- Copy stable/10@285827 to releng/10.2 in preparation for 10.2-RC1
builds.
- Update newvers.sh to reflect RC1.
- Update __FreeBSD_version to reflect 10.2.
- Update default pkg(8) configuration to use the quarterly branch.[1]

Discussed with: re, portmgr [1]
Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

MFC r271349:

Add diagnostic printfs and disable test # 4 on i386

Reviewed by: jmmv, rpaulo
Phabric: D749
PR: 191676
Sponsored by: EMC / Isilon Storage Division

MFC r271336:

Expand the tests structure in test_small(..) to workaround the
"initializer not constant" warning with gcc

Approved by: jmmv (mentor)
Phabric: D744
Sponsored by: EMC / Isilon Storage Division

Copy head (r256279) to stable/10 as part of the 10.0-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Fix some bugs in the complex trig tests so that they test both double
and float precision properly.

Add more tests for log functions. A few are commented out because the
long double versions don't pass yet. (They are rather nit-picky cases,
so there's ongoing discussion with Bruce about whether it is worth the
performance cost.)

Factor out some common code from the libm tests. This is a bit messy
because different tests have different ideas about what it means to be
"close enough" to the right answer, depending on the properties of the
function being tested. In the process, I fixed some warnings and
added a few more 'volatile' hacks, which are sufficient to make all
the tests pass at -O2 with clang.

Fix some harmless bugs in a test.

Basic tests for complex inverse trig and hyperbolic functions.

Fix cexp regression tests that have an infinite real part. The signs of the
result depend on the cosine and sine of the imaginary part.
Small values are used in the new tests such that cosine and sine are well
defined.

Reviewed by: das

Fix some regressions caused by the switch from gcc to clang. The fixes
are workarounds for various symptoms of the problem described in clang
bugs 3929, 8100, 8241, 10409, and 12958.

The regression tests did their job: they failed, someone brought it
up on the mailing lists, and then the issue got ignored for 6 months.
Oops. There may still be some regressions for functions we don't have
test coverage for yet.

Add some tests from PR 166463. Also make sure that all of the tests,
old and new, check the sign bits of both the remainder and the
quotient.

Fix a test that doesn't work on architectures where long double is no
wider than double. Thanks to Ian Lepore for catching the bug.

Add regression tests for modf{,f,l}().

Tests for complex trig and hyperbolic functions.

Tests for cancellation in fma(). Also include more tests for 128-bit
long doubles. Thanks for clusteradm (simon) for making the needed
hardware available.

Add some tests for corner cases of log() in unusual rounding modes.
I wrote these ages ago, but they've been failing until now.

Add some tests for double-rounding bugs in fma().

Add some tests for cexp() and cexpf(). (I need to clean up all of
these tests some day, but in the mean time, they're a useful sanity
check for future changes.)

Work around gcc constant folding bugs.

signbit() returns nonzero for negative arguments, but we shouldn't assume
that it always returns the same nonzero value.

Fix some warnings.

Add regression tests for logarithmic functions in the math library.

Add some basic regression tests for nearbyint().

Delete some empty mergeinfo.

Add tests for conj{,f,l}() that I wrote some time ago. These test the
versions in libm, not the gcc builtins.

Test that the result is correctly rounded when |y/x| is huge.

Add some tests for acos*(), asin*(), atan*(), and atan2*().

Add regression tests for fmin{,f,l} and fmax{,f,l}.
I wrote these to test amd64 asm functions that used
maxss, maxsd, minss, and minsd, but it turns out that
those instructions don't handle NaNs and signed zero
in the same way as fmin() and fmax() are required to,
so we're stuck with the C versions for now.

Add some tests for fma(), fmaf(), and fmal().

Test remainderl() and remquol() as well.

Add some minimal tests for csqrtl().

Don't run tests that assume <= 64-bit precision on machines with quad
precision.

Some basic regression tests for {sin,cos,tan}{,f,l}().

There are 3 tests here, not 2.

expm1(-big) generates an inexact exception but not underflow.

Add some regression tests for libm's exponential functions. These
mostly just test corner cases rather than accuracy. Some of the
tests don't pass right now if you compile libm at -O2 due to gcc
constant-folding some things that it shouldn't. I'll fix that
shortly.

Tests for lrintl() and llrintl(). I didn't add anything specially
tailored for the long double format; instead, I just modified the existing
tests to test lrintl() and llrintl() as well.

Don't try the long double tests on i386. Our reduced precision
can cause them to fail.

Remove another Alpha remnant.

Regression tests for nan{,f,l}().

Regression tests for csqrt(3).

Fixes to avoid overzealous constant folding.

gcc 4 does some overzealous constant folding, and since it doesn't
support FENV_ACCESS, that was causing this test to fail. Use a volatile
to avoid the constant folding.

Fix a few of the tests so that correct implementations actually pass them.

Remove alpha left-overs.

Add missing test-rem.t.

A few simple regression tests for remainder(), remainderf(),
remquo(), and remquof().

- Replace fe[gs]etmask() with feenableexcept(), fedisableexcept(), and
fegetexcept().
- Add additional tests for the above.
- Output a separate success message for each of the 8 components of
this set of tests.

Test cases for nextafter{,f,l} and nexttoward{,f,l}.

Regression tests for [l]lrint[f]() and [l]lround[f]().

Switch over to a different, more flexible test output protocol that's
understood by Perl's Test::Harness module and prove(1) commands.

Update README to describe the new protocol. The work's broken down into
two main sets of changes.

First, update the existing test programs (shell scripts and C programs)
to produce output in the ok/not ok format, and to, where possible, also
produce a header describing the number of tests that are expected to be
run.

Second, provide the .t files that actually run the tests. In some cases
these are copies of, or very similar too, scripts that already existed.
I've kept the old scripts around so that it's possible to verify that
behaviour under this new system (in terms of whether or not a test fails)
is identical to the behaviour under the old system.

Add a TODO file.

Add regression tests for ilogb{,f,l}().

While testing fe[gs]etround(), make sure FLT_ROUNDS reflects the
proper rounding mode as well.

Add some fenv.h regression tests.