xunjieli@ could you please look into this.please feel free to re-assigned back if needed. thanks in advance !

Thanks! I will look into it asap.
cc-ing my reviewers so they know I am on it.

The bug is in the new net::BrotliSourceStream. I will upload a fix shortly and will send it for review.

Proposed fix https://codereview.chromium.org/2451833004

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/c4780eac0d48c8b57244a9cbcf8bfe7271b532f4

commit c4780eac0d48c8b57244a9cbcf8bfe7271b532f4
Author: xunjieli <xunjieli@chromium.org>
Date: Wed Oct 26 19:05:29 2016

Fix net::BrotliSourceStream to ignore trailing data.

When net::BrotliSourceStream is given a valid input but
with trailing data, BrotliSourceStream will consume valid
input but not the trailing data. This triggers a DCHECK in
FilterSourceStream which asserts that either all bytes
are consumed or bytes written is not 0.

This CL matches the original implementation
(net::BrotliFilter) more closely.
If there is any trailing data, the data will be ignored.

This CL adds a regression test which without the patch
will trigger the DCHECK.

BUG= 659311 

Review-Url: https://codereview.chromium.org/2451833004
Cr-Commit-Position: refs/heads/master@{#427770}

[modify] https://crrev.com/c4780eac0d48c8b57244a9cbcf8bfe7271b532f4/net/filter/brotli_source_stream.cc
[modify] https://crrev.com/c4780eac0d48c8b57244a9cbcf8bfe7271b532f4/net/filter/brotli_source_stream_unittest.cc

ClusterFuzz has detected this issue as fixed in range 427755:427846.

Detailed report: https://cluster-fuzz.appspot.com/testcase?key=5598807900553216

Fuzzer: libfuzzer_net_url_request_fuzzer
Job Type: libfuzzer_chrome_asan_debug
Platform Id: linux

Crash Type: UNKNOWN
Crash Address: 0x03e900001fe2
Crash State:
  base::debug::DebugBreak
  net::FilterSourceStream::DoFilterData
  net::FilterSourceStream::DoLoop
  
Regressed: https://cluster-fuzz.appspot.com/revisions?job=libfuzzer_chrome_asan_debug&range=427323:427365
Fixed: https://cluster-fuzz.appspot.com/revisions?job=libfuzzer_chrome_asan_debug&range=427755:427846

Minimized Testcase (0.04 Kb): https://cluster-fuzz.appspot.com/download/AMIfv94zSE-kF9p7_1gvQ4OkRF574rLqBCCqAVWxIBqqpsvHWUBHN5ZIC09tYWhUPbJzb7dvT99sKKlUMTOfVlfIMIWGuIs3XPQf21Vb3IUoiwGT5YpsD3bj7qUCkPBJpFd2sCcFrhpXYhoOl1SlLRycWL67ORJwpA?testcase_id=5598807900553216

See https://chromium.googlesource.com/chromium/src/+/master/testing/libfuzzer/reproducing.md for more information.

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

Removing EditIssue view restrictions from ClusterFuzz filed bugs. If you believe that this issue should still be restricted, please reapply the label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot