Fix RowSizeForBufferFormatChecked use of base/numerics

Fix RowSizeForBufferFormatChecked use of base/numerics

R=danakj@chromium.org
TBR=danakj@chromium.org
BUG=668713
Committed: https://crrev.com/8794c4be74f56e495da28f388ee30098dc16eb88
Cr-Commit-Position: refs/heads/master@{#435341}

[jschuh, 2016-11-30 17:21:52]

The CQ bit was checked by jschuh@chromium.org

[commit-bot: I haz the power, 2016-11-30 17:22:22]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[jschuh, 2016-11-30 17:25:05]

danakj@ - This cleanup in preparation for landing this CL:
https://codereview.chromium.org/2528243002/

It shouldn't alter the current behavior of the code, hence the TBR.

[commit-bot: I haz the power, 2016-11-30 18:23:06]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1480526512251440, "parent_rev":
"87b57fb60d759dffd92e36a044b7013622f021b0", "commit_rev":
"f728de90c163416789d85ce786f73b26eb8079ff"}

[commit-bot: I haz the power, 2016-11-30 18:23:34]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-11-30 18:28:03]

Description was changed from

==========
Fix RowSizeForBufferFormatChecked use of base/numerics

R=danakj@chromium.org
TBR=danakj@chromium.org
BUG=668713
==========

to

==========
Fix RowSizeForBufferFormatChecked use of base/numerics

R=danakj@chromium.org
TBR=danakj@chromium.org
BUG=668713

Committed: https://crrev.com/8794c4be74f56e495da28f388ee30098dc16eb88
Cr-Commit-Position: refs/heads/master@{#435341}
==========

[commit-bot: I haz the power, 2016-11-30 18:28:03]

Patchset 1 (id:??) landed as
https://crrev.com/8794c4be74f56e495da28f388ee30098dc16eb88
Cr-Commit-Position: refs/heads/master@{#435341}

[danakj, 2016-11-30 23:13:20]

https://codereview.chromium.org/2545523002/diff/1/ui/gfx/buffer_format_util.cc
File ui/gfx/buffer_format_util.cc (right):

https://codereview.chromium.org/2545523002/diff/1/ui/gfx/buffer_format_util.c...
ui/gfx/buffer_format_util.cc:56: *size_in_bytes = (checked_size &
~0x3).ValueOrDie();
I'm not clear why this is better. bitmask operations will not overflow.

[jschuh, 2016-11-30 23:21:35]

On 2016/11/30 23:13:20, danakj wrote:
> https://codereview.chromium.org/2545523002/diff/1/ui/gfx/buffer_format_util.cc
> File ui/gfx/buffer_format_util.cc (right):
> 
>
https://codereview.chromium.org/2545523002/diff/1/ui/gfx/buffer_format_util.c...
> ui/gfx/buffer_format_util.cc:56: *size_in_bytes = (checked_size &
> ~0x3).ValueOrDie();
> I'm not clear why this is better. bitmask operations will not overflow.

You are totally correct. The current code isn't wrong, and there's no added
checking on the "checked" bitwise operations (so there shouldn't be any added
overhead). But, the change clears a blocker for this CL:
https://codereview.chromium.org/2528243002/

The gist is that the CheckedNumeric value functions will be returning a proxy
class instead of the raw value, so it can detect truncations. But, I didn't want
to add math operations to the proxy class, because I want to restrict the math
to the CheckedNumeric class.