Add a CongestionController fuzzer.

Add a CongestionController fuzzer.

BUG=

Committed: https://crrev.com/bded44b79bf24340c34b4e788e30261665c0ba8a
Cr-Commit-Position: refs/heads/master@{#13497}

[stefan-webrtc, 2016-07-18 07:56:13]

stefan@webrtc.org changed reviewers:
+ pbos@webrtc.org

[stefan-webrtc, 2016-07-18 07:57:27]

Not sure if it's possible to have several fuzzers in one executable in a
reasonable way? If not maybe I should rename it to clarify it's the feedback
generation/receive-side that is being fuzzed?

[pbos-webrtc, 2016-07-18 08:02:52]

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD.gn
File webrtc/test/fuzzers/BUILD.gn (right):

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD...
webrtc/test/fuzzers/BUILD.gn:118: "//testing/gmock",
Can you try to build this in a chromium checkout? I'm not sure if gmock is
available under //testing there.

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD...
webrtc/test/fuzzers/BUILD.gn:119: "//testing/gtest",
Does this need gtest?

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
File webrtc/test/fuzzers/congestion_controller_fuzzer.cc (right):

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
webrtc/test/fuzzers/congestion_controller_fuzzer.cc:21: const size_t kMinSize =
sizeof(size_t) + sizeof(uint16_t) + sizeof(uint8_t);
Something like kMinPacketSize?

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
webrtc/test/fuzzers/congestion_controller_fuzzer.cc:22: if (size <
sizeof(int64_t) + sizeof(uint8_t) + sizeof(uint32_t) + kMinSize)
You can permit this to run with 0 packets as well, kMinSize is sufficient to
check at line 37. Then it has possibly better coverage (running process without
inserting packets).

[pbos-webrtc, 2016-07-18 08:05:35]

On 2016/07/18 07:57:27, stefan-webrtc (holmer) wrote:
> Not sure if it's possible to have several fuzzers in one executable in a
> reasonable way? If not maybe I should rename it to clarify it's the feedback
> generation/receive-side that is being fuzzed?

I think it's reasonable to make separate binaries of them with a common
source_set. I'd rename it so it doesn't sound like all of CongestionController
is already fuzzed.

[stefan-webrtc, 2016-07-18 15:46:48]

Comments addressed, still catches the bug.

[stefan-webrtc, 2016-07-18 15:46:56]

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD.gn
File webrtc/test/fuzzers/BUILD.gn (right):

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD...
webrtc/test/fuzzers/BUILD.gn:118: "//testing/gmock",
On 2016/07/18 08:02:52, pbos-webrtc wrote:
> Can you try to build this in a chromium checkout? I'm not sure if gmock is
> available under //testing there.

Removed gmock and gtest

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/BUILD...
webrtc/test/fuzzers/BUILD.gn:119: "//testing/gtest",
On 2016/07/18 08:02:52, pbos-webrtc wrote:
> Does this need gtest?

Done.

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
File webrtc/test/fuzzers/congestion_controller_fuzzer.cc (right):

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
webrtc/test/fuzzers/congestion_controller_fuzzer.cc:21: const size_t kMinSize =
sizeof(size_t) + sizeof(uint16_t) + sizeof(uint8_t);
On 2016/07/18 08:02:52, pbos-webrtc wrote:
> Something like kMinPacketSize?

Roughly, it also includes the size of the packet. But sure, it's a better name
:)

https://codereview.webrtc.org/2157783002/diff/20001/webrtc/test/fuzzers/conge...
webrtc/test/fuzzers/congestion_controller_fuzzer.cc:22: if (size <
sizeof(int64_t) + sizeof(uint8_t) + sizeof(uint32_t) + kMinSize)
On 2016/07/18 08:02:52, pbos-webrtc wrote:
> You can permit this to run with 0 packets as well, kMinSize is sufficient to
> check at line 37. Then it has possibly better coverage (running process
without
> inserting packets).

Good suggestion, fixing.

[pbos-webrtc, 2016-07-18 15:50:21]

lgtm, consider moving the null implementations to the other headers

[stefan-webrtc, 2016-07-18 15:52:56]

Rename

[stefan-webrtc, 2016-07-18 15:53:46]

Also change BUILD.gn

[stefan-webrtc, 2016-07-18 15:54:20]

The CQ bit was checked by stefan@webrtc.org

[stefan-webrtc, 2016-07-18 15:54:20]

The patchset sent to the CQ was uploaded after l-g-t-m from pbos@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2157783002/#ps80001 (title:
"Also change BUILD.gn")

[commit-bot: I haz the power, 2016-07-18 15:54:28]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2016-07-18 16:26:10]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-07-18 16:26:21]

Description was changed from

==========
Add a CongestionController fuzzer.

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

to

==========
Add a CongestionController fuzzer.

BUG=

Committed: https://crrev.com/bded44b79bf24340c34b4e788e30261665c0ba8a
Cr-Commit-Position: refs/heads/master@{#13497}
==========

[commit-bot: I haz the power, 2016-07-18 16:26:23]

Patchset 5 (id:??) landed as
https://crrev.com/bded44b79bf24340c34b4e788e30261665c0ba8a
Cr-Commit-Position: refs/heads/master@{#13497}