Disk cache: Disable block file internal consistency dchecks for normal builds.

Disk cache: Disable block file internal consistency dchecks for normal builds.

BUG=169960
TEST=none
R=gavinp@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=250843

[gavinp, 2014-02-12 01:25:38]

You've been bit by a rietveld bug.
https://codereview.chromium.org/148293016/diff/1/net/disk_cache/block_files.cc
and.... "error: old chunk mismatch"

Anyway, this change LGTM.

However, it really, really freaks me out how different the two definitions of
STRESS_NOTREACHED and STRESS_DCHECK are! Shouldn't the empty form either be
empty, or at least semicolon swallowing? Can we create a bug for that, please?

[gavinp, 2014-02-12 01:37:31]

On 2014/02/12 01:25:38, gavinp wrote:
> You've been bit by a rietveld bug.
> https://codereview.chromium.org/148293016/diff/1/net/disk_cache/block_files.cc
> and.... "error: old chunk mismatch"
> 
> Anyway, this change LGTM.
> 
> However, it really, really freaks me out how different the two definitions of
> STRESS_NOTREACHED and STRESS_DCHECK are! Shouldn't the empty form either be
> empty, or at least semicolon swallowing? Can we create a bug for that, please?

One more difference: DCHECK always references its arguments.

I guess the win here is that the stress test build isn't common, so we're
unlikely to enconter that one...

[rvargas (doing something else), 2014-02-12 19:47:42]

On 2014/02/12 01:37:31, gavinp wrote:
> On 2014/02/12 01:25:38, gavinp wrote:
> > You've been bit by a rietveld bug.
> >
https://codereview.chromium.org/148293016/diff/1/net/disk_cache/block_files.cc
> > and.... "error: old chunk mismatch"
> > 
> > Anyway, this change LGTM.
> > 
> > However, it really, really freaks me out how different the two definitions
of
> > STRESS_NOTREACHED and STRESS_DCHECK are! Shouldn't the empty form either be
> > empty, or at least semicolon swallowing? Can we create a bug for that,
please?
> 
> One more difference: DCHECK always references its arguments.
> 
> I guess the win here is that the stress test build isn't common, so we're
> unlikely to enconter that one...

Thanks!.

Unfortunately, I don't see what is bothering you. Feel free to expand your
concerns on a bug.

[rvargas (doing something else), 2014-02-12 19:48:17]

The CQ bit was checked by rvargas@chromium.org

[commit-bot: I haz the power, 2014-02-12 19:52:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rvargas@chromium.org/148293016/1

[gavinp, 2014-02-12 21:18:22]

On 2014/02/12 19:47:42, rvargas wrote:
> On 2014/02/12 01:37:31, gavinp wrote:
> > On 2014/02/12 01:25:38, gavinp wrote:
> > > You've been bit by a rietveld bug.
> > >
> https://codereview.chromium.org/148293016/diff/1/net/disk_cache/block_files.cc
> > > and.... "error: old chunk mismatch"
> > > 
> > > Anyway, this change LGTM.
> > > 
> > > However, it really, really freaks me out how different the two definitions
> of
> > > STRESS_NOTREACHED and STRESS_DCHECK are! Shouldn't the empty form either
be
> > > empty, or at least semicolon swallowing? Can we create a bug for that,
> please?
> > 
> > One more difference: DCHECK always references its arguments.
> > 
> > I guess the win here is that the stress test build isn't common, so we're
> > unlikely to enconter that one...
> 
> Thanks!.
> 
> Unfortunately, I don't see what is bothering you. Feel free to expand your
> concerns on a bug.

Consider this code:

int foo = bar() + quux();
if (frob())
  STRESS_DCHECK(foo);
else
  nitz();

That doesn't compile for two reasons outside of stress cache mode; because "foo"
is unreferenced, and because the else statement is invalid. If we had used a
regular DCHECK here, it would have compiled.

[commit-bot: I haz the power, 2014-02-12 23:38:08]

Change committed as 250843