With reference to the  Issue 902762 , assigning it to mstensho@

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b19fca3183f1af6e281043f2eba79fdb8e04c0f4

commit b19fca3183f1af6e281043f2eba79fdb8e04c0f4
Author: Morten Stenshorne <mstensho@chromium.org>
Date: Mon Dec 03 10:11:06 2018

Consistently round table row heights.

Do in LogicalHeightForRow() as in CalcRowLogicalHeight() - i.e. round
instead of floor, to convert from LayoutUnit to int. We calculated the
pagination strut incorrectly for the line that followed, because it was
based on the floored LayoutUnit height of the preceding table row.
Then, later on, it got adjusted (rounded from LayoutUnit), and the
balancing code got confused, because the line had a pagination strut,
but didn't appear to be exactly at the start of a column (there should
be no strut in such a case).

Bug:  908247 
Change-Id: I11fa5430b9ff6e83b20dc25ce1d945e54b80f6ea
Reviewed-on: https://chromium-review.googlesource.com/c/1356551
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: David Grogan <dgrogan@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#613033}
[modify] https://crrev.com/b19fca3183f1af6e281043f2eba79fdb8e04c0f4/third_party/blink/renderer/core/layout/layout_table_section.cc
[modify] https://crrev.com/b19fca3183f1af6e281043f2eba79fdb8e04c0f4/third_party/blink/web_tests/TestExpectations
[add] https://crrev.com/b19fca3183f1af6e281043f2eba79fdb8e04c0f4/third_party/blink/web_tests/external/wpt/css/css-multicol/balance-table-with-fractional-height-row.html

ClusterFuzz has detected this issue as fixed in range 613032:613033.

Detailed report: https://clusterfuzz.com/testcase?key=4999503415083008

Fuzzer: marty_html_twiddler
Job Type: linux_debug_chrome
Platform Id: linux

Crash Type: CHECK failure
Crash Address: 
Crash State:
  IsFirstAfterBreak(line_top_in_flow_thread) || !line.PaginationStrut() || !IsLogi
  blink::MinimumSpaceShortageFinder::ExamineLine
  blink::ColumnBalancer::TraverseLines
  
Sanitizer: address (ASAN)

Fixed: https://clusterfuzz.com/revisions?job=linux_debug_chrome&range=613032:613033

Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=4999503415083008

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 4999503415083008 is verified as fixed, so closing issue as verified.

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