Allan McRae [Sat, 27 Apr 2013 05:13:12 +0000 (15:13 +1000)]
Update i386 libm-test ULPs
Joseph Myers [Fri, 26 Apr 2013 19:26:22 +0000 (19:26 +0000)]
Fix catan, catanh missing underflows (bug 15406).
Joseph Myers [Fri, 26 Apr 2013 19:25:19 +0000 (19:25 +0000)]
Fix csin, csinh, ccos, ccosh missing underflows (bug 15405).
Adhemerval Zanella [Fri, 26 Apr 2013 18:00:56 +0000 (13:00 -0500)]
PowerPC: modf optimization fix
This patch fix the
3c0265394d9ffedff2b0de508602dc52e077ce5c commits
by correctly setting minimum architecture for modf PPC optimization
to power5+ instead of power5 (since only on power5+ round/ceil will
be inline to inline assembly).
Markus Trippelsdorf [Fri, 26 Apr 2013 07:30:46 +0000 (09:30 +0200)]
Update x86_64 ULPs
2013-04-26 Markus Trippelsdorf <markus@trippelsdorf.de>
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
Joseph Myers [Thu, 25 Apr 2013 19:34:30 +0000 (19:34 +0000)]
Add missing bug numbers to NEWS.
Joseph Myers [Thu, 25 Apr 2013 19:23:11 +0000 (19:23 +0000)]
Move x86_64-specific audit tests to sysdeps/x86_64/.
Paul Pluzhnikov [Thu, 25 Apr 2013 18:08:31 +0000 (11:08 -0700)]
Get rid of __STDC_FORMAT_MACROS, __STDC_LIMIT_MACROS and __STDC_CONSTANT_MACROS
Maciej W. Rozycki [Thu, 25 Apr 2013 15:09:19 +0000 (16:09 +0100)]
Declare __ehdr_start with hidden visibility.
This avoids a linker bug triggering for MIPS SVR4 binaries:
http://sourceware.org/bugzilla/show_bug.cgi?id=15365
and regardless serves as a documentation of intent.
Maciej W. Rozycki [Thu, 25 Apr 2013 15:01:21 +0000 (16:01 +0100)]
soft-fp: s/sNAN/NAN/ -- no sNaNs are ever produced.
Carlos O'Donell [Wed, 24 Apr 2013 20:29:33 +0000 (16:29 -0400)]
math: Use accurate answers for cos and sincos.
Use the most accurate hex literals possible for the answers to the
cos and sincos tests that vary according to the error in the rounding
of PI/2.
---
2013-04-24 Carlos O'Donell <carlos@redhat.com>
* math/libm-test.inc (cos_test): Use accurate hex constants.
(sincost_test): Likewise.
Joseph Myers [Wed, 24 Apr 2013 20:14:15 +0000 (20:14 +0000)]
Add catan, catanh tests at +/- 1 and +/- i.
Joseph Myers [Wed, 24 Apr 2013 18:59:37 +0000 (18:59 +0000)]
Use suffixed floating-point constants in float and long double catan/catanh.
Joseph Myers [Wed, 24 Apr 2013 18:49:13 +0000 (18:49 +0000)]
Fix catan, catanh inaccuracy through use of log (bug 15394).
Siddhesh Poyarekar [Wed, 24 Apr 2013 08:36:02 +0000 (14:06 +0530)]
Mention files in which fast/slow paths of math functions are implemented
Roland McGrath [Tue, 23 Apr 2013 22:57:25 +0000 (15:57 -0700)]
Fix name space use in last commit.
Roland McGrath [Tue, 23 Apr 2013 22:28:59 +0000 (15:28 -0700)]
Add generic POSIX implementation of C11 timespec_get.
Adhemerval Zanella [Mon, 25 Mar 2013 21:10:06 +0000 (16:10 -0500)]
PowerPC: modf optimization
This patch implements modf/modff optimization for POWER by focus
on FP operations instead of relying in integer ones.
Siddhesh Poyarekar [Tue, 23 Apr 2013 10:02:42 +0000 (15:32 +0530)]
Consistently use ISSPACE to check for whitespace
Resolves #14888.
This only really manifests itself when there are no spaces between
format specifiers, which is not allowed by POSIX, but is allowed by
the glibc implementation.
Andreas Schwab [Sat, 20 Apr 2013 10:12:35 +0000 (12:12 +0200)]
Remove non-standard initialisation of flexible array member
This avoids GCC bug 28865.
Heiko Carstens [Tue, 23 Apr 2013 06:53:44 +0000 (08:53 +0200)]
S/390: Change struct statfs[64] member types to unsigned values
Kay Sievers reported that coreutils' stat tool has a problem with
s390's statfs[64] definition:
> The definition of struct statfs::f_type needs a fix. s390 is the only
> architecture in the kernel that uses an int and expects magic
> constants lager than INT_MAX to fit into.
>
> A fix is needed to make Fedora boot on s390, it currently fails to do
> so. Userspace does not want to add code to paper-over this issue.
[...]
> Even coreutils cannot handle it:
> #define RAMFS_MAGIC 0x858458f6
> # stat -f -c%t /
>
ffffffff858458f6
>
> #define BTRFS_SUPER_MAGIC 0x9123683E
> # stat -f -c%t /mnt
>
ffffffff9123683e
The bug is caused by an implicit sign extension within the stat tool:
out_uint_x (pformat, prefix_len, statfsbuf->f_type);
where the format finally will be "%lx".
A similar problem can be found in the 'tail' tool.
s390 is the only architecture which has an int type f_type member in
struct statfs[64]. Other architectures have either unsigned ints or
long values, so that the problem doesn't occur there.
Therefore change the type of the f_type member to unsigned int, so
that we get zero extension instead sign extension when assignment to
a long value happens.
Reported-by: Kay Sievers <kay@vrfy.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Jan-Benedict Glaw [Mon, 22 Apr 2013 17:44:31 +0000 (10:44 -0700)]
Fix getent to call endspent rather than endpwent for shadow database.
Siddhesh Poyarekar [Mon, 22 Apr 2013 04:58:31 +0000 (10:28 +0530)]
Consolidate pthread_attr value validation
Define inline functions that wrap around validation for each of the
pthread attributes to reduce duplication in code.
Siddhesh Poyarekar [Mon, 22 Apr 2013 04:54:00 +0000 (10:24 +0530)]
Minor cleanup in getaddrinfo
Replace repeated computations of alloca size with a local variable
that stores the computed value.
David S. Miller [Sun, 21 Apr 2013 19:36:14 +0000 (15:36 -0400)]
Russian translations update.
* po/ru.po: Update Russion translation from translation project.
Roland McGrath [Fri, 19 Apr 2013 18:33:01 +0000 (11:33 -0700)]
ARM: Macroize assembly use of EABI unwind directives.
Adam Conrad [Fri, 19 Apr 2013 05:58:16 +0000 (23:58 -0600)]
Remove __wur from setfsuid and setfsgid.
David Holsgrove [Tue, 10 Jan 2012 05:38:42 +0000 (15:38 +1000)]
MicroBlaze Port
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Carlos O'Donell [Thu, 18 Apr 2013 00:17:59 +0000 (17:17 -0700)]
Update ChangeLog.
Update ChangeLog for
0f122b8d12875181a4fba3449e3549d34d3dc398.
Carlos O'Donell [Thu, 18 Apr 2013 00:13:20 +0000 (17:13 -0700)]
Configuring for i386 is no longer supported.
We no longer support configuring for i386, nor do we
elide such a configuration to i686. Configuring with
i386-* is a failure, and we provide an example of
how to fix that.
---
2013-04-17 Carlos O'Donell <carlos@redhat.com>
* configure.in: Remove i386 configure warning. Remove i386 case.
* configure: Regenerate.
* sysdeps/i386/configure.in: Raise error if config_machine is i386.
Add example to error message.
* sysdeps/i386/configure: Regenerate.
Siddhesh Poyarekar [Wed, 17 Apr 2013 12:15:55 +0000 (17:45 +0530)]
Add benchmark inputs for cos and tan
Roland McGrath [Tue, 16 Apr 2013 19:11:26 +0000 (12:11 -0700)]
Don't test O_RDONLY case in tst-aio7.
Siddhesh Poyarekar [Tue, 16 Apr 2013 13:04:03 +0000 (18:34 +0530)]
Define NOT_IN_libc when compiling benchmark programs
Siddhesh Poyarekar [Tue, 16 Apr 2013 08:37:21 +0000 (14:07 +0530)]
Add target bench-clean
David Holsgrove [Wed, 4 Jan 2012 03:56:48 +0000 (13:56 +1000)]
Adding MicroBlaze support to elf/elf.h
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
Thomas Schwinge [Mon, 15 Apr 2013 20:11:30 +0000 (22:11 +0200)]
Fix coding-style violation.
Introduced in commit
ee091edf1a16312fff0cc3c6ae0116aded7f1dbf.
Andreas Schwab [Tue, 4 Dec 2012 15:14:13 +0000 (16:14 +0100)]
Properly check for short writes when sending the response in nscd
Siddhesh Poyarekar [Mon, 15 Apr 2013 08:22:26 +0000 (13:52 +0530)]
Write to bench.out-tmp only once
Appending benchmark program output on every run could result in a case
where the benchmark run was cancelled, resulting in a partially
written file. This file gets used again on the next run, resulting in
results being appended to old results.
It could have been possible to remove the file before every benchmark
run, but it is easier to just write the output to bench.out-tmp only
once.
Andreas Schwab [Sun, 23 Dec 2012 08:45:07 +0000 (09:45 +0100)]
nscd: don't fork twice
Siddhesh Poyarekar [Mon, 15 Apr 2013 06:05:43 +0000 (11:35 +0530)]
Fix off-by-one bug in tst-fwrite-error
tst-fwrite-error did not allocate enough space for the string 'world'
and its NULL terminator. Fixed.
Siddhesh Poyarekar [Mon, 15 Apr 2013 05:47:01 +0000 (11:17 +0530)]
Rebuild benchmark sources when Makefile is updated
Benchmark programs are generated using parameters from the Makefile,
so it is necessary to rebuild them whenever the parameters in the
Makefile are updated. Hence, added a dependency for the generated C
source on the Makefile so that it gets regenerated when the Makefile
is updated.
Roland McGrath [Fri, 12 Apr 2013 20:11:20 +0000 (13:11 -0700)]
BZ#15361: Make aio_fsync not check open modes.
Siddhesh Poyarekar [Fri, 12 Apr 2013 09:31:44 +0000 (15:01 +0530)]
Move bench target to benchtests
The bench target will only be used within the benchtests directory.
Andreas Schwab [Thu, 11 Apr 2013 14:32:31 +0000 (16:32 +0200)]
m68k: update libm test ULPs
Carlos O'Donell [Thu, 11 Apr 2013 12:52:18 +0000 (08:52 -0400)]
libm-test.inc: Fix tests where cos(PI/2) != 0.
The value of PI is never exactly PI in any floating point representation,
and the value of PI/2 is never PI/2. It is wrong to expect cos(M_PI_2l)
to return 0, instead it will return an answer that is non-zero because
M_PI_2l doesn't round to exactly PI/2 in the type used.
That is to say that the correct answer is to do the following:
* Take PI or PI/2.
* Round to the floating point representation.
* Take the rounded value and compute an infinite precision cos or sin.
* Use the rounded result of the infinite precision cos or sin as the
answer to the test.
I used printf to do the type rounding, and Wolfram's Alpha to do the
infinite precision cos calculations.
The following changes bring x86-64 and x86 to 1/2 ulp for two tests.
It shows that the x86 cos implementation is quite good, and that
our test are flawed.
Unfortunately given that the rounding errors are type dependent we
need to fix this for each type. No regressions on x86-64 or x86.
---
2013-04-11 Carlos O'Donell <carlos@redhat.com>
* math/libm-test.inc (cos_test): Fix PI/2 test.
(sincos_test): Likewise.
* sysdeps/x86_64/fpu/libm-test-ulps: Regenerate.
* sysdeps/i386/fpu/libm-test-ulps: Regenerate.
Andreas Schwab [Thu, 22 Nov 2012 11:57:37 +0000 (12:57 +0100)]
Extend i486 pthread_cond_timedwait to use futex syscall with absolute timeout
Andreas Schwab [Mon, 14 Jan 2013 16:32:20 +0000 (17:32 +0100)]
Don't accept exp char without preceding digits in scanf float parsing
Andreas Schwab [Tue, 15 Jan 2013 15:39:07 +0000 (16:39 +0100)]
Fix invalid free of memory allocated during rtld init
Siddhesh Poyarekar [Thu, 11 Apr 2013 04:07:50 +0000 (09:37 +0530)]
Don't use run-via-rtld-prefix for anything other than tests
run-via-rtld-prefix checks whether the program to be run is a static
test and skips if it is. This is fine, except that it assumes that
the program to be run is the second $^, which is true only for tests.
This change creates an rtld-prefix, which is simply the dynamic linker
prefix with the necessary arguments and uses that in the non-test
targets.
Siddhesh Poyarekar [Thu, 11 Apr 2013 03:45:52 +0000 (09:15 +0530)]
Remove eval
It's not needed.
Siddhesh Poyarekar [Wed, 10 Apr 2013 06:01:46 +0000 (11:31 +0530)]
Accept leading and trailing spaces in getdate input string
Fixes #15346.
The POSIX description of getdate allows for extra spaces in the
getdate input string. __getdate_r uses strptime internally, which
works fine with extra spaces between format strings (and hence within
an input string) but not with leading and trailing spaces. So we trim
off the leading and trailing spaces before we pass it on to strptime.
Roland McGrath [Mon, 8 Apr 2013 21:31:38 +0000 (14:31 -0700)]
BZ#14280: Fix Hurd ioctl macro to avoid warning.
Carlos O'Donell [Sun, 7 Apr 2013 20:13:02 +0000 (16:13 -0400)]
sem_post.c: Include atomic.h.
The sem_post.c file uses atomic functions without including
atomic.h. Add `#include <atomic.h>' to the file to prevent
any compile time warnings when other headers change and
atomic.h isn't implicitly included.
---
nptl/
2013-04-07 Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/sem_post.c: Include atomic.h.
Carlos O'Donell [Sat, 6 Apr 2013 21:13:19 +0000 (17:13 -0400)]
README.libm-test: Use testrun.sh to run libm test.
Document the use of the convenience testrun.sh script for
running the libm test.
---
2013-04-06 Carlos O'Donell <carlos@redhat.com>
* math/README.libm-test (How can I generate "libm-test-ulps"?):
Use testrun.sh to run libm tests.
Carlos O'Donell [Sat, 6 Apr 2013 21:00:02 +0000 (17:00 -0400)]
dl_open_worker: Memset all of seen array.
The seen array was doubled in size recently, but the memset to clear
the array was not adjusted. We adjust the memset to always be correct
regardless of the size of seen.
---
2013-04-06 Carlos O'Donell <carlos@redhat.com>
[BZ #15309]
* elf/dl-open.c (dl_open_worker): memset all of seen array.
Marko Myllynen [Sat, 6 Apr 2013 20:33:26 +0000 (16:33 -0400)]
fi_FI: Define yesstr, nostr
Define yesstr/nostr in fi_FI (as "Kyllä" and "Ei").
Fixes part of BZ#15264.
---
2013-04-06 Marko Myllynen <myllynen@redhat.com>
[BZ #15264]
* locales/fi_FI (LC_MESSAGES): Define yesstr and nostr.
Carlos O'Donell [Sat, 6 Apr 2013 20:22:47 +0000 (16:22 -0400)]
New Makefile target `regen-ulps'.
The wiki "Regeneration" page has this to say about update ULPs.
"The libm-test-ulps files are semiautomatically updated. To
update an ulps baseline, run each of the failing tests (test-float,
test-double, etc.) with -u; this will generate a file called ULPs;
concatenate each of those files with the existing libm-test-ulps
file, after removing any entries for particularly huge numbers of
ulps that you do not want to mark as expected. Then run
gen-libm-test.pl -n -u FILE where FILE is the concatenated file
produced in the previous step. This generates a file called
NewUlps which is the new sorted version of libm-test-ulps."
The same information is listed in math/README.libm-test, and is a
lot of manual work that you often want to run over-and-over again
while working on a particular test.
The `regen-ulps' convenience target does this automatically for
developers.
We strictly assume the source tree is readonly and add a
new --output-dir option to libm-test.inc to allow for writing
out ULPs to $(objpfx).
When run the new target does the following:
* Starts with the baseline ULPs file.
* Runs each of the libm math tests with -u.
* Adds new changes seen with -u to the baseline.
* Sorts and prepares the test output with gen-libm-test.pl.
* Leaves math/NewUlps in your build tree to copy to your source
tree, cleanup, and checkin.
The math test documentation in math/README.libm-test is updated
document the new Makefile target.
---
2013-04-06 Carlos O'Donell <carlos@redhat.com>
* Makefile.in (regen-ulps): New target.
* math/Makefile [ifneq (no,$(PERL)]: Declare regen-ulps with .PHONY.
[ifneq (no,$(PERL)] (run-regen-ulps): New variable.
[ifneq (no,$(PERL)] (regen-ulps): New target.
[ifeq (no,$(PERL)] (regen-ulps): New target.
* math/libm-test.inc (ulps_file_name): Define.
(output_dir): New variable.
(options): Add "output-dir" option.
(parse_opt): Handle 'o' case.
(main): If output_dir is non-NULL use it as a prefix
otherwise use "".
* math/README.libm-test: Update `How can I generate "libm-test-ulps"?'
Carlos O'Donell [Sat, 6 Apr 2013 16:00:35 +0000 (12:00 -0400)]
i386: Fail at configure time for i386 builds.
This change does two things:
* Treats a target i386-* as if it were i686.
* Fails configure if the user is generating code
for i386.
We no longer support i386 code-generation because the i386
lacks the atomic operations we need in glibc.
You can still configure for i386-*, but you get i686 code.
You can't build with --march=i386, --mtune=i386 or a compiler
that defaults to i386 code-generation.
I've added two i386 entries in the master todo list to discuss
merging and renaming:
http://sourceware.org/glibc/wiki/Development_Todo/Master#i386
The failure modes are fail-safe here. You compile for i386,
get i686, and try to run on i386 and it fails. The configure
log has a warning saying we elided to i686. There is no situation
that I can see where we run into any serious problems.
The patch makes the current state better in that we get less
confused users and we build successfully in more default
configurations.
The next enhancement would be to add --march=i?86
as suggested in #c20 of BZ#10062 for any i?86-* builds, which
would solve the problem of a 32-bit compiler that defaults to
i386 code-gen and glibc configured for i686-* target. Which
previously failed at build time, and now will fail at configure
time (requires adding --march=i686).
Updated NEWS with BZ #10060 and #10062.
No regressions.
---
2013-04-06 Carlos O'Donell <carlos@redhat.com>
[BZ #10060, #10062]
* aclocal.m4 (LIBC_COMPILER_BUILTIN_INLINED): New macro.
* sysdeps/i386/configure.in: Use LIBC_COMPILER_BUILTIN_INLINED and
fail configure if __sync_val_compare_and_swap is not inlined.
* sysdeps/i386/configure: Regenerate.
* configure.in: Build for i686 when configured for i386.
* configure: Regenerate.
* README: Remove i386 reference.
Carlos O'Donell [Sat, 6 Apr 2013 15:41:44 +0000 (11:41 -0400)]
Update ChangeLog.
Update ChangeLog for commit:
24116fcdbe55d58aa35fc77444ae71a9dec9fff6
Carlos O'Donell [Sat, 6 Apr 2013 14:53:54 +0000 (10:53 -0400)]
s390/s390x: Undef PSEUDO before redef.
The s390 and s390x sysdep.h files include the more generic sysdep.h.
The more generic sysdep.h defines PSEUDO. This causes an annoying
CPP warning saying the PSEUDO was redefined. This patch removes the
warning by undefining PSEUDO before the redefinition. This is in line
with what all the other machines do.
---
2013-04-06 Carlos O'Donell <carlos@redhat.com>
* sysdeps/s390/s390-32/sysdep.h: Undefine PSEUDO before redefinition.
* sysdeps/s390/s390-64/sysdep.h: Likewise.
Thomas Schwinge [Fri, 5 Apr 2013 18:28:47 +0000 (20:28 +0200)]
More tests for external charater sequences representing NaNs.
Thomas Schwinge [Fri, 5 Apr 2013 13:58:27 +0000 (15:58 +0200)]
Properly collect individual tests' results.
Fixup for commit
003c9895a89e71767ad64bafac1ca99622be2eb7.
Thomas Schwinge [Thu, 4 Apr 2013 15:35:12 +0000 (17:35 +0200)]
[BZ #14686, #15336] Fix standard compliance. Don't use hard-coded qNaN values.
Thomas Schwinge [Wed, 3 Apr 2013 20:00:21 +0000 (22:00 +0200)]
[BZ #15335, #15342] Fix standard compliance. Don't use hard-coded qNaN values.
Thomas Schwinge [Wed, 3 Apr 2013 19:10:02 +0000 (21:10 +0200)]
Remove unreachable code.
The case of y == 0 is handled at the beginning of the function.
Thomas Schwinge [Wed, 3 Apr 2013 20:17:47 +0000 (22:17 +0200)]
Remove unused hard-coded qNaN definition.
Siddhesh Poyarekar [Thu, 4 Apr 2013 14:17:41 +0000 (19:47 +0530)]
Add BZ #15337 to NEWS
Siddhesh Poyarekar [Thu, 4 Apr 2013 14:13:56 +0000 (19:43 +0530)]
Fix static build when configured with --disable-hidden-plt
Fixes BZ #15337.
Static builds fail with the following warning:
/home/tools/glibc/glibc/nptl/../nptl/sysdeps/unix/sysv/linux/x86_64/cancellation.S:80:
undefined reference to `__GI___pthread_unwind'
When the source is configured with --disable-hidden-plt. This is
because the preprocessor conditional in cancellation.S only checks if
the build is for SHARED, whereas hidden_def is defined appropriately
only for a SHARED build that will have symbol versioning *and* hidden
defs are enabled. The last case is false here.
Joseph Myers [Wed, 3 Apr 2013 16:51:46 +0000 (16:51 +0000)]
Fix missing underflow from cexp (bug 14478).
Andreas Schwab [Wed, 3 Apr 2013 15:53:57 +0000 (17:53 +0200)]
Update NEWS
Andreas Schwab [Thu, 21 Mar 2013 14:50:27 +0000 (15:50 +0100)]
Fix stack overflow in getaddrinfo with many results
Thomas Schwinge [Wed, 3 Apr 2013 12:13:44 +0000 (14:13 +0200)]
Refer to two GCC PRs.
Siddhesh Poyarekar [Wed, 3 Apr 2013 08:50:56 +0000 (14:20 +0530)]
Update bench.out and bench.out.old only upon completion
Write output from the currently running benchmark into a temporary
file and move files around only once the current run is complete.
That way we don't lose data from the last two runs due to an
incomplete run.
Siddhesh Poyarekar [Wed, 3 Apr 2013 08:02:37 +0000 (13:32 +0530)]
Echo benchmark that is currently running
This will be useful when the benchmark has many functions.
Siddhesh Poyarekar [Wed, 3 Apr 2013 08:08:39 +0000 (13:38 +0530)]
Add benchmark inputs for atan
Add separate inputs for slow and fast paths of atan
Siddhesh Poyarekar [Wed, 3 Apr 2013 05:44:39 +0000 (11:14 +0530)]
Remove TWO
Minor cleanup to remove the macro TWO and use the value directly
instead.
Siddhesh Poyarekar [Wed, 3 Apr 2013 05:26:45 +0000 (10:56 +0530)]
Preserve errno across _PC_CHOWN_RESTRICTED call on XFS
Fix BZ #15305.
On kernel versions earlier than 2.6.29, the Linux kernel exported a
sysctl called restrict_chown for xfs, which could be used to allow
chown to users other than the owner. 2.6.29 removed this support,
causing the open_not_cancel_2 to fail and thus modify errno. The fix
is to save and restore errno so that the caller sees it as unmodified.
Additionally, since the code to check the sysctl is not useful on
newer kernels, we add an ifdef so that in future the code block gets
rmeoved completely.
Joseph Myers [Tue, 2 Apr 2013 22:54:00 +0000 (22:54 +0000)]
Fix cacosh inaccuracy and spurious exceptions (bug 15327).
Siddhesh Poyarekar [Tue, 2 Apr 2013 12:23:09 +0000 (17:53 +0530)]
Use mantissa_t in mpexp
Siddhesh Poyarekar [Mon, 1 Apr 2013 09:15:54 +0000 (14:45 +0530)]
Add benchmark inputs for sin
Siddhesh Poyarekar [Tue, 2 Apr 2013 05:06:01 +0000 (10:36 +0530)]
Add benchmark tests for slowpow and slowexp
Separate benchmarks for the fast and slow implementations of pow and
exp since measuring both together doesn't make sense. Adjust the
iterations for pow and exp accordingly so that they run long enough
for the measurements to be meaningful.
Adhemerval Zanella [Tue, 2 Apr 2013 12:08:09 +0000 (07:08 -0500)]
Thomas Schwinge [Tue, 2 Apr 2013 11:51:02 +0000 (13:51 +0200)]
New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
It is based on draft TS 18661 and currently enabled as a GNU extension.
Adhemerval Zanella [Fri, 29 Mar 2013 23:15:28 +0000 (18:15 -0500)]
PowerPC: remove branch prediction from rint implementation
The branch prediction hints is actually hurts performance in this case.
The assembly implementation make two assumptions: 1. 'fabs (x) < 2^52'
is unlikely and 2. 'x > 0.0' is unlike (if 1. is true). Since it a
general floating point function, expected input is not bounded and then
it is better to let the hardware handle the branches.
David S. Miller [Sat, 30 Mar 2013 20:58:52 +0000 (16:58 -0400)]
Update German translations.
* po/de.po: Update from translation team.
Joseph Myers [Sat, 30 Mar 2013 13:31:53 +0000 (13:31 +0000)]
Fix casinh inaccuracy for imaginary part < 1.0, real part small (bug 10357).
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:23:05 +0000 (16:53 +0530)]
Remove usage of ONE macro
I missed this instance, which caused a build failure in powerpc.
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:10:36 +0000 (16:40 +0530)]
Remove ONE and MONE
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:08:27 +0000 (16:38 +0530)]
Format s_tan.c
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:07:26 +0000 (16:37 +0530)]
Remove ZERO and MZERO macros
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:04:28 +0000 (16:34 +0530)]
Format s_atan.c
Siddhesh Poyarekar [Fri, 29 Mar 2013 11:01:52 +0000 (16:31 +0530)]
Format e_log.c
Roland McGrath [Thu, 28 Mar 2013 23:52:57 +0000 (16:52 -0700)]
Avoid unconditional __call_tls_dtors calls in static linking.
Roland McGrath [Thu, 28 Mar 2013 23:15:48 +0000 (16:15 -0700)]
Use __ehdr_start, if available, as fallback for AT_PHDR.
Roland McGrath [Thu, 28 Mar 2013 22:39:32 +0000 (15:39 -0700)]
Make _dl_phdr pointer to const.
Roland McGrath [Thu, 28 Mar 2013 22:33:57 +0000 (15:33 -0700)]
Consolidate declarations of _dl_phdr, _dl_phnum.
Joseph Myers [Thu, 28 Mar 2013 21:43:56 +0000 (21:43 +0000)]
Fix typo in bug number in NEWS.
Roland McGrath [Thu, 28 Mar 2013 17:38:37 +0000 (10:38 -0700)]
Never call syslog in __libc_message.
Alan Modra [Thu, 28 Mar 2013 17:16:28 +0000 (12:16 -0500)]
PowerPC: .eh_frame info in crt1.o isn't useful and triggers gold bug 14675.
The .eh_frame info in crt1.o isn't useful and this patch prevents it from
being generated on PowerPC. It triggers the following gold bug:
http://sourceware.org/bugzilla/show_bug.cgi?id=14675
Adhemerval Zanella [Thu, 28 Mar 2013 12:52:48 +0000 (09:52 -0300)]
Fix e_logl (128ibm) spurious underflow