Bug 144256 - REGRESSION(r183072): dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit fails on AArch64 Linux
Summary: REGRESSION(r183072): dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cji...
Status: REOPENED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks: 108645 143734
  Show dependency treegraph
 
Reported: 2015-04-27 02:41 PDT by Csaba Osztrogonác
Modified: 2017-10-18 02:41 PDT (History)
7 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Csaba Osztrogonác 2015-04-27 02:41:28 PDT
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit
started to fail intermittently on the EFL Linux AArch64 bot. Maybe it fails on iOS AArch64 too,
but I don't know, because there isn't public iOS buildbot.

The first failure: https://build.webkit.org/builders/EFL%20Linux%20AArch64%20Release/builds/455 (r183204)
183202-183204 seems to be absolutely unrelated. There are many JSC changes near here:
https://trac.webkit.org/changeset/183201
https://trac.webkit.org/changeset/183199
https://trac.webkit.org/changeset/183162
https://trac.webkit.org/changeset/183161
...

I'm going to find which commit caused this regression.
Comment 1 Csaba Osztrogonác 2015-04-27 03:28:03 PDT
failure log:

jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: DIFF FAILURE!
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: --- ../.tests/jsc-layout-tests.yaml/js/dfg-put-by-id-prototype-check-expected.txt	2015-04-27 00:51:01.000000000 -0700
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: +++ ../jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit.out	2015-04-27 03:26:24.470000000 -0700
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: @@ -903,8 +903,8 @@
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS o.f is 897
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS o.f is 898
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS o.f is 899
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: -PASS stuff is 900
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: -PASS o.f is undefined
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: +FAIL stuff should be 900 (of type number). Was undefined (of type undefined).
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit: +FAIL o.f should be undefined (of type undefined). Was 900 (of type number).
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS stuff is 901
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS o.f is undefined
jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit:  PASS stuff is 902
FAIL: jsc-layout-tests.yaml/js/script-tests/dfg-put-by-id-prototype-check.js.layout-dfg-eager-no-cjit
Comment 2 Csaba Osztrogonác 2015-04-27 03:40:15 PDT
I can reproduce it on r183160 too, I'll check older revisions too.
Comment 3 Csaba Osztrogonác 2015-04-27 05:02:19 PDT
I found, http://trac.webkit.org/changeset/183072 is the culprit. 
The failure can be reproduced on r183072, but not before r183072.
Comment 4 Michael Saboff 2015-04-27 12:51:49 PDT
We are not seeing this failure on iOS AArch64 testbots.
Comment 5 Csaba Osztrogonác 2015-05-08 02:40:19 PDT
I painted the AArch64 Linux bot green with skipping this flakey test:
https://trac.webkit.org/changeset/183989
Comment 6 Zan Dobersek 2017-10-18 02:31:59 PDT
These failures aren't exhibited anymore.
Comment 7 Zan Dobersek 2017-10-18 02:41:28 PDT
(In reply to Zan Dobersek from comment #6)
> These failures aren't exhibited anymore.

Reopening because the test hasn't been unskipped yet. Otherwise, yes, this failure isn't exhibited anymore.