Automatically applying components based on crash stacktrace and information from OWNERS files.

If this is incorrect, please apply the Test-Predator-Wrong-Components label.

Running the clusterfuzz repro tool fails to reproduce this issue.  it gets a different issue:

New crash type: Pointer-overflow
New crash state:
  SkRasterPipelineBlitter::blitMask
  blitClippedMask
  draw_nine_clipped

However, the fact that we're computing the tail time of an unstable IIR filter in the repro test case makes it pretty clear that we are in fact trying to convert a double float infinity value to a single-float.  In this case, the conversion is ok, but we can fix this (mostly) in WebAudio.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/949538bb77dce13f23abe328338f50f39c6350e0

commit 949538bb77dce13f23abe328338f50f39c6350e0
Author: Raymond Toy <rtoy@chromium.org>
Date: Tue May 01 17:51:27 2018

Bypass tail time computation if IIR filter is unstable

If the IIRFilter is known to be unstable, we can bypass the
computation of the tail time because we know a priori that the impulse
response won't converge to 0.

So, in construction of the IIRFilterNode where we determine if the
filter is stable or not, pass a stability flag to the IIRProcessor.
Then computation of the tail time can check the IIRProcess to see if
the filter is stable or not.  If not, just return the max tail time
value.

Bug:  837872 
Change-Id: I9f0e030c2a9e5a5b5635c964214fdafdd57e8668
Reviewed-on: https://chromium-review.googlesource.com/1035722
Reviewed-by: Hongchan Choi <hongchan@chromium.org>
Commit-Queue: Raymond Toy <rtoy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#555105}
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/modules/webaudio/iir_filter_node.cc
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/modules/webaudio/iir_filter_node.h
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/modules/webaudio/iir_processor.cc
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/modules/webaudio/iir_processor.h
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/modules/webaudio/iirdsp_kernel.cc
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/platform/audio/iir_filter.cc
[modify] https://crrev.com/949538bb77dce13f23abe328338f50f39c6350e0/third_party/blink/renderer/platform/audio/iir_filter.h

ClusterFuzz has detected this issue as fixed in range 555098:555107.

Detailed report: https://clusterfuzz.com/testcase?key=5330908266364928

Fuzzer: inferno_twister
Job Type: linux_ubsan_chrome
Platform Id: linux

Crash Type: Float-cast-overflow
Crash Address: 
Crash State:
  blink::IIRFilter::Process
  blink::IIRFilter::TailTime
  blink::IIRDSPKernel::IIRDSPKernel
  
Sanitizer: undefined (UBSAN)

Regressed: https://clusterfuzz.com/revisions?job=linux_ubsan_chrome&range=552707:552711
Fixed: https://clusterfuzz.com/revisions?job=linux_ubsan_chrome&range=555098:555107

Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=5330908266364928

See https://github.com/google/clusterfuzz-tools for more information.

If you suspect that the result above is incorrect, try re-doing that job on the test case report page.

ClusterFuzz testcase 5330908266364928 is verified as fixed, so closing issue as verified.

If this is incorrect, please add ClusterFuzz-Wrong label and re-open the issue.