David S. Miller
ae251b0b58
Fix underflow generation in soft-fp.
...
* soft-fp/soft-fp.h (FP_CUR_EXCEPTIONS): Define.
(FP_TRAPPING_EXCEPTIONS): Provide default implementation.
* sysdeps/sparc/sparc32/soft-fp/sfp-machine.h
(FP_TRAPPING_EXCEPTIONS): Define.
* sysdeps/sparc/sparc64/soft-fp/sfp-machine.h
(FP_TRAPPING_EXCEPTIONS): Define.
* soft-fp/op-common.h (_FP_PACK_SEMIRAW): Signal underflow for
subnormals only when inexact has been signalled or underflow
exceptions are enabled.
(_FP_PACK_CANONICAL): Likewise.
2012-05-30 13:41:01 -07:00
H.J. Lu
d5c9086734
Use __UWORD_TYPE for __NLINK_T_TYPE with -m32
2012-05-30 11:35:47 -07:00
Richard Henderson
8d8f2279e7
alpha: Don't symbol_version syscalls outside libc
...
* sysdeps/unix/make-syscalls.sh: Protect symbol_version output
with #ifndef NOT_IN_libc.
2012-05-30 11:05:51 -07:00
Richard Henderson
63f1549e0e
alpha: Handle ST_OTHER in objdump output
...
* scripts/abilist.awk: Accept 8 fields. Handle Alpha functions
marked to avoid plt entry.
2012-05-30 11:05:51 -07:00
H.J. Lu
0ab0291b84
Convert WORDSIZE[32|64]/ld entries to abi-variants
2012-05-30 08:33:26 -07:00
Joseph Myers
3a85279c0b
Remove __ASSUME_TRUNCATE64_SYSCALL.
2012-05-30 14:52:34 +00:00
Joseph Myers
4da0431d5f
Weaken PIE configure test to a compile test.
2012-05-30 09:50:40 +00:00
Andreas Schwab
865eac65c2
Move CL entry
2012-05-30 08:49:59 +02:00
Andreas Schwab
4d17e68350
Remove use of INTDEF/INTUSE in nptl
2012-05-30 00:45:53 +02:00
Andreas Krebbel
11ef492c80
S/390: ABI baseline update.
2012-05-29 11:04:22 +00:00
David S. Miller
d66ef399f5
Fix underflow reporting and tie up loose ends in sparc soft-fp.
...
* sysdeps/sparc/sparc32/soft-fp/q_util.c (___Q_numbers): Delete.
(___Q_zero): New.
(__Q_simulate_exceptions): Return void. Change to simulate
exceptions by writing into the %fsr.
* sysdeps/sparc/sparc64/soft-fp/qp_util.c
(__Qp_handle_exceptions): Likewise.
(numbers): Delete.
* sysdeps/sparc/sparc64/soft-fp/Versions: Remove entry for
__Qp_handle_exceptions.
* sysdeps/unix/sysv/linux/sparc/sparc64/nptl/libc.abilist: Remove
__Qp_handle_exceptions.
* sysdeps/sparc/sparc32/soft-fp/sfp-machine.h (_FP_DECL_EX): Mark
as unused and give dummy FP_RND_NEAREST initializer.
(FP_INHIBIT_RESULTS): Define.
(___Q_simulate_exceptions): Update declaration.
(FP_HANDLE_EXCEPTIONS): Use ___Q_zero and tidy inline asm
formatting.
* sysdeps/sparc/sparc64/soft-fp/sfp-machine.h (_FP_DECL_EX): Mark
as unused and give dummy FP_RND_NEAREST initializer.
(__Qp_handle_exceptions): Update declaration.
(FP_HANDLE_EXCEPTIONS, QP_NO_EXCEPTIONS): Tidy inline asm
formatting.
2012-05-27 21:11:42 -07:00
Thomas Schwinge
04fb54b507
SH: Use fpu_control_t for the temporary FPU control word.
2012-05-28 01:03:18 +02:00
Kaz Kojima
3f99608f82
SH: feraiseexcept: Restore fpscr flag fields.
2012-05-28 01:02:34 +02:00
Chung-Lin Tang
65a4de4e06
SH: Add CFI directives.
2012-05-28 00:11:06 +02:00
Ulrich Drepper
15749d40db
New Croatian translation
2012-05-27 03:00:50 -04:00
Andreas Schwab
3d3f8e5582
Fix missing declaration
2012-05-26 22:33:51 +02:00
Marek Polacek
7ad47a8080
Fix CL date.
2012-05-26 14:41:28 +02:00
Andreas Schwab
25dbcb277a
Optimize handling of denormals in logb/logbf/logbl
2012-05-26 13:53:22 +02:00
Marek Polacek
d81dcb3569
Fix fma test for underflow.
2012-05-26 13:49:49 +02:00
Siddhesh Poyarekar
9c6ea9facb
Fix stack size and address inconsistency due to executable stack
...
When a stack is marked executable due to loading a DSO that requires
an executable stack, the logic tends to leave out a portion of stack
after the first frame, thus causing a difference in the value returned
by pthread_getattr_np before and after the stack is marked
executable. It ought to be possible to fix this by marking the rest of
the stack as executable too, but in the interest of marking as less of
the stack as executable as possible, the path this fix takes is to
make pthread_getattr_np also look at the first frame as the underflow
end of the stack and compute size and stack top accordingly.
The above happens only for the main process stack. NPTL thread stacks
are not affected by this change.
2012-05-26 09:48:26 +05:30
Joseph Myers
a82392228a
Remove __ASSUME_NEW_GETRLIMIT_SYSCALL.
2012-05-25 22:30:06 +00:00
H.J. Lu
918b56067a
Don't use header files in glibc configure
2012-05-25 14:16:19 -07:00
Roland McGrath
8422c9a560
Add systemtap static probe points in setjmp/longjmp on x86.
2012-05-25 13:41:03 -07:00
Roland McGrath
3a097cc7a1
Add --enable-systemtap configuration to define static probe points.
2012-05-25 13:40:20 -07:00
Joseph Myers
21708942c9
Remove pre-2.4 Linux kernel support.
2012-05-25 19:19:30 +00:00
Richard Henderson
a386f1cc22
ppc: Pass hwcap to ifuncs.
...
* sysdeps/powerpc/powerpc32/dl-irel.h (elf_ifunc_invoke): Pass
dl_hwcap to ifunc resolver.
* sysdeps/powerpc/powerpc32/dl-machine.h (elf_machine_rela): Use
elf_ifunc_invoke.
* sysdeps/powerpc/powerpc64/dl-irel.h (elf_ifunc_invoke): Pass
dl_hwcap to ifunc resolver.
* sysdeps/powerpc/powerpc64/dl-machine.h (resolve_ifunc): Likewise.
2012-05-25 10:30:42 -07:00
Joseph Myers
b65504975c
Fix acosf underflow (bug 14153).
2012-05-25 11:07:07 +00:00
Jeff Law
b0bc23a177
2012-05-24 Jeff Law <law@redhat.com>
...
* stdio-common/Makefile (tests): Add bug25.
* stdio-common/bug25.c: New test.
2012-05-24 21:30:43 -06:00
H.J. Lu
347c92e9e7
Make free chunk size a multiple of MALLOC_ALIGNMENT
2012-05-24 17:50:28 -07:00
Joseph Myers
6bcc8b3ff9
conformtest: Update expectations for stdio.h and stdlib.h.
2012-05-24 21:58:04 +00:00
Andreas Schwab
56d25bb888
Remove use of INTDEF/INTUSE in intl
2012-05-24 23:08:18 +02:00
Andreas Schwab
d18ea0c5e6
Remove use of INTDEF/INTUSE in libio
2012-05-24 23:06:20 +02:00
Roland McGrath
c14874927b
syscalls.list support for vDSO IFUNCs, use it for x32 gettimeofday and time.
2012-05-24 13:13:46 -07:00
Roland McGrath
d6c33fda03
Switch gettimeofday from INTUSE to libc_hidden_proto.
2012-05-24 13:12:34 -07:00
H.J. Lu
b5a2bbe6cc
Properly handle MALLOC_ALIGNMENT > 2 * SIZE_SZ
2012-05-24 11:57:23 -07:00
H.J. Lu
7f90742178
Add x32 abilist files
2012-05-24 11:20:12 -07:00
H.J. Lu
cfba4fda12
Add x32 C++ type data
2012-05-24 11:18:30 -07:00
Joseph Myers
7a25eb06bf
Add tests for underflow exception bugs 10846, 14036.
2012-05-24 15:36:54 +00:00
Joseph Myers
795405f956
Don't include exceptions in libm-test-ulps test names.
2012-05-24 12:14:03 +00:00
Joseph Myers
80bad0ccae
Test underflow exceptions in libm-test.inc.
2012-05-24 11:26:36 +00:00
David S. Miller
9e0e3d0b2b
Forgot changelog in previous commit.
2012-05-23 21:53:01 -07:00
H.J. Lu
ccd0a08f4e
Add sysdeps/unix/sysv/linux/x86_64/x32/configure
2012-05-23 20:03:20 -07:00
Paul Eggert
f8591f8049
mktime: avoid signed integer overflow
...
* time/mktime.c (__mktime_internal): Do not mishandle the case
where diff == INT_MIN.
2012-05-23 14:36:25 -07:00
Paul Eggert
94c7d826c9
mktime: simplify computation of average
...
* time/mktime.c (ranged_convert): Use new time_t_avg function
instead of rolling our own (probably-slower) code.
2012-05-23 14:36:25 -07:00
Paul Eggert
ce73d68397
mktime: do not assume signed right shift propagates sign bit
...
* time/mktime.c (isdst_differ): New static function.
(__mktime_internal): No need to normalize tm_isdst now.
(__mktime_internal, not_equal_tm): Use isdst_differ to compare
tm_isdst values.
2012-05-23 14:36:24 -07:00
Paul Eggert
72a22e596c
mktime: merge another wrapv change from gnulib
...
* time/mktime.c (TYPE_MAXIMUM): Rework slightly to avoid diagnostics
from some compilers.
2012-05-23 14:36:24 -07:00
Paul Eggert
6860543348
mktime: remove incorrect attempt at unusual arithmetics
...
* time/mktime.c (TYPE_ONES_COMPLEMENT, TYPE_SIGNED_MAGNITUDE): Remove.
The code didn't really work on such machines anyway.
(TYPE_MINIMUM): Assume two's complement.
(twos_complement_arithmetic): Verify that long_int and time_t
are two's complement (or unsigned, in the latter case).
2012-05-23 14:36:24 -07:00
Paul Eggert
03cf7fe31b
mktime: check signed shifts on long_int and time_t, too
...
* time/mktime.c (SHR): Check that shifts work as desired
on the types long_int and time_t too, as SHR is used on
such types.
2012-05-23 14:36:24 -07:00
Paul Eggert
f04dfbc244
mktime: do not assume 'long' is wide enough
...
* time/mktime.c (verify): Move decl up.
(long_int): New type.
(leapyear, ydhms_diff, guess_time_tm, __mktime_internal): Use it,
to remove assumption in the code that 'long' is wide enough to
store year values. This assumption is not true on x32 and on
some non-glibc platforms.
2012-05-23 14:36:24 -07:00
Paul Eggert
62bdf9a683
mktime: merge wrapv change from gnulib
...
* time/mktime.c (WRAPV): New macro.
(time_t_avg, time_t_add_ok, time_t_int_add_ok): New static functions.
(guess_time_tm, __mktime_internal): Do not assume that signed
integer overflow wraps around; modern compilers generate code
where this assumption is no longer valid.
2012-05-23 14:36:24 -07:00