Revert of Generic version of caching cc switches to avoid searching of switches...

Revert of https://codereview.chromium.org/69123002/
Reason for revert: This CL caused a regression in sizes bots, see https://chromeperf.appspot.com/report?masters=Chromium&bots=chromium-rel-linux&tests=sizes%2Fchrome-si&checked=all&start_rev=236784&end_rev=237073

TBR=enne@chromium.org,tomhudson@chromium.org,wiltzius@chromium.org,danakj@chromium.org,jamesr@chromium.org,vivekg@chromium.org,siva.gunturi@samsung.com
NOTREECHECKS=true
NOTRY=true

[Roger Tawa OOO till Jul 10th, 2013-11-25 15:01:04]

Created Revert of Generic version of caching cc switches  to avoid  searching of
switches on each function call

[sivag, 2013-11-25 15:23:07]

On 2013/11/25 15:01:04, Roger Tawa wrote:
> Created Revert of Generic version of caching cc switches  to avoid  searching
of
> switches on each function call

Hi Roger , What kind of regression is this causing ..How can i test this..?

[Tom Hudson, 2013-11-25 15:43:18]

You created a static object at file-level scope. That's a no-no; it causes
severe start-up time hits on some platforms.
We should have caught this in review of the initial patch. :(

[vivekg, 2013-11-25 15:49:11]

On 2013/11/25 15:43:18, Tom Hudson wrote:
> You created a static object at file-level scope. That's a no-no; it causes
> severe start-up time hits on some platforms.
> We should have caught this in review of the initial patch. :(

Is having a static local variable in a function and returning a reference to it
for further modification allowed? This way we will avoid the global variable
with lazy initialization at the function call.

[tfarina, 2013-11-26 15:34:28]

Not sure if it happens in this case, but base/ has LazyInstance API. You may
want to look into that to see if it fits your needs.

[danakj, 2013-11-26 15:57:35]

On 2013/11/25 15:43:18, Tom Hudson wrote:
> You created a static object at file-level scope. That's a no-no; it causes
> severe start-up time hits on some platforms.
> We should have caught this in review of the initial patch. :(

Sorry for missing this, I thought a POD struct would be the same as a POD bool,
but I guess I mixed that up.