Fix -Wsign-compare bugs with GCC 4.9.2

Fix -Wsign-compare bugs with GCC 4.9.2

R=svenpanne@chromium.org
LOG=N
BUG=

Committed: https://crrev.com/09ae1c3ddc31b4eeea70937ab21b98d4eddf1eb8
Cr-Commit-Position: refs/heads/master@{#28039}

[wingo, 2015-04-23 14:11:52]

No idea why these checks are causing warnings and the above "if (value==0) {}"
check isn't, but so it is.

In file included from .././src/checks.h:9:0,
                 from .././src/v8.h:31,
                 from ../src/deoptimizer.cc:5:
.././src/base/logging.h: In instantiation of ‘std::string*
v8::base::CheckEQImpl(const Lhs&, const Rhs&, const char*) [with Lhs = int; Rhs
= long unsigned int; std::string = std::basic_string<char>]’:
../src/deoptimizer.cc:2370:101:   required from here
.././src/base/logging.h:123:154: error: comparison between signed and unsigned
integer expressions [-Werror=sign-compare]
 DEFINE_CHECK_OP_IMPL(EQ, ==)
                                                                                
                                                                         ^
cc1plus: error: unrecognized command line option "-Wno-format-pedantic"
[-Werror]
cc1plus: all warnings being treated as errors

[Sven Panne, 2015-04-24 06:01:51]

On 2015/04/23 14:11:52, wingo wrote:
> No idea why these checks are causing warnings and the above "if (value==0) {}"
> check isn't, but so it is. [...]

The DCHECK_FOO macros are totally fragile because they highly depend on the
complex interaction between integral promotion, candidate selection during
overload resolution  and normal functions "winning" over templates (even when
they are fully specialized), see e.g.
http://en.cppreference.com/w/cpp/language/overload_resolution. We regularly have
trouble with them.

My personal opinion here is: Nuke this complete nonsense, it just doesn't work
reliably. Simple DCHECK (without any _EQ, _NE etc. suffix) is all you need and
all that is reliable. Sadly enough, our presubmit.py thinks differently... :-P

[Sven Panne, 2015-04-24 06:02:00]

The CQ bit was checked by svenpanne@chromium.org

[Sven Panne, 2015-04-24 06:02:00]

lgtm

[commit-bot: I haz the power, 2015-04-24 06:02:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1105653002/1

[commit-bot: I haz the power, 2015-04-24 06:57:02]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-04-24 06:57:17]

Patchset 1 (id:??) landed as
https://crrev.com/09ae1c3ddc31b4eeea70937ab21b98d4eddf1eb8
Cr-Commit-Position: refs/heads/master@{#28039}