
+ "<U4E8C>";/
+ "<U4E09>";/
+ "<U56DB>";/
+ "<U4E94>";/
+ "<U516D>"
+% %YÃÂÂ%mÃÂÂ%dÃÂÂ (%A) %HÃÂÂ%MÃÂÂ%SÃÂÂ
+d_t_fmt "<U0025><U0059><U5E74><U0025><U006D><U6708><U0025><U0064><U65E5><U0020><U0028><U0025><U0041><U0029><U0020><U0025><U0048><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% %YÃÂÂ%mÃÂÂ%dÃÂÂ
+d_fmt "<U0025><U0059><U5E74><U0025><U006D><U6708><U0025><U0064><U65E5>"
+% %HÃÂÂ%MÃÂÂ%SÃÂÂ
+t_fmt "<U0025><U0048><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% ÃÂÂÃÂÂ, ÃÂÂÃÂÂ
+am_pm "<U4E0A><U665D>";"<U4E0B><U665D>"
+% %p %IÃÂÂ%MÃÂÂ%SÃÂÂ
+t_fmt_ampm "<U0025><U0070><U0020><U0025><U0049><U9EDE><U0025><U004D><U5206><U0025><U0053><U79D2>"
+% %YÃÂÂ %b %eÃÂÂ %A %H:%M:%S %Z
+date_fmt "<U0025><U0059><U5E74><U0020><U0025><U0062><U0020><U0025><U0065><U65E5><U0020><U0025><U0041><U0020><U0025><U0048><U003A><U0025><U004D><U003A><U0025><U0053><U0020><U0025><U005A>"
+END LC_TIME
+
+LC_NUMERIC
+% .
+decimal_point "<U002E>"
+% ,
+thousands_sep "<U002C>"
+grouping 4
+END LC_NUMERIC
+
+LC_MONETARY
+% NT$
+currency_symbol "<U004E><U0054><U0024>"
+% TWD
+int_curr_symbol "<U0054><U0057><U0044><U0020>"
+% .
+mon_decimal_point "<U002E>"
+% ,
+mon_thousands_sep "<U002C>"
+mon_grouping 4
+positive_sign ""
+% -
+negative_sign "<U002D>"
+frac_digits 2
+int_frac_digits 2
+p_cs_precedes 1
+p_sep_by_space 0
+n_cs_precedes 1
+n_sep_by_space 0
+p_sign_posn 1
+n_sign_posn 1
+int_p_cs_precedes 1
+int_p_sep_by_space 0
+int_n_cs_precedes 1
+int_n_sep_by_space 0
+int_p_sign_posn 1
+int_n_sign_posn 1
+END LC_MONETARY
+
+LC_PAPER
+height 297
+width 210
+END LC_PAPER
+
+LC_MEASUREMENT
+measurement 1
+END LC_MEASUREMENT
+
+LC_MESSAGES
+% ^[yYÃÂÂ]
+yesexpr "<U005E><U005B><U0079><U0059><U4FC2><U005D>"
+% ^[nNÃÂÂ]
+noexpr "<U005E><U005B><U006E><U004E><U6BCB><U005D>"
+END LC_MESSAGES
+
+LC_NAME
+% %f%t%g%t%d
+name_fmt "<U0025><U0066><U0025><U0074><U0025><U0067><U0025><U0074><U0025><U0064>"
+% ÃÂÂ
+name_gen "<U541B>"
+% ÃÂÂÃÂÂ
+name_miss "<U5C0F><U59D0>"
+% ÃÂÂÃÂÂ
+name_mr "<U5148><U751F>"
+% ÃÂÂÃÂÂ
+name_mrs "<U592B><U4EBA>"
+% ÃÂÂÃÂÂ
+name_ms "<U5973><U58EB>"
+END LC_NAME
+
+LC_ADDRESS
+% %c%N%T%N%s %h %e %r%N%b%N%d%N%f%N%a%N
+postal_fmt "<U0025><U0063><U0025><U004E><U0025><U0054><U0025><U004E><U0025><U0073><U0020><U0025><U0068><U0020><U0025><U0065><U0020><U0025><U0072><U0025><U004E><U0025><U0062><U0025><U004E><U0025><U0064><U0025><U004E><U0025><U0066><U0025><U004E><U0025><U0061><U0025><U004E>"
+% ÃÂÂÃÂÂÃÂÂÃÂÂ
+country_name "<U4E2D><U83EF><U6C11><U570B>"
+% TW
+country_post "<U0054><U0057>"
+% TW
+country_ab2 "<U0054><U0057>"
+% TWN
+country_ab3 "<U0054><U0057><U004E>"
+country_num 158
+country_isbn 957
+% ÃÂÂÃÂÂÃÂÂÃÂÂÃÂÂ
+lang_name "<U6F22><U8A9E><U5BA2><U5BB6><U8A9E>"
+% hak
+lang_term "<U0068><U0061><U006B>"
+% hak
+lang_lib "<U0068><U0061><U006B>"
+END LC_ADDRESS
+
+LC_TELEPHONE
+% +%c-%a-%l
+tel_int_fmt "<U002B><U0025><U0063><U002D><U0025><U0061><U002D><U0025><U006C>"
+% %A-%l
+tel_dom_fmt "<U0025><U0041><U002D><U0025><U006C>"
+% 00
+int_select "<U0030><U0030>"
+% 886
+int_prefix "<U0038><U0038><U0036>"
+END LC_TELEPHONE
Modified: fsf/trunk/libc/manual/freemanuals.texi
==============================================================================
--- fsf/trunk/libc/manual/freemanuals.texi (original)
+++ fsf/trunk/libc/manual/freemanuals.texi Wed Sep 25 00:02:24 2013
@@ -1,4 +1,7 @@
-@appendix Free Software Needs Free Documentation
+@c freemanuals.texi - blurb for free documentation.
+@c This file is intended to be included within another document,
+@c hence no sectioning command or @node.
+
@cindex free documentation
The biggest deficiency in the free software community today is not in
Modified: fsf/trunk/libc/manual/libc.texinfo
==============================================================================
--- fsf/trunk/libc/manual/libc.texinfo (original)
+++ fsf/trunk/libc/manual/libc.texinfo Wed Sep 25 00:02:24 2013
@@ -118,6 +118,7 @@
@include chapters.texi
@node Free Manuals, Copying, Contributors, Top
+@appendix Free Software Needs Free Documentation
@include freemanuals.texi
@node Copying, Documentation License, Free Manuals, Top
Modified: fsf/trunk/libc/ports/ChangeLog.aarch64
==============================================================================
--- fsf/trunk/libc/ports/ChangeLog.aarch64 (original)
+++ fsf/trunk/libc/ports/ChangeLog.aarch64 Wed Sep 25 00:02:24 2013
@@ -1,3 +1,12 @@
+2013-09-24 Venkataramanan Kumar <venkataramanan.kumar@xxxxxxxxxx>
+
+ * sysdeps/aarch64/machine-gmon.h (__mcount): Accept parameter and
+ pass it to mcount_internal.
+
+2013-09-24 Venkataramanan Kumar <venkataramanan.kumar@xxxxxxxxxx>
+
+ * sysdeps/unix/sysv/linux/aarch64/sigcontextinfo.h: New file.
+
2013-07-26 Marcus Shawcroft <marcus.shawcroft@xxxxxxxxxx>
* sysdeps/aarch64/Versions: New file.
Modified: fsf/trunk/libc/ports/ChangeLog.arm
==============================================================================
--- fsf/trunk/libc/ports/ChangeLog.arm (original)
+++ fsf/trunk/libc/ports/ChangeLog.arm Wed Sep 25 00:02:24 2013
@@ -1,3 +1,8 @@
+2013-09-24 Will Newton <will.newton@xxxxxxxxxx>
+
+ * ports/sysdeps/arm/nptl/tls.h (TLS_INIT_TP_EXPENSIVE): Remove
+ macro.
+
2013-09-20 Andreas Schwab <schwab@xxxxxxxxxxxxxx>
* sysdeps/arm/armv6t2/ffs.S (__ffs): Define as hidden.
Modified: fsf/trunk/libc/ports/sysdeps/aarch64/machine-gmon.h
==============================================================================
--- fsf/trunk/libc/ports/sysdeps/aarch64/machine-gmon.h (original)
+++ fsf/trunk/libc/ports/sysdeps/aarch64/machine-gmon.h Wed Sep 25 00:02:24 2013
@@ -16,10 +16,9 @@
License along with the GNU C Library; if not, see
<http://www.gnu.org/licenses/>. */
-/* GCC version 2 gives us a perfect magical function to get
- just the information we need:
- void *__builtin_return_address (unsigned int N)
- returns the return address of the frame N frames up. */
+/* Accept 'frompc' address as argument from the function that calls
+ __mcount for profiling. Use __builtin_return_address (0)
+ for the 'selfpc' address. */
#include <sysdep.h>
@@ -28,8 +27,8 @@
#define _MCOUNT_DECL(frompc, selfpc) \
static inline void mcount_internal (u_long frompc, u_long selfpc)
-#define MCOUNT \
-void __mcount (void) \
-{ \
- mcount_internal ((u_long) RETURN_ADDRESS (1), (u_long) RETURN_ADDRESS (0)); \
+#define MCOUNT \
+void __mcount (void *frompc) \
+{ \
+ mcount_internal ((u_long) frompc, (u_long) RETURN_ADDRESS (0)); \
}
Modified: fsf/trunk/libc/ports/sysdeps/arm/nptl/tls.h
==============================================================================
--- fsf/trunk/libc/ports/sysdeps/arm/nptl/tls.h (original)
+++ fsf/trunk/libc/ports/sysdeps/arm/nptl/tls.h Wed Sep 25 00:02:24 2013
@@ -119,10 +119,6 @@
#define THREAD_SETMEM_NC(descr, member, idx, value) \
descr->member[idx] = (value)
-/* Initializing the thread pointer will generate a SIGILL if the syscall
- is not available. */
-#define TLS_INIT_TP_EXPENSIVE 1
-
/* Get and set the global scope generation counter in struct pthread. */
#define THREAD_GSCOPE_FLAG_UNUSED 0
#define THREAD_GSCOPE_FLAG_USED 1
Added: fsf/trunk/libc/ports/sysdeps/unix/sysv/linux/aarch64/sigcontextinfo.h
==============================================================================
--- fsf/trunk/libc/ports/sysdeps/unix/sysv/linux/aarch64/sigcontextinfo.h (added)
+++ fsf/trunk/libc/ports/sysdeps/unix/sysv/linux/aarch64/sigcontextinfo.h Wed Sep 25 00:02:24 2013
@@ -1,0 +1,35 @@
+/* AArch64 definitions for signal handling calling conventions.
+ Copyright (C) 1996-2013 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; if not, see
+ <http://www.gnu.org/licenses/>. */
+
+#include <sys/ucontext.h>
+#include "kernel-features.h"
+
+#define SIGCONTEXT siginfo_t *_si, struct ucontext *
+#define GET_PC(ctx) ((void *) (ctx)->uc_mcontext.pc)
+
+/* There is no reliable way to get the sigcontext unless we use a
+ three-argument signal handler. */
+#define __sigaction(sig, act, oact) ({ \
+ (act)->sa_flags |= SA_SIGINFO; \
+ (__sigaction) (sig, act, oact); \
+})
+
+#define sigaction(sig, act, oact) ({ \
+ (act)->sa_flags |= SA_SIGINFO; \
+ (sigaction) (sig, act, oact); \
+})
Modified: fsf/trunk/libc/stdio-common/tst-long-dbl-fphex.c
==============================================================================
--- fsf/trunk/libc/stdio-common/tst-long-dbl-fphex.c (original)
+++ fsf/trunk/libc/stdio-common/tst-long-dbl-fphex.c Wed Sep 25 00:02:24 2013
@@ -28,9 +28,9 @@
int result = 0;
const long double x = 24.5;
wchar_t a[16];
- swprintf (a, sizeof (a), L"%La\n", x);
+ swprintf (a, sizeof a / sizeof a[0], L"%La\n", x);
wchar_t A[16];
- swprintf (A, sizeof (A) / sizeof (A[0]), L"%LA\n", x);
+ swprintf (A, sizeof A / sizeof A[0], L"%LA\n", x);
/* Here wprintf can return four valid variants. We must accept all
of them. */
Modified: fsf/trunk/libc/sysdeps/mach/hurd/i386/tls.h
==============================================================================
--- fsf/trunk/libc/sysdeps/mach/hurd/i386/tls.h (original)
+++ fsf/trunk/libc/sysdeps/mach/hurd/i386/tls.h Wed Sep 25 00:02:24 2013
@@ -118,7 +118,6 @@
operation can cause a failure 'errno' must not be touched. */
# define TLS_INIT_TP(descr, secondcall) \
_hurd_tls_init ((tcbhead_t *) (descr), (secondcall))
-# define TLS_INIT_TP_EXPENSIVE 1
/* Return the TCB address of the current thread. */
# define THREAD_SELF \
_______________________________________________
Commits mailing list
Commits@xxxxxxxxxx
http://eglibc.org/cgi-bin/mailman/listinfo/commits