Add functions to interact with ASan and MSan, and some sample uses

Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the try bots when I tried to land this CL (a test was passing uninitialized packet data to NetEq).

BUG=chromium:617124
Committed: https://crrev.com/ac554eebb9365ca1bb383b62bb436ff3ae470bd8
Cr-Commit-Position: refs/heads/master@{#14034}

[kwiberg-webrtc, 2016-08-30 19:17:39]

Patchset #1 (id:1) has been deleted

[kwiberg-webrtc, 2016-08-30 19:18:57]

Patchset #1 (id:20001) has been deleted

[kwiberg-webrtc, 2016-08-30 19:30:19]

Description was changed from

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124.

BUG=chromium:617124
==========

to

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in
neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the
try bots when I tried to land this CL (a test was passing uninitialized packet
data to NetEq).

BUG=chromium:617124
==========

[kwiberg-webrtc, 2016-08-30 19:40:24]

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

[kwiberg-webrtc, 2016-08-30 19:40:25]

Primary review duty split:

hlundin@: common_audio, audio_coding
tommi@: base

But both of you probably need to look at the entire CL to form an opinion.

Basically, I used these new functions as a debug aid when chasing fuzzer bugs,
and had the idea that like [D]CHECKs, these sanitizer calls seem like they'd be
useful to actually put in the code permanently.

And as with [D]CHECK, in addition to whether the new tool is useful in general,
there's also the question of whether each individual call is worth the clutter
or not.

[kwiberg-webrtc, 2016-08-30 19:55:28]

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
File webrtc/common_audio/signal_processing/filter_ma_fast_q12.c (right):

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
webrtc/common_audio/signal_processing/filter_ma_fast_q12.c:32: B_length + length
- 1);
This is the sort of rtc_MsanCheckInitialized() call that one doesn't want to
see, since it shows that the code was written by crazy people who were
nevertheless brilliant since they got it to mostly work. However, removing it
would be like shooting the messenger and sweeping the body under the rug.

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/modules/audio_codi...
File webrtc/modules/audio_coding/codecs/ilbc/create_augmented_vec.c (right):

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/codecs/ilbc/create_augmented_vec.c:45: index +
interp_len);
This is another example of a rtc_MsanCheckInitialized() call that one would
really rather not have, but which is probably best to leave in given that it
unfortunately describes the way the surrounding code works.

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/modules/audio_codi...
File webrtc/modules/audio_coding/neteq/neteq_impl.cc (right):

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/neteq_impl.cc:138:
rtc::MsanCheckInitialized(payload);
This is the sort of rtc::MsanCheckInitialized() call one would like to see. It's
very easy to see what it does.

[tommi, 2016-08-31 18:18:55]

very nice! lgtm.  oh and thanks for the background :)

[hlundin-webrtc, 2016-09-01 13:41:35]

LGTM. But we should probably be a bit cautious before using it in every corner
of the codebase.

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
File webrtc/common_audio/signal_processing/filter_ma_fast_q12.c (right):

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
webrtc/common_audio/signal_processing/filter_ma_fast_q12.c:32: B_length + length
- 1);
On 2016/08/30 19:55:28, kwiberg-webrtc wrote:
> This is the sort of rtc_MsanCheckInitialized() call that one doesn't want to
> see, since it shows that the code was written by crazy people who were
> nevertheless brilliant since they got it to mostly work. However, removing it
> would be like shooting the messenger and sweeping the body under the rug.

What?! You don't like functions that use memory "to the left" of the data
pointer given?

[kwiberg-webrtc, 2016-09-01 19:41:53]

The CQ bit was checked by kwiberg@webrtc.org

[commit-bot: I haz the power, 2016-09-01 19:42:07]

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, 2016-09-01 21:42:36]

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

[commit-bot: I haz the power, 2016-09-01 21:42:37]

Try jobs failed on following builders:
  android_arm64_rel on master.tryserver.webrtc (JOB_TIMED_OUT, no build URL)
  android_rel on master.tryserver.webrtc (JOB_TIMED_OUT, no build URL)

[kwiberg-webrtc, 2016-09-02 07:30:16]

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
File webrtc/common_audio/signal_processing/filter_ma_fast_q12.c (right):

https://codereview.webrtc.org/2293893002/diff/40001/webrtc/common_audio/signa...
webrtc/common_audio/signal_processing/filter_ma_fast_q12.c:32: B_length + length
- 1);
On 2016/09/01 13:41:35, hlundin-webrtc wrote:
> On 2016/08/30 19:55:28, kwiberg-webrtc wrote:
> > This is the sort of rtc_MsanCheckInitialized() call that one doesn't want to
> > see, since it shows that the code was written by crazy people who were
> > nevertheless brilliant since they got it to mostly work. However, removing
it
> > would be like shooting the messenger and sweeping the body under the rug.
> 
> What?! You don't like functions that use memory "to the left" of the data
> pointer given?

Well, as long as you don't document what memory range your function accesses I
guess it's OK.

[kwiberg-webrtc, 2016-09-02 07:30:26]

The CQ bit was checked by kwiberg@webrtc.org

[commit-bot: I haz the power, 2016-09-02 07:30:39]

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, 2016-09-02 07:39:38]

Description was changed from

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in
neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the
try bots when I tried to land this CL (a test was passing uninitialized packet
data to NetEq).

BUG=chromium:617124
==========

to

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in
neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the
try bots when I tried to land this CL (a test was passing uninitialized packet
data to NetEq).

BUG=chromium:617124
==========

[commit-bot: I haz the power, 2016-09-02 07:39:39]

Committed patchset #1 (id:40001)

[commit-bot: I haz the power, 2016-09-02 07:39:42]

Description was changed from

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in
neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the
try bots when I tried to land this CL (a test was passing uninitialized packet
data to NetEq).

BUG=chromium:617124
==========

to

==========
Add functions to interact with ASan and MSan, and some sample uses

The sample uses are from when I debugged bug 617124. The change in
neteq_network_stats_unittest.cc is a fix for a minor unrelated bug found by the
try bots when I tried to land this CL (a test was passing uninitialized packet
data to NetEq).

BUG=chromium:617124

Committed: https://crrev.com/ac554eebb9365ca1bb383b62bb436ff3ae470bd8
Cr-Commit-Position: refs/heads/master@{#14034}
==========

[commit-bot: I haz the power, 2016-09-02 07:39:45]

Patchset 1 (id:??) landed as
https://crrev.com/ac554eebb9365ca1bb383b62bb436ff3ae470bd8
Cr-Commit-Position: refs/heads/master@{#14034}