Prevents a segfault in the noise suppression code on ARMv7 with NEON instructions and Mips platform…

Prevents a segfault in the noise suppression code on ARMv7 with NEON instructions and Mips platforms.

Integer wraparound when casting from int32 to int16 can cause invalid array indices to be accessed.
Fix for wraparound issue.

BUG=webrtc:5781
Committed: https://crrev.com/3bc9566f5aa0fae0015419c02f73cf22d6d55c2f
Cr-Commit-Position: refs/heads/master@{#12449}

[ivoc, 2016-04-15 08:04:58]

ivoc@webrtc.org changed reviewers:
+ hlundin@google.com

[ivoc, 2016-04-15 08:04:58]

Please have a look :)

[hlundin-webrtc, 2016-04-15 08:20:06]

henrik.lundin@webrtc.org changed reviewers:
+ henrik.lundin@webrtc.org
- hlundin@google.com

[hlundin-webrtc, 2016-04-15 08:20:06]

(Changing instantiation of myself.)

[hlundin-webrtc, 2016-04-15 08:28:31]

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/ns/nsx_core_c.c (right):

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/ns/nsx_core_c.c:99: if (tmp32no1 < (16<<14) &&
tmp32no1 >= 0) {
Add spaces around <<
Here and below.

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/ns/nsx_core_mips.c (right):

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/ns/nsx_core_mips.c:166: if (tmp32no1 < (16<<14)
&& tmp32no1 >= 0) {
This is wrong. The old code deals with tmpU32no1, not tmp32no1. And you should
then not check for non-negativity.

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/ns/nsx_core_mips.c:213: if (tmp32no1 < (16<<14)
&& tmp32no1 >= 0) {
Same here. Wrong variable.

[ivoc, 2016-04-15 08:37:25]

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/ns/nsx_core_mips.c (right):

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/ns/nsx_core_mips.c:166: if (tmp32no1 < (16<<14)
&& tmp32no1 >= 0) {
On 2016/04/15 08:28:31, hlundin-webrtc wrote:
> This is wrong. The old code deals with tmpU32no1, not tmp32no1. And you should
> then not check for non-negativity.

Oops, you're right. These wonderfully descriptive variable names make it easy to
miss.

https://codereview.webrtc.org/1894483002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/ns/nsx_core_mips.c:213: if (tmp32no1 < (16<<14)
&& tmp32no1 >= 0) {
On 2016/04/15 08:28:31, hlundin-webrtc wrote:
> Same here. Wrong variable.

Fixed.

[hlundin-webrtc, 2016-04-15 09:09:54]

lgtm

[ivoc, 2016-04-15 09:12:34]

The CQ bit was checked by ivoc@webrtc.org

[commit-bot: I haz the power, 2016-04-15 09:12:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1894483002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1894483002/20001

[commit-bot: I haz the power, 2016-04-15 11:13:10]

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

[commit-bot: I haz the power, 2016-04-15 11:13:11]

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

[ivoc, 2016-04-19 09:58:37]

The CQ bit was checked by ivoc@webrtc.org

[commit-bot: I haz the power, 2016-04-19 09:58:47]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1894483002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1894483002/20001

[commit-bot: I haz the power, 2016-04-19 11:59:19]

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

[commit-bot: I haz the power, 2016-04-19 11:59:20]

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

[ivoc, 2016-04-20 14:07:44]

The CQ bit was checked by ivoc@webrtc.org

[commit-bot: I haz the power, 2016-04-20 14:07:55]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1894483002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1894483002/20001

[commit-bot: I haz the power, 2016-04-20 15:25:26]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-20 15:29:23]

Description was changed from

==========
Prevents a segfault in the noise suppression code on ARMv7 with NEON
instructions and Mips platforms.

Integer wraparound when casting from int32 to int16 can cause invalid array
indices to be accessed.
Fix for wraparound issue.

BUG=webrtc:5781
==========

to

==========
Prevents a segfault in the noise suppression code on ARMv7 with NEON
instructions and Mips platforms.

Integer wraparound when casting from int32 to int16 can cause invalid array
indices to be accessed.
Fix for wraparound issue.

BUG=webrtc:5781

Committed: https://crrev.com/3bc9566f5aa0fae0015419c02f73cf22d6d55c2f
Cr-Commit-Position: refs/heads/master@{#12449}
==========

[commit-bot: I haz the power, 2016-04-20 15:29:24]

Patchset 2 (id:??) landed as
https://crrev.com/3bc9566f5aa0fae0015419c02f73cf22d6d55c2f
Cr-Commit-Position: refs/heads/master@{#12449}