Moved bits.h from O3D to Chrome base.

Moved bits.h from O3D to Chrome base.
TEST=none
BUG=none

[apatrick_chromium, 2009-11-05 22:15:27]

All I really did to these was change the formatting and naming to be consistent
with Chrome. I didn't bother with a svn copy because nothing of interest has
happened to these files since we moved from perforce.

[apatrick_chromium, 2009-11-05 22:16:04]

Oh and the reason for this is because the command buffer code is dependent on
it.

[jam, 2009-11-05 23:23:10]

We're generally conservative with what goes into base .  I don't feel I'm
qualified to say which new stuff should go in there, or whether this can be put
into an existing file, or whether this doesn't belong in base if only one module
uses it.  It's best to check with Brett or Darin, they have stronger opinions on
this.

[apatrick_chromium, 2009-11-05 23:26:52]

Changed reviewers to Brett and Darin.

[apatrick_chromium, 2009-11-11 01:03:25]

bump

[apatrick_chromium, 2009-11-11 01:03:42]

bump

http://codereview.chromium.org/373001

[brettw, 2009-11-11 23:48:58]

I think adding this to base is OK. I thought there was an existing math file we
could add this to but couldn't find one. Might as well start.

I didn't check the actual logic.

http://codereview.chromium.org/373001/diff/1/2
File base/bits.h (right):

http://codereview.chromium.org/373001/diff/1/2#newcode1
Line 1: // Copyright (c) 2006-2009 The Chromium Authors. All rights reserved.
Nit: "2009"

http://codereview.chromium.org/373001/diff/1/2#newcode16
Line 16: inline int Log2Floor(unsigned int n) {
How about using uint32 from basictypes here? Since this is specifically for "bit
manipulation" it seems more natural to specify how many bits there are (same
below).

[apatrick_chromium, 2009-11-12 00:02:54]

http://codereview.chromium.org/373001/diff/1/2
File base/bits.h (right):

http://codereview.chromium.org/373001/diff/1/2#newcode1
Line 1: // Copyright (c) 2006-2009 The Chromium Authors. All rights reserved.
On 2009/11/11 23:48:59, brettw wrote:
> Nit: "2009"

Done.

http://codereview.chromium.org/373001/diff/1/2#newcode16
Line 16: inline int Log2Floor(unsigned int n) {
On 2009/11/11 23:48:59, brettw wrote:
> How about using uint32 from basictypes here? Since this is specifically for
"bit
> manipulation" it seems more natural to specify how many bits there are (same
> below).

Done.

[apatrick_chromium, 2009-11-16 20:53:11]

Should I interpret your last comment as LGTM, Brett?

[gman, 2009-11-17 23:56:31]

LGTM