Add missing EnsureOpenSSLInit to QuicStreamFactory.

Add missing EnsureOpenSSLInit to QuicStreamFactory.

BUG=512783
Committed: https://crrev.com/78406baf2210d015caa04c59918e77142a487193
Cr-Commit-Position: refs/heads/master@{#339937}

[davidben, 2015-07-22 14:07:22]

davidben@chromium.org changed reviewers:
+ rch@chromium.org

[Alexander Potapenko, 2015-07-22 14:19:39]

On 2015/07/22 14:07:22, David Benjamin wrote:

I've started a tryjob on linux_chromium_tsan_rel_ng.

[Alexander Potapenko, 2015-07-22 15:07:32]

On 2015/07/22 14:19:39, Alexander Potapenko wrote:
> On 2015/07/22 14:07:22, David Benjamin wrote:
> 
> I've started a tryjob on linux_chromium_tsan_rel_ng.

TSan looks happy: https://chromium-swarm.appspot.com/user/task/28cf1c77957b8a10

[Ryan Hamilton, 2015-07-22 18:03:24]

LGTM (even if it's a bit magic :>) I wonder if we should put this invocation
higher in the stack somewhere? Perhaps in io_thread.cc or some such.

[davidben, 2015-07-22 18:05:13]

On 2015/07/22 18:03:24, Ryan Hamilton wrote:
> LGTM (even if it's a bit magic :>) I wonder if we should put this invocation
> higher in the stack somewhere? Perhaps in io_thread.cc or some such.

We've tended to put Ensure(OpenSSL|NSS)Init invocations close to code that uses
them, rather than have it happen early in startup. So this is at least
consistent, though it is certainly precarious; I'm sure we have lots of places
where things accidentally worked due to other code.

[davidben, 2015-07-22 18:05:29]

The CQ bit was checked by davidben@chromium.org

[commit-bot: I haz the power, 2015-07-22 18:05:46]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1244413002/20001

[commit-bot: I haz the power, 2015-07-22 19:37:34]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-07-22 19:38:18]

Patchset 2 (id:??) landed as
https://crrev.com/78406baf2210d015caa04c59918e77142a487193
Cr-Commit-Position: refs/heads/master@{#339937}