Remove use of _USE_32BIT_TIME_T

Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256
Committed: https://crrev.com/dc4a1cef180152d90464271dd1a0ba7beacf15fa
Cr-Commit-Position: refs/heads/master@{#386639}

[kjellander_chromium, 2016-04-05 11:25:10]

Description was changed from

==========
Remove use of _USE_32BIT_TIME_T

BUG=
==========

to

==========
Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256
==========

[kjellander_chromium, 2016-04-05 11:25:10]

kjellander@chromium.org changed reviewers:
+ tommi@chromium.org, torbjorng@chromium.org

[kjellander_chromium, 2016-04-05 11:31:20]

So, removing _USE_32BIT_TIME_T doesn't seem to affect the regular tests being
run in Chromium at least. 
Unfortunately there's no way to see how it affects our end-to-end tests that run
in our own chromium.webrtc[.fyi]  waterfalls. Should we commit it and see or is
it risky to do with the cut only a few days away?

[torbjorng, 2016-04-05 12:14:24]

I cannot really tell risky this is.  All (or most) matches seem to be in sqlite.
 Possibly more in \usr\include?

Perhaps bring this up with the sqlite OWNER (or grope around properly in sqlite
sources yourself).

[kjellander_chromium, 2016-04-06 12:53:26]

On 2016/04/05 12:14:24, torbjorng wrote:
> I cannot really tell risky this is.  All (or most) matches seem to be in
sqlite.
>  Possibly more in \usr\include?
> 
> Perhaps bring this up with the sqlite OWNER (or grope around properly in
sqlite
> sources yourself).

Is sqlite relevant? We cannot possibly depend on it from third_party/libjingle
so it can't be affected by this change, right? 

At least GN tells me this is its dependency tree:
 $ gn refs out/Default //third_party/sqlite
//components/history/core/browser:browser
//sql:sql
//sql:sql_unittests
//sql:test_support
//storage/browser:browser
//testing/libfuzzer/fuzzers:sqlite3_prepare_v2_fuzzer
//third_party/WebKit/Source/core:core_generated
//third_party/WebKit/Source/core:generated
//third_party/WebKit/Source/core:prerequisites
//third_party/WebKit/Source/modules:modules
//third_party/sqlite:sqlite_shell

[torbjorng, 2016-04-06 13:48:19]

Right. Let's forget sqlite then.

I tried to read up a bit about _USE_32BIT_TIME_T. Rumour has it that it affects
the definition of time_t and the APIs that use that type on Win32 (while Win64
enforces a 64-bit time_t). 

If you have the Microsoft compiler installed on a system, you should look for
logic confirming this in crtdefs.h.  (This file should live in the same
directory as stdio.h.)

Unfortunately I cannot tell if something in our code base depends on this being
set (or unset).

I'd suggest to put off this cleanup until after the 51 cut.

Ref: https://msdn.microsoft.com/en-us/library/4wacf567.aspx

[kjellander_chromium, 2016-04-06 15:15:19]

On 2016/04/06 13:48:19, torbjorng wrote:
> Right. Let's forget sqlite then.
> 
> I tried to read up a bit about _USE_32BIT_TIME_T. Rumour has it that it
affects
> the definition of time_t and the APIs that use that type on Win32 (while Win64
> enforces a 64-bit time_t). 
> 
> If you have the Microsoft compiler installed on a system, you should look for
> logic confirming this in crtdefs.h.  (This file should live in the same
> directory as stdio.h.)
> 
> Unfortunately I cannot tell if something in our code base depends on this
being
> set (or unset).
> 
> I'd suggest to put off this cleanup until after the 51 cut.
> 
> Ref: https://msdn.microsoft.com/en-us/library/4wacf567.aspx

I completely agree. We should wait with this until after the M51 cut. Or at
least have a Windows guru (tommi or someone else) speak up first.

[tommi (sloooow) - chröme, 2016-04-12 08:50:13]

let's land this now and see what happens. lgtm

[torbjorng, 2016-04-12 08:52:41]

lgtm

[kjellander_chromium, 2016-04-12 09:02:38]

The CQ bit was checked by kjellander@chromium.org

[commit-bot: I haz the power, 2016-04-12 09:02:47]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1862443003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1862443003/20001

[commit-bot: I haz the power, 2016-04-12 10:02:18]

Description was changed from

==========
Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256
==========

to

==========
Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256
==========

[commit-bot: I haz the power, 2016-04-12 10:02:19]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-12 10:03:50]

Description was changed from

==========
Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256
==========

to

==========
Remove use of _USE_32BIT_TIME_T

BUG=webrtc:4256

Committed: https://crrev.com/dc4a1cef180152d90464271dd1a0ba7beacf15fa
Cr-Commit-Position: refs/heads/master@{#386639}
==========

[commit-bot: I haz the power, 2016-04-12 10:03:50]

Patchset 2 (id:??) landed as
https://crrev.com/dc4a1cef180152d90464271dd1a0ba7beacf15fa
Cr-Commit-Position: refs/heads/master@{#386639}