Fix two UBSan warnings in NetEq

Fix two UBSan warnings in NetEq

Both were related to very large jumps in RTP timestamps.

BUG=webrtc:5488
Committed: https://crrev.com/07c51e3f79242b7750cc8719d804c65107d2271d
Cr-Commit-Position: refs/heads/master@{#11569}

[hlundin-webrtc, 2016-02-10 13:19:43]

The CQ bit was checked by henrik.lundin@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-10 13:19:55]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1685103002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1685103002/1

[hlundin-webrtc, 2016-02-10 13:37:32]

henrik.lundin@webrtc.org changed reviewers:
+ kwiberg@webrtc.org

[hlundin-webrtc, 2016-02-10 13:37:33]

Karl,

Please, take a look at this CL.

Thanks!
/HL

[commit-bot: I haz the power, 2016-02-10 14:35:20]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-02-10 14:35:21]

Dry run: This issue passed the CQ dry run.

[kwiberg-webrtc, 2016-02-10 15:14:21]

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
File webrtc/modules/audio_coding/neteq/delay_manager.cc (right):

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
webrtc/modules/audio_coding/neteq/delay_manager.cc:101:
rtc::saturated_cast<int>(1000 * packet_len_samp / sample_rate_hz);
Am I guessing right when I think that this calculation is guaranteed to never
overflow in real life? packet_len_samp is at most 2^32 - 1, and sample_rate_hz >
2000.

Hmm. If so, a checked_cast may be better?

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
File webrtc/modules/audio_coding/neteq/timestamp_scaler.cc (right):

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
webrtc/modules/audio_coding/neteq/timestamp_scaler.cc:82: return internal_ref_;
This sort of math is scary. It's hard to verify that no operation will overflow.
It shouldn't be worse than before, though. :-)

Whenever I think too hard about this, I have an urge to use these gcc builtins
to make a small library...:
https://gcc.gnu.org/onlinedocs/gcc/Integer-Overflow-Builtins.html

[hlundin-webrtc, 2016-02-10 15:35:04]

Change to checked_cast

[hlundin-webrtc, 2016-02-10 15:36:13]

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
File webrtc/modules/audio_coding/neteq/delay_manager.cc (right):

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
webrtc/modules/audio_coding/neteq/delay_manager.cc:101:
rtc::saturated_cast<int>(1000 * packet_len_samp / sample_rate_hz);
On 2016/02/10 15:14:21, kwiberg-webrtc wrote:
> Am I guessing right when I think that this calculation is guaranteed to never
> overflow in real life? packet_len_samp is at most 2^32 - 1, and sample_rate_hz
>
> 2000.
> 
> Hmm. If so, a checked_cast may be better?

Done.

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
File webrtc/modules/audio_coding/neteq/timestamp_scaler.cc (right):

https://codereview.webrtc.org/1685103002/diff/1/webrtc/modules/audio_coding/n...
webrtc/modules/audio_coding/neteq/timestamp_scaler.cc:82: return internal_ref_;
On 2016/02/10 15:14:21, kwiberg-webrtc wrote:
> This sort of math is scary. It's hard to verify that no operation will
overflow.
> It shouldn't be worse than before, though. :-)
> 
> Whenever I think too hard about this, I have an urge to use these gcc builtins
> to make a small library...:
> https://gcc.gnu.org/onlinedocs/gcc/Integer-Overflow-Builtins.html

Acknowledged.

[kwiberg-webrtc, 2016-02-11 09:53:46]

lgtm

[hlundin-webrtc, 2016-02-11 10:35:30]

The CQ bit was checked by henrik.lundin@webrtc.org

[commit-bot: I haz the power, 2016-02-11 10:35:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1685103002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1685103002/20001

[commit-bot: I haz the power, 2016-02-11 11:35:48]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-11 11:35:48]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-11 11:36:13]

Description was changed from

==========
Fix two UBSan warnings in NetEq

Both were related to very large jumps in RTP timestamps.

BUG=webrtc:5488
==========

to

==========
Fix two UBSan warnings in NetEq

Both were related to very large jumps in RTP timestamps.

BUG=webrtc:5488

Committed: https://crrev.com/07c51e3f79242b7750cc8719d804c65107d2271d
Cr-Commit-Position: refs/heads/master@{#11569}
==========

[commit-bot: I haz the power, 2016-02-11 11:36:13]

Patchset 2 (id:??) landed as
https://crrev.com/07c51e3f79242b7750cc8719d804c65107d2271d
Cr-Commit-Position: refs/heads/master@{#11569}