[Layout] Fix clusterfuzz DCHECK
We were hitting DCHECK(HasSetAbsoluteColumnIndex()); when recalculating borders. Cells do not have AbsoluteColumnIndex set when section needs cell recalc. Fix is to make section recalc if needed before computing borders. I am not sure if this will have any performance implications... Clusterfuzz is quite elaborate. Several steps are involved in triggering the bug: - set needs_cell_recalc on section - without triggering layout on section, cause child to ask section's cell to compute its borders. Figuring a reduced test case would take about a day. Considering that I am working on eliminating this code completely, and that this bug has been around forever, I think no test case is ok. Bug: 999861 Change-Id: Ia1abcd23646fd534d740ece6d9cae52cf101829e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1796570Reviewed-by:Emil A Eklund <eae@chromium.org> Reviewed-by:
David Grogan <dgrogan@chromium.org> Commit-Queue: Aleks Totic <atotic@chromium.org> Cr-Commit-Position: refs/heads/master@{#695847}
Showing
Please register or sign in to comment