Make SSLClientSocketFalseStartTest.NoSessionResumptionBeforeFinish not a no-op.

Make SSLClientSocketFalseStartTest.NoSessionResumptionBeforeFinish not a no-op.

The test actually didn't test anything. Calling sock1.reset() would cause the
socket to disconnect before tlslite completed the handshake. tlslite doesn't
insert the session into the session cache.

This is a preparatory CL to get the tests in shape before reworking the session
cache implementation.

BUG=454044
Committed: https://crrev.com/6528761cc85a3a17559ee689b0758da9baef76c5
Cr-Commit-Position: refs/heads/master@{#319756}

[davidben, 2015-03-09 21:04:30]

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

[davidben, 2015-03-09 21:04:31]

There's a more interesting CL in your future, but this to start with.

[Ryan Sleevi, 2015-03-09 21:56:12]

LGTM % wording

https://codereview.chromium.org/988223004/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_unittest.cc (right):

https://codereview.chromium.org/988223004/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_unittest.cc:3183: // cache.
Suggested reword:

// Continue to block the client (|sock1|) from processing the Finished
// message, but allow it to arrive on the socket. This ensures that
// from the server's point of view, it has completed the handshake and
// added the session to its session cache.
// The actual read on |sock1| will not complete until the Finished message
// is processed; however, pump the underlying transport so that it is
// read from the socket. This still has the potential to race, but is
// generally unlikely due to socket buffer sizes.

[davidben, 2015-03-09 22:00:38]

Patchset #2 (id:20001) has been deleted

[davidben, 2015-03-09 22:01:13]

https://codereview.chromium.org/988223004/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_unittest.cc (right):

https://codereview.chromium.org/988223004/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_unittest.cc:3183: // cache.
On 2015/03/09 21:56:12, Ryan Sleevi wrote:
> Suggested reword:
> 
> // Continue to block the client (|sock1|) from processing the Finished
> // message, but allow it to arrive on the socket. This ensures that
> // from the server's point of view, it has completed the handshake and
> // added the session to its session cache.
> // The actual read on |sock1| will not complete until the Finished message
> // is processed; however, pump the underlying transport so that it is
> // read from the socket. This still has the potential to race, but is
> // generally unlikely due to socket buffer sizes.

Done.

[davidben, 2015-03-09 22:01:27]

The CQ bit was checked by davidben@chromium.org

[davidben, 2015-03-09 22:01:27]

The patchset sent to the CQ was uploaded after l-g-t-m from rsleevi@chromium.org
Link to the patchset: https://codereview.chromium.org/988223004/#ps40001 (title:
"sleevi comments, also appease windows")

[commit-bot: I haz the power, 2015-03-09 22:02:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/988223004/40001

[commit-bot: I haz the power, 2015-03-09 23:10:21]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-03-09 23:11:03]

Patchset 2 (id:??) landed as
https://crrev.com/6528761cc85a3a17559ee689b0758da9baef76c5
Cr-Commit-Position: refs/heads/master@{#319756}