win/clang: Roll clang 239674:242415

win/clang: Roll clang 239674:242415

32-bit SEH, member pointers with virtual bases, inttypes.h warning fix...

BUG=509256, 82385
R=hans@chromium.org

Committed: https://chromium.googlesource.com/chromium/src/+/2776ee378ee7d309d5d837e1e4405460403556ff

Committed: https://chromium.googlesource.com/chromium/src/+/0d1b17f598c72f8672c5223439735ece5e6860f0

[Nico, 2015-07-13 18:03:08]

thakis@chromium.org changed reviewers:
+ hans@chromium.org

[hans, 2015-07-13 18:07:15]

lgtm

[Nico, 2015-07-15 15:03:18]

Committed patchset #3 (id:40001) manually as
2776ee378ee7d309d5d837e1e4405460403556ff (presubmit successful).

[Nico, 2015-07-15 19:36:44]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/1240813002/ by thakis@chromium.org.

The reason for reverting is: Looks like this caused http://crbug.com/510555.

[Nico, 2015-07-16 19:38:47]

patch set 3 goes to 242415 (and unconditionally changes VERSION to 3.8.0)

[hans, 2015-07-16 19:41:13]

On 2015/07/16 19:38:47, Nico wrote:
> patch set 3 goes to 242415 (and unconditionally changes VERSION to 3.8.0)

lgtm

Maybe mention Michael's size improvement too?

[Nico, 2015-07-16 19:42:30]

Committed patchset #4 (id:60001) manually as
0d1b17f598c72f8672c5223439735ece5e6860f0 (presubmit successful).

[Nico, 2015-07-16 19:45:54]

On 2015/07/16 19:41:13, hans wrote:
> On 2015/07/16 19:38:47, Nico wrote:
> > patch set 3 goes to 242415 (and unconditionally changes VERSION to 3.8.0)
> 
> lgtm
> 
> Maybe mention Michael's size improvement too?

Well, they were what caused 242130 to be reverted and aren't in 242415 (242415
is in fact the revision that reverted mkuper's change), so… :-)

(On the plus side, we'll see if this roll improves sizes too, and from the diff
between the improvement of the 239674:242130 roll and this roll we can figure
out how much of the win is due to mkuper's patch)

[hans, 2015-07-16 19:50:45]

On 2015/07/16 19:45:54, Nico wrote:
> On 2015/07/16 19:41:13, hans wrote:
> > On 2015/07/16 19:38:47, Nico wrote:
> > > patch set 3 goes to 242415 (and unconditionally changes VERSION to 3.8.0)
> > 
> > lgtm
> > 
> > Maybe mention Michael's size improvement too?
> 
> Well, they were what caused 242130 to be reverted and aren't in 242415 (242415
> is in fact the revision that reverted mkuper's change), so… :-)
> 
> (On the plus side, we'll see if this roll improves sizes too, and from the
diff
> between the improvement of the 239674:242130 roll and this roll we can figure
> out how much of the win is due to mkuper's patch)

Oh, I see.