Convert a bunch of BPF_TEST tests to use BPF_TEST_C

Convert a bunch of BPF_TEST tests to use BPF_TEST_C

This CL converts all BPF_TEST tests that didn't make use of BPF_AUX to
instead use BPF_TEST_C and a proper SandboxBPFPolicy instead of a
SyscallEvaluator.  Simultaneously, it removes a lot of unnecessary
code that was only necessary to support these tests that don't use
BPF_AUX.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=272014

[mdempsky, 2014-05-16 23:39:30]

Apologies in advance for this boring cleanup CL. :)  I tried to structure things
to minimize indentation churn, so hopefully it's not too bad.

(And again, builds on the previous two CLs I mailed out.)

[jln (very slow on Chromium), 2014-05-19 21:22:20]

lgtm

In this very localized unittest, I think it would be silly to add the
DISALLOW_COPY_AND_ASSIGN() clutter.
With C++03, doing a copy of such a policy would be UB, but since C++11 wouldn't
let us compile it, and since we always have a clang-building bot, I think this
is fine.

I wonder if we should make it an official Chrome policy to accept this.

https://chromiumcodereview.appspot.com/289193003/diff/40001/sandbox/linux/sec...
File sandbox/linux/seccomp-bpf/sandbox_bpf_unittest.cc (right):

https://chromiumcodereview.appspot.com/289193003/diff/40001/sandbox/linux/sec...
sandbox/linux/seccomp-bpf/sandbox_bpf_unittest.cc:137: }
It would be fine in C++11 to punt on DISALLOW_COPY_AND_ASSIGN() even though
strictly speaking the style guide requires it.
In C++03, the implicitly declared copy constructor has undefined behavior if the
base class has a private copy constructor, so it's a little more problematic.
Since we always build with Clang, our bots should catch this anyways.

I'm not sure what to do here. Maybe a question for chromium-dev?

[mdempsky, 2014-05-19 21:47:15]

On 2014/05/19 21:22:20, jln wrote:
> I'm not sure what to do here. Maybe a question for chromium-dev?

As discussed, this seems to be safe even in C++03, but I'll check with
chromium-dev to see what the general rule here is.

[mdempsky, 2014-05-21 04:24:42]

The CQ bit was checked by mdempsky@chromium.org

[commit-bot: I haz the power, 2014-05-21 04:26:20]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mdempsky@chromium.org/289193003/80001

[commit-bot: I haz the power, 2014-05-21 23:29:51]

Change committed as 272014