Issue metadata
Sign in to add a comment
|
Bad-cast to blink::LayoutBox from invalid vptr in blink::LayoutBlockFlow::XPositionForFloatIncludingMargin |
||||||||||||||||||||||||||||
Issue descriptionDetailed report: https://clusterfuzz.com/testcase?key=5914353736613888 Fuzzer: inferno_twister Job Type: linux_ubsan_vptr_chrome Platform Id: linux Crash Type: Bad-cast Crash Address: 0x18350041c1f0 Crash State: Bad-cast to blink::LayoutBox from invalid vptr blink::LayoutBlockFlow::XPositionForFloatIncludingMargin blink::LayoutBlockFlow::AddOverflowFromFloats Sanitizer: undefined (UBSAN) Recommended Security Severity: High Regressed: https://clusterfuzz.com/revisions?job=linux_ubsan_vptr_chrome&range=531696:531702 Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=5914353736613888 Issue filed automatically. See https://github.com/google/clusterfuzz-tools for more information.
,
Apr 20 2018
wangxianzhu@ -- can you please help triage this and find the right owner? Thanks.
,
Apr 21 2018
Let's go through the Blink>Layout triage process.
,
Apr 21 2018
,
Apr 21 2018
,
Apr 21 2018
,
Apr 22 2018
Add eae@ for triage.
,
Apr 23 2018
eae@ -- this is a high severity security issue, so please triage this on priority. thanks.
,
Apr 23 2018
There are no blink or even rendering process changes in the regression range and it does not reproduce. Without either a way to reproduce or a regression range there really isn't much we can do.
,
Apr 24 2018
,
Apr 24 2018
,
Apr 24 2018
eae@: Agree that the regression range looks wrong. When you say "it does not reproduce", are you using the ClusterFuzz reproduce tool (e.g. "clusterfuzz reproduce 5914353736613888")? Loading the POC HTML in Canary (68.0.3405.0 on Windows) results in a crash; crash/a67d4ddb142aa083 with a slightly different crash stack.
,
Apr 24 2018
You are probably looking for a change made after 513865 (known good), but no later than 513886 (first known bad). CHANGELOG URL: https://chromium.googlesource.com/chromium/src/+log/4f31578028a9aff1f6535153d0dc0271889493dc..7a22edaa3ae7ef77e5d2499062bbccf24f1f5f1c This looks suspect? https://chromium.googlesource.com/chromium/src/+/7a22edaa3ae7ef77e5d2499062bbccf24f1f5f1c%5E%21/#F2
,
Apr 24 2018
Bisecting against the continuous builds indeed shows https://chromium.googlesource.com/chromium/src/+/7a22edaa3ae7ef77e5d2499062bbccf24f1f5f1c as the culprit.
,
Apr 24 2018
eae@ is currently out of office. kojii@, is this something you might be able to help with?
,
Apr 25 2018
Assigning to culprit author. Robert, this is crashing in a bunch of places. can you please take a look or revert.
,
Apr 30 2018
,
Apr 30 2018
,
Apr 30 2018
,
May 1 2018
Can you please revert https://chromium-review.googlesource.com/c/chromium/src/+/1036489. i am getting conflicts during to i think code move to blink/.
,
May 1 2018
ClusterFuzz has detected this issue as fixed in range 555009:555011. Detailed report: https://clusterfuzz.com/testcase?key=5914353736613888 Fuzzer: inferno_twister Job Type: linux_ubsan_vptr_chrome Platform Id: linux Crash Type: Bad-cast Crash Address: 0x18350041c1f0 Crash State: Bad-cast to blink::LayoutBox from invalid vptr blink::LayoutBlockFlow::XPositionForFloatIncludingMargin blink::LayoutBlockFlow::AddOverflowFromFloats Sanitizer: undefined (UBSAN) Recommended Security Severity: High Regressed: https://clusterfuzz.com/revisions?job=linux_ubsan_vptr_chrome&range=531696:531702 Fixed: https://clusterfuzz.com/revisions?job=linux_ubsan_vptr_chrome&range=555009:555011 Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=5914353736613888 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.
,
May 1 2018
Bulls-eye!
,
May 1 2018
,
May 1 2018
,
May 1 2018
We should definitely merge the revert, this is a bad regression, see list of dupes. https://chromium-review.googlesource.com/c/chromium/src/+/1036856
,
May 1 2018
This bug requires manual review: M67 has already been promoted to the beta branch, so this requires manual review Please contact the milestone owner if you have questions. Owners: cmasso@(Android), cmasso@(iOS), kbleicher@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
May 1 2018
,
May 1 2018
,
May 1 2018
This is one reason why i hate two things in layout - re-layout with floats, clearing of anonymous blocks. this is the floats one, eight years later, these uafs are still annoying (https://trac.webkit.org/search?q=inferno+float).
,
May 1 2018
Good thing we're, finally, about to replace our float and clearance implementation :)
,
May 1 2018
govind - yep, we should revert in 67 too
,
May 1 2018
Approving merge for revert to M67 branch 3396 based on comments #25 and #31. Please do revert merge ASAP so we can pick it up for this week beta release. Thank you.
,
May 1 2018
Merged into 3396 as dc36611260ea8e98f26e9384b0471531bd4d331e.
,
May 2 2018
,
May 2 2018
,
May 29 2018
,
Aug 7
This bug has been closed for more than 14 weeks. Removing security view restrictions. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot |
|||||||||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||||||||
Comment 1 by ClusterFuzz
, Apr 20 2018Labels: Test-Predator-Auto-Components