MIPS/ARM: Fix benchmarks failures on simulator.

MIPS/ARM: Fix benchmarks failures on simulator.

The default x87 FPU precision is usually 80-bit extended precision,
which can lead to precision failures on several tests.
This patch fixes the precision failures on simulator by
setting the x87 FPU precision to 64-bit double precision.

TEST=benchmarks/sunspider/3d-cube, benchmarks/sunspider/3d-raytrace,
     benchmarks/sunspider/access-nbody, benchmarks/sunspider/math-partial-sums

BUG=

[palfia, 2013-09-09 13:41:53]

PTAL.

[Paul Lind, 2013-09-09 15:23:46]

LTGM, but we need v8 team to review the ARM fix as well. PTAL.

[Yang, 2013-09-10 16:47:34]

On 2013/09/09 15:23:46, Paul Lind wrote:
> LTGM, but we need v8 team to review the ARM fix as well. PTAL.

I don't see results being verified in either of the benchmarks. Why is this
necessary?

[Paul Lind, 2013-09-10 17:33:14]

On 2013/09/10 16:47:34, Yang wrote:
> I don't see results being verified in either of the benchmarks. Why is this
> necessary?

Some SunSpider benchmarks do internal data validation, for instance lines
171-173 of access-nbody.js.

If I run the test like this on Ubuntu 12.04 LTS:

$ make arm.release -j8
$ tools/run-tests.py -j8 --arch=arm --mode=release --timeout=240 --time
--no-presubmit --report benchmarks/sunspider

I get 3 errors, the first one is this:

=== benchmarks/sunspider/access-nbody ===                                
/home/paul/v8/gv8/test/benchmarks/sunspider/access-nbody.js:173: ERROR: bad
result: expected -1.3524862408537381 but got -1.352486240853738
    throw "ERROR: bad result: expected " + expected + " but got " + ret;

The problem is that the x86 defaults to 80-bit precision for results that stay
inside the FP unit. So we do not get the answer expected if all operations had
64-bit precision as they do on real ARM or MIPS cores.

Note that this does NOT happen on Mac running 10.8.4, so apparently the FPU
defaults are different on Linux & OS-X. We could put this change behind a
platform check, I suppose.

[Paul Lind, 2013-09-10 17:38:11]

On 2013/09/10 17:33:14, Paul Lind wrote:
> We could put this change behind a platform check, I suppose.

Oops, we will HAVE to do this, as the fix does not build on Mac, it is Linux
specific. Sorry for not checking this earlier.

Ákos, let's work this out, and upload a new patchset before we do further
reviews.