Fixing a Dr. memory failure in the remoting unit tests.

Fixing a Dr. memory failure in the remoting unit tests.

The test I checked in yesterday had a race condition in it which was detected
by the Dr. Memory bot (they run more slowly than the normal bots).  On a
normal bot, my runloop quits when the first (of ~11) message is written but
the other messages are processed and added to the vector by the time it is
checked.  On the Dr. Memory bot, this condition was exposed and the size check
failed.  The fix is to ensure that we are processing each message as it is
written/received and validating the size afterwards.

This is a reland of: https://codereview.chromium.org/1844423002/
Due to changes to the base files, I am sending this out as a new patch instead
of using the old CL.

BUG=599494
Committed: https://crrev.com/c705be10c12997d771afcd7c0eb301a311adfb8a
Cr-Commit-Position: refs/heads/master@{#385093}

[joedow, 2016-04-05 00:19:19]

joedow@chromium.org changed reviewers:
+ sergeyu@chromium.org

[joedow, 2016-04-05 00:19:20]

PTAL!

Thanks,
Joe

[Sergey Ulanov, 2016-04-05 00:39:38]

lgtm

https://codereview.chromium.org/1860693003/diff/1/remoting/host/security_key/...
File
remoting/host/security_key/remote_security_key_message_reader_impl_unittest.cc
(right):

https://codereview.chromium.org/1860693003/diff/1/remoting/host/security_key/...
remoting/host/security_key/remote_security_key_message_reader_impl_unittest.cc:86:
void RemoteSecurityKeyMessageReaderImplTest::Run(bool close_file) {
Instead of the close_file flag maybe add CloseFileAndRunLoop()?

[joedow, 2016-04-05 02:24:06]

Thanks!

https://codereview.chromium.org/1860693003/diff/1/remoting/host/security_key/...
File
remoting/host/security_key/remote_security_key_message_reader_impl_unittest.cc
(right):

https://codereview.chromium.org/1860693003/diff/1/remoting/host/security_key/...
remoting/host/security_key/remote_security_key_message_reader_impl_unittest.cc:86:
void RemoteSecurityKeyMessageReaderImplTest::Run(bool close_file) {
On 2016/04/05 00:39:37, Sergey Ulanov wrote:
> Instead of the close_file flag maybe add CloseFileAndRunLoop()?

Done.

[joedow, 2016-04-05 02:24:14]

The CQ bit was checked by joedow@chromium.org

[joedow, 2016-04-05 02:24:14]

The patchset sent to the CQ was uploaded after l-g-t-m from sergeyu@chromium.org
Link to the patchset: https://codereview.chromium.org/1860693003/#ps2 (title:
"Addressing feedback")

[commit-bot: I haz the power, 2016-04-05 02:24:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1860693003/2
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1860693003/2

[commit-bot: I haz the power, 2016-04-05 02:55:37]

Committed patchset #2 (id:2)

[commit-bot: I haz the power, 2016-04-05 02:56:41]

Description was changed from

==========
Fixing a Dr. memory failure in the remoting unit tests.

The test I checked in yesterday had a race condition in it which was detected
by the Dr. Memory bot (they run more slowly than the normal bots).  On a
normal bot, my runloop quits when the first (of ~11) message is written but
the other messages are processed and added to the vector by the time it is
checked.  On the Dr. Memory bot, this condition was exposed and the size check
failed.  The fix is to ensure that we are processing each message as it is
written/received and validating the size afterwards.

This is a reland of: https://codereview.chromium.org/1844423002/
Due to changes to the base files, I am sending this out as a new patch instead
of using the old CL.

BUG=599494
==========

to

==========
Fixing a Dr. memory failure in the remoting unit tests.

The test I checked in yesterday had a race condition in it which was detected
by the Dr. Memory bot (they run more slowly than the normal bots).  On a
normal bot, my runloop quits when the first (of ~11) message is written but
the other messages are processed and added to the vector by the time it is
checked.  On the Dr. Memory bot, this condition was exposed and the size check
failed.  The fix is to ensure that we are processing each message as it is
written/received and validating the size afterwards.

This is a reland of: https://codereview.chromium.org/1844423002/
Due to changes to the base files, I am sending this out as a new patch instead
of using the old CL.

BUG=599494

Committed: https://crrev.com/c705be10c12997d771afcd7c0eb301a311adfb8a
Cr-Commit-Position: refs/heads/master@{#385093}
==========

[commit-bot: I haz the power, 2016-04-05 02:56:42]

Patchset 2 (id:??) landed as
https://crrev.com/c705be10c12997d771afcd7c0eb301a311adfb8a
Cr-Commit-Position: refs/heads/master@{#385093}