[patch] tree-cfg.c: Speed up cleanup_tree_cfg().

[Giovanni Bajo]

Andrew MacLeod wrote:

> So gcc_assert() isn't a direct replacement for ENABLE_CHECKING? I was
> under the impression that the contents of the gcc_assert went away when
> you disabled checking...

No, they are orthogonal. On trunk, we have everything enabled. On release
branches, we disable checking but asserts are still enabled. The idea is
that you disable strong verifications (like verify_* on trees), but you
still have some quick consistency checking which avoids some wrong code
generation.

>   gcc_assert (cond);
> becomes
>   0 && cond;
> or some such thing.

This happens only if not ENABLE_ASSERT_CHECKING.

Giovanni Bajo